GPU Wars: Nvidia vs. AMD

Possible AMD Radeon HD 7700 Series Specs Revealed

2012-02-13T08:01:00.000-08:00

As report first of February in a leaked roadmap, AMD is set to release its HD 7700 series graphics cards based on the Cape Verde. Based on information coming out of German-based 3dcenter.org, we may have our first glimpse at the full specifications for the new Cape Verde series. The HD 7700 series will utilize AMD's GCN stream processors seen with the HD 7900 series. With the HD 7770 series, you'll have a base core-clock speed of 1 GHz, with performance falling between the HD 6850 and HD 6790, at a suggested price around $150 dollars. The HD 7750 should equal performance of the HD 6770/5770 in performance, at a suggested price of around $125 dollars.

With AMD reportedly set to release its Radeon HD 7700 series next week, we get our first peek at the specifications of the HD 7770 and HD 7750.

As report first of February in a leaked roadmap, AMD is set to release its HD 7700 series graphics cards based on the Cape Verde. Based on information coming out of German-based 3dcenter.org, we may have our first glimpse at the full specifications for the new Cape Verde series. The HD 7700 series will utilize AMD's GCN stream processors seen with the HD 7900 series. With the HD 7770 series, you'll have a base core-clock speed of 1 GHz, with performance falling between the HD 6850 and HD 6790, at a suggested price around $150 dollars. The HD 7750 should equal performance of the HD 6770/5770 in performance, at a suggested price of around $125 dollars.

Cape Verde Physical
•Built on TSMC 28 nm process, ~1.5 billion transistors
•10 Graphics CoreNext Compute Units (CUs)
•640 stream processors
•40 TMUs, 16 ROPs
•128-bit wide GDDR5 memory interface

Radeon HD 7770
•All CUs enabled, 640 stream processors
•1 GB GDDR5 memory
•40 TMUs, 16 ROPs
•1000 MHz core clock-speed
•1125 MHz (actual), 4500 MHz (effective) memory clock-speed
•72 GB/s memory bandwidth
•1280 GFLOP/s single-precision floating-point performance
•Typical board power: 80W

Radeon HD 7750
•8 CUs enabled, 512 stream processors
•1 GB GDDR5 memory
•32 TMUs, 16 ROPs
•800 MHz core clock-speed
•1125 MHz (actual), 4500 MHz (effective) memory clock-speed
•72 GB/s memory bandwidth
•819 GFLOP/s single-precision floating-point performance
•Typical board power: 55W

Please keep in mind, of course, that these specifications are from 3dcenter's supposed reliable source. We won't know for sure until AMD shows its hand. Stay tuned!

Source: tomshardware.com

There are two GK104/Kepler variants

2012-02-13T07:59:00.000-08:00

Sources are now telling SemiAccurate that Nvidia has two variants of the GK104 in the pipe. These two variants hint at a finer grained fusing ability for the end product.

The two siblings are said to be GK104-400 and GK104-335, basically a full working and partially fused off version of the same chip. The -400 is said to be an “8 group” device, the -335 described as “7 group’. If you recall the sad tale of Fermi/GF100, the chip had large swathes of shaders turned off, the ability to do less radical surgery was not there. This is a fairly painful way to deal with defects, the more granular you can make the disabling, the better off you are.

Not much between them this time

Sources are now telling SemiAccurate that Nvidia has two variants of the GK104 in the pipe. These two variants hint at a finer grained fusing ability for the end product.

The two siblings are said to be GK104-400 and GK104-335, basically a full working and partially fused off version of the same chip. The -400 is said to be an “8 group” device, the -335 described as “7 group’. If you recall the sad tale of Fermi/GF100, the chip had large swathes of shaders turned off, the ability to do less radical surgery was not there. This is a fairly painful way to deal with defects, the more granular you can make the disabling, the better off you are.

Nothing comes for free in the silicon world, and the art of chip design is balancing granularity with cost. Nvidia botched this badly in Fermi, and paid a high price. The only good that came of it was the entertainment in seeing their spokespeople spin ever increasing leaps of logic in public. This however doesn’t placate investors much, even if they do smile.

With this new description of the -400 and -335 variant of GK104, it looks like Nvidia has implemented what AMD has been doing since at least the R700 (HD4000) chips, if not earlier. Instead of being forced to fuse off large blocks of shaders as a minimum, it looks like they can now do much smaller chunks. In Fermi terms, instead of taking a CU at a whack, they can now do portions of a CU too.

This should greatly improve yields, allow for endless SKU variations, and generally make things better. Of course, it comes at a die size penalty, but after the last learning experience, it would be foolish to do otherwise.

Source: semiaccurate.com

AMD Radeon HD 7950 Game Benchmarks

2012-02-10T08:18:00.000-08:00

A collections of real-world gaming tests using AMD Radeon HD 7950 Game Benchmarks, listed alphabetically.

"At an anticipated price tag of $450, it's a cheaper, cooler, and more power-friendly alternative to GeForce GTX 580. In games, there's really no contest in a decision between the two." -tomshardware.com

Rambus and Nvidia sign patent agreement

2012-02-09T07:42:00.000-08:00

Rambus and Nvidia have signed a patent license agreement that covers the use of Ramubs patented innovations in a brad range of integrated circuit products offered by Nvidia. This patent agreement also settles all claims that Rambus had been pushing against Nvidia for years.

The details regarding the new agreement have been safely locked away and kept confidential. The only thing that was released to public is the actual duration of the agreement that will cover a period of five years. Harold Hughes, president and CEO of Rambus didn't hesitate to point out the importance of this licence agreement that also allows Rambus to move forward with Nvidia and "focus on developing innovative solutions in concert with our licensees to help bring compelling, innovative products to market."

Burry the lawsuit hatchet

Rambus and Nvidia have signed a patent license agreement that covers the use of Ramubs patented innovations in a brad range of integrated circuit products offered by Nvidia. This patent agreement also settles all claims that Rambus had been pushing against Nvidia for years.

The details regarding the new agreement have been safely locked away and kept confidential. The only thing that was released to public is the actual duration of the agreement that will cover a period of five years. Harold Hughes, president and CEO of Rambus didn't hesitate to point out the importance of this licence agreement that also allows Rambus to move forward with Nvidia and "focus on developing innovative solutions in concert with our licensees to help bring compelling, innovative products to market."

You can check out more here.

Source: fudzilla.com

AMD Announces Turks Based FirePro V3900

2012-02-09T07:24:00.000-08:00

While AMD’s consumer GPU division is well into its deployment of their first 28nm products, the long validation and certification period for business hardware means that AMD’s business GPU division is still in the process of wrapping up the last of their 40nm product launches. In November AMD launched the Turks based FirePro V4900, and today they’re launching the final member of the current generation FirePro product stack: the FirePro V3900.

If the FirePro V4900 was a business version of the Radeon HD 6670, then the FirePro V3900 is a business version of the Radeon HD 6570 DDR3. Clocked at 650MHz and coupled with 1GB of 900MHz DDR3, the hardware specs are identical to the DDR3 version of the Radeon HD 6570.

	AMD FirePro V7900	AMD FirePro V5900	AMD FirePro V4900	AMD FirePro V3900
Stream Processors	1280	512	480	480
Texture Units	80	32	24	24
ROPs	32	32	8	8
Core Clock	725MHz	600MHz	800MHz	650MHz
Memory Clock	1.25GHz (5GHz data rate) GDDR5	500MHz (2GHz data rate) GDDR5	1GHz (4GHz data rate) GDDR5	900MHz (1.8GHz data rate) DDR3
Memory Bus Width	256-bit	256-bit	128-bit	128-bit
VRAM	2GB	2GB	1GB	1GB
FP64	N/A	N/A	N/A	N/A
Transistor Count	2.64B	2.64B	716M	716M
TDP	<150W	<75W	<75W	<50W
Manufacturing Process	TSMC 40nm	TSMC 40nm	TSMC 40nm	TSMC 40nm
Price Point	N/A	N/A	$189	$119

If the FirePro V4900 was a business version of the Radeon HD 6670, then the FirePro V3900 is a business version of the Radeon HD 6570 DDR3. Clocked at 650MHz and coupled with 1GB of 900MHz DDR3, the hardware specs are identical to the DDR3 version of the Radeon HD 6570.

For this reason the FirePro V3900 compares to the V4900 in much the same way the 6570 and 6670 do. While the V3900 has a lower core clock (650MHz vs. 800MHz), it’s otherwise a fully functional Turks GPU just like the V4900. The bigger reason for their performance difference is that while the V4900 uses GDDR5, the V3900 uses DDR3, giving it less than half the memory bandwidth and a similar overall performance drop compared to the V4900.

Of course the tradeoff for this drop in performance is size and power consumption. While the V4900 was a full profile card rated for 75W the V3900 is a low-profile card rated for 50W, with most of those power savings coming from switching out GDDR5 for DDR3. This makes the V3900 unique in that it’s the only low-profile FirePro card in AMD’s lineup – though it should be noted that for compatibility purposes it will be shipping with its full-profile bracket installed while the low-profile bracket will be in the box.

AMD will be releasing the V3900 today, with a price of $119. This positions it directly against NVIDIA’s GT216 based Quadro 400, and roughly $50 below NVIDIA’s GF108 based Quadro 600. For the V3900 AMD will be heavily leaning upon the fact that Turks can drive 5 monitors. However as with the V4900 this feature is effectively MIA until DisplayPort MST hubs ship this summer, as without the hub the card can only drive up to 2 monitors via its DP 1.2 and DL-DVI ports.

More immediately, on paper the V3900 should be far more powerful than the architecturally ancient Quadro 400. But as this is the professional market AMD’s real competition is NVIDIA’s certification and support, more so than their performance at any given price.

On that note, since AMD already launched a Turks based FirePro last year the certification process should be rather straightforward. Products (rather than GPUs) are individually certified, but as AMD already worked out any Turks driver kinks for the V4900 there shouldn’t be any surprises in store for the V3900.

Finally, it’s interesting to note that with this launch AMD has effectively committed to keeping Turks around for quite some time. AMD’s 3 year FirePro lifecycle means that the V3900 will be available until at least February of 2015, some 4 years after the first Turks products launched. Given Turks’ continual recurrence through 2012 in OEM laptops, desktops, and now professional cards, it’s clear that it’s living up to its position of being AMD’s low cost, high volume anchor GPU for the 40nm generation.

Source: anandtech.com

HD 7950 in healthy quantity in EU, US

2012-02-04T07:12:00.000-08:00

AMD’s new HD 7950 is the talk of town these days and for good reason, too. The second iteration of AMD’s Tahiti GPU costs quite a bit less than the HD 7970, yet it’s fast enough to wipe the floor with Nvidia’s GTX 580.

From what we’ve been hearing, demand is good, but luckily there is no shortage of HD 7950’s in retail. A few dozen shops in mainland Europe have plenty of cards in stock and the prices are pretty good for a brand new product. In Germany, you can pick up a PowerColor reference card for as little as €386, or a Sapphire card for €397. You can even get a slightly overclocked 830MHz card from MSI for €393. In case you are looking for something in the 880MHz – 900MHz range, you have a choice of overclocked cards from Gigabyte, MSI, Sappire, PowerColor and XFX, with prices starting at €441. The HD 7970 sells for €519+, while the cheapest GTX 580 3GB cards go for €460+.

Prices starting at $449, sub-€400

AMD’s new HD 7950 is the talk of town these days and for good reason, too. The second iteration of AMD’s Tahiti GPU costs quite a bit less than the HD 7970, yet it’s fast enough to wipe the floor with Nvidia’s GTX 580.

From what we’ve been hearing, demand is good, but luckily there is no shortage of HD 7950’s in retail. A few dozen shops in mainland Europe have plenty of cards in stock and the prices are pretty good for a brand new product. In Germany, you can pick up a PowerColor reference card for as little as €386, or a Sapphire card for €397. You can even get a slightly overclocked 830MHz card from MSI for €393. In case you are looking for something in the 880MHz – 900MHz range, you have a choice of overclocked cards from Gigabyte, MSI, Sappire, PowerColor and XFX, with prices starting at €441. The HD 7970 sells for €519+, while the cheapest GTX 580 3GB cards go for €460+.

In the US, prices start at $449 for reference cards, while overclocked models retail for €489 or €499, depending on the clock and vendor. This is a full $100 less than the HD 7970 or GTX 580 3GB.

All in all AMD has done a pretty good job with the HD 7950. Quite a few consumers find the HD 7970 a tad too expensive, so the HD 7950 looks like a great deal, particularly in Europe, as it is up to 30 percent cheaper than the flagship HD 7970.

Source: fudzilla.com

Physics hardware makes Kepler/GK104 like crazy

2012-02-04T06:33:00.000-08:00

Nvidia’s Kepler/GK104 chip has an interesting secret, a claimed Ageia PhysX hardware block that really isn’t. If you were wondering why Nvidia has been beating the dead horse called PhysX for so long, now you know, but it only gets more interesting from there.

Sources tell SemiAccurate that the ‘big secret’ lurking in the Kepler chips are optimisations for physics calculations. Some are calling this PhysX block a dedicated chunk of hardware, but more sources have been saying that it is simply shaders, optimisations, and likely a dedicated few new ops. In short, marketing may say it is, but under the heat spreader, it is simply shaders and optimisations.

That is the marketing claim anyway

Nvidia’s Kepler/GK104 chip has an interesting secret, a claimed Ageia PhysX hardware block that really isn’t. If you were wondering why Nvidia has been beating the dead horse called PhysX for so long, now you know, but it only gets more interesting from there.

Sources tell SemiAccurate that the ‘big secret’ lurking in the Kepler chips are optimisations for physics calculations. Some are calling this PhysX block a dedicated chunk of hardware, but more sources have been saying that it is simply shaders, optimisations, and likely a dedicated few new ops. In short, marketing may say it is, but under the heat spreader, it is simply shaders and optimisations.

The market has treated hardware PhysX like an unexplained sore that shows up a week after a night you can’t remember through a tequila induced haze. Numbers vary about the absolute magnitude of PhysX’s overwhelming success, but counts of 2011 game releases supporting hardware acceleration range from a low of two to a high of six. The snowball has pretty much stopped rolling, or to be more accurate, it never started, all the developers SemiAccurate spoke with indicate that their use of PhysX hardware acceleration was a cash flow positive experience, but we didn’t talk to all six listed.

With this new bit of information, one big question is answered, but specific hardware implementations details are a bit murky. Is the ‘hardware block’ dedicated to physics calculations when there are some being issued, or is it a AMD/GCN like multiple instruction issue? Is it just shaders with an added op or two that speed up math routines heavily used by physics simulations? How much die area is spent on this functionality? This isn’t very clear, and given the marketing materials Semiaccurate has seen, explanations will only serve to impede the impending hype.

That said, SemiAccurate is told Kepler/GK104 will be marketed as having a dedicated block, and this will undoubtedly be repeated everywhere, truth not withstanding. Luckily, since most of the target audience isn’t technically literate, it may “become fact” through the VIECOOCDF (Vast Internet Echo Chamber Of Often Repeated Dubious Facts). Lowering the collective intelligence can be profitable if not ethical. Luckily, the story doesn’t end here, it gets much worse.

This part ties in to the story SemiAccurate published a few weeks ago saying that Nvidia would win this generation. A lot of people have been asking about Kepler/GK104 performance and if it is really that good. The short story is yes and no, depending on your views on some very creative ‘optimisations’ around physics.

We stated earlier, Kepler wins in most ways vs the current AMD video cards. How does Nvidia do it with a $299 card? Is it raw performance? Massive die size? Performance per metric? The PhysX ‘hardware block’? Cheating? The easy answer is yes, but lets go in to a lot more detail.

GK104 is the mid-range GPU in Nvidia’s Kepler family, has a very small die, and the power consumption is far lower than the reported 225W. How low depends on what is released and what clock bins are supported by the final silicon. A1 stepping cards seen by SemiAccurate had much larger heatsinks than the A2 versions, and recent rumours suggest there may be an A3 to fix persistent PCIe3 headaches.

To date, an A3 spin has not been confirmed, but if it is necessary, it will likely push out the late March/early April release date by at least a month. One other possibility is for Nvidia to pull an Intel and release cards without the official PCI SIG stamp, adding it when A3 silicon is available. In any case, the number of PCIe3 supporting computers on the market is minimal, so functionally speaking, it doesn’t matter. You may loose a small bit of theoretical performance, but for a mid-range part, it is unlikely to be noticeable. Marketing is a completely different story though, one not closely tied to the reality most of us live in.

The architecture itself is very different from Fermi, SemiAccurate’s sources point to a near 3TF card with a 256-bit memory bus. Kepler is said to have a very different shader architecture from Fermi, going to much more AMD-like units, caches optimised for physics/computation, and clocks said to be close to the Cayman/Tahiti chips. The initial target floating among the informed is in the 900-1000MHz range. Rumours have it running anywhere from about 800MHz in early silicon to 1.1+GHz later on, with early stepping being not far off later ones. Contrary to some floating rumours, yields are not a problem for either GK104 or TSMC’s 28nm process in general.

Performance is likewise said to be a tiny bit under 3TF from a much larger shader count than previous architectures. This is comparable to the 3.79TF and 2048 shaders on AMD’s Tahiti, GK104 isn’t far off either number. With the loss of the so called “Hot Clocked” shaders, this leaves two main paths to go down, two CUs plus hardware PhysX unit or three. Since there is no dedicated hardware physics block, the math says each shader unit will probably do two SP FLOPs per clock or one DP FLOP.

This would be in line with the company’s earlier claims of a large jump in compute capabilities, but also leads to questions of how those shaders will be fed with only a 256-bit memory path. Given the small die sizes floating around, it is unlikely to be Itanium-esque brute forcing through large caches. The net result is that shader utilisation is likely to fall dramatically, with a commensurate loss of real world performance compared to theoretical peak.

In the same way that AMD’s Fusion chips count GPU FLOPS the same way they do CPU FLOPS in some marketing materials, Kepler’s 3TF won’t measure up close to AMD’s 3TF parts. Benchmarks for GK104 shown to SemiAccurate have the card running about 10-20% slower than Tahiti. On games that both heavily use physics related number crunching and have the code paths to do so on Kepler hardware, performance should seem to be well above what is expected from a generic 3TF card. That brings up the fundamental question of whether the card is really performing to that level?

This is where the plot gets interesting. How applicable is the “PhysX block”/shader optimisations to the general case? If physics code is the bottleneck in your app, A goal Nvidia appears to actively code for, then uncorking that artificial impediment should make an app positively fly. On applications that are written correctly without artificial performance limits, Kepler’s performance should be much more marginal. Since Nvidia is pricing GK104 against AMD’s mid-range Pitcairn ASIC, you can reasonably conclude that the performance will line up against that card, possibly a bit higher. If it could reasonably defeat everything on the market in a non-stacked deck comparison, it would be priced accordingly, at least until the high end part is released.

All of the benchmark numbers shown by Nvidia, and later to SemiAccurate, were overwhelmingly positive. How overwhelmingly positive? Far faster than an AMD HD7970/Tahiti for a chip with far less die area and power use, and it blew an overclocked 580GTX out of the water by unbelievable margins. That is why we wrote this article. Before you take that as a backpedal, we still think those numbers are real, the card will achieve that level of performance in the real world on some programs.

The problem for Nvidia is that once you venture outside of that narrow list of tailored programs, performance is likely to fall off a cliff, with peaky performance the likes of which haven’t been seen in a long time. On some games, GK104 will handily trounce a 7970, on others, it will probably lose to a Pitcairn. Does this mean it won’t actually do what is promised? No, it will. Is this a problem? Depends on how far review sites dare to step outside of the ‘recommenced’ list of games to benchmark in the reviewers guide.

Ethically, this could go either way, and in a vacuum, we would be more than willing to say that the cards are capable of very high performance. The problem is that the numbers that Nvidia will likely show off at the launch are not in a vacuum, nor are they very real, even considering the above caveats. Nvidia is going out of their way to have patches coded for games that tend to be used as benchmarks by popular sites.

Once again, this is nothing new, and has been done many times before. One example that is often mentioned is Starcraft II’s use of stencil buffers. People with inside knowledge of that game’s development have said that Nvidia gave Blizzard help in coding some parts of the game during the final ‘crunch’ period. The code is said to heavily use stencil buffers to fix some issues and patch over minor glitches. Again nothing unusual, AMD, Intel, and almost everyone else does this on a case by case basis, especially for AAA titles released in conjunction with new hardware.

Since Nvidia’s Fermi generation GPUs are very good at handling stencil buffers, they perform very well on this code. Again, this is normal practice, Nvidia put in the effort and now reaps the benefits, good for them. What is odd about this case, is that several knowledgeable sources have said that the code actually net decreases performance on both cards. The above tale may be anecdotal, but Starcraft 2′s release code sure seemed to use stencil buffers a lot more than you would expect, unreasonably so according to many coders. This however doesn’t constitute proof in any way, but it fits what SemiAccurate has seen Nvidia do in prior cases.

More to the point is antialiasing (AA) in Batman: Arkham Asylum. If you recall, AMD stated complaining about that game’s AA routines upon release. They directly stated that if AMD cards were detected, the game would disable AA for non-technical reasons. (Note: The original post that TechPowerUp refers to has the pertinent sections in the comments, not on the front page. It takes a little searching to find the post that also talks about several other games having similar ‘bugs’.) It goes on to state that if the card IDs were changed, the AA in the game functioned correctly on ATI hardware.

Short story, this turned in to the proverbial “epic pissing match”, with Nvidia claiming that it was Eidos that owned the code, and they were free to do with it as they feel fit. This is technically true. Unfortunately, emails seen by SemiAccurate directly contradict this. Those emails state unequivocally that Eidos should not change code written by Nvidia and provided to Eidos as a part of Batman: Arkham Asylum. At the point they were questioned on why, Eidos says they could not do anything due to advice of their attorneys.

Since it was the attorneys objecting, not the coders, we can only speculate that this was due to Nvidia’s financial sponsorship of the game, not any technical reason. Since sources tell SemiAccurate that Batman: Arkham Asylum only uses standard DirectX calls to implement AA, and it appears to function if the graphics card IDs are changed, this seems to be nothing other than Nvidia directly sabotaging their competition and not allowing AMD remove the lockout. Go and re-read the statements from AMD/ATI, Nvidia, and Eidos, then draw your own conclusions.

Why do we bring these two cases up in a Kepler article? Well, we hear that it is happening again. Both AMD and Nvidia have developers that they can and do ‘embed’ at game companies. This is an old and quite legitimate practice for GPU and non-GPU hardware companies. Everyone does it. It can be done ethically or not, with net performance gains for the end user or not, and with the intent to hurt or harm. In general, the more marketing money involved, the more most developers are willing to go out on a shaky ethical limbs.

One last really good example, tesselation. High end Fermi cards, GF100/110/GTX480/GTX580 are heavily biased toward geometry performance. Since most modern GPUs can compute multiple triangles per displayable pixel on any currently available monitor, usually multiple monitors, doubling that performance is a rather dubious win. Doubling it again makes you wonder why so die area was wasted.

Since Nvidia did waste that die area, helping games show that prowess off is a good thing for users, right? Look at Crysis 2, a AAA title that is heavily promoted by Nvidia, it positively flies on Fermi based cards, but performance on AMD GPUs is far less impressive. Why? The amazing detail in things like the concrete blocks, brick walls, and vast expanses of realistically modelled water. Breathtaking isn’t it? All thanks to Nvidia’s efforts to make the game experience better on their hardware. How could this be interpreted as anything but a win for users by a reasonable observer?

Nvidia is said to have around 15 developers they can embed at companies to help ‘optimise’ their code, ‘fix bugs’, and work out ‘performance problems’, even if those problems are not on Nvidia hardware. The count for other companies is less clear, but unlikely to be much different. Sources tell SemiAccurate that about half of them are currently working at Eidos on, wait for it, a patch for the recently released Batman: Arkham City game. Since Both the original and and the new Batman games are flag bearers for Nvidia’s hardware/GPU PhysX acceleration, it doesn’t take a genius to connect the dots. Since neither the patch or Kepler based video cards are out yet, we can only wait to see what the end result is.

If the purported patch does change performance radically on specific cards, is this legitimate GPU performance? Yes. How about if it raises performance on Kepler cards while decreasing performance on non-Kepler cards to a point lower than pre-patch levels? How about if it raises performance on Kepler cards while decreasing performance only on non-Nvidia cards? Which scenario will it be? Time will tell.

How many other games have had this level of attention and optimisation gifted upon them is another open question. One thing we can say is that the list of benchmarks shown off by Nvidia where Kepler has an overwhelming advantage all support PhysX. This is not to say that they are all hardware/GPU PhysX accelerated, they are not, most use the software API.

This is important because it strongly suggests that Nvidia is accelerating their own software APIs on Kepler without pointing it out explicitly. Since Kepler is a new card with new drivers, there is no foul play here, and it is a quite legitimate use of the available hardware. Then again, they have been proven to degrade the performance of the competition through either passive or active methods. Since Nvidia controls the APIs and middleware used, the competition can not ‘fix’ these ‘problems with the performance of their hardware’.

Going back to Kepler, we see that this happy and completely ethical game is going to be starting round 3, or round 17, depending on how you count. Nvidia appears to be stacking the playing field to both cripple the competition and raise their own performance. Is the performance of Kepler cards legitimate? Yes. Is it the general case? No. If you look at the most comprehensive list of supported titles we can find, it is long, but the number of titles released per year isn’t all that impressive, and anecdotally speaking, appears to be slowing.

When Kepler is released, you can reasonably expect extremely peaky performance. For some games, specifically those running Nvidia middleware, it should fly. For the rest, performance is likely to fall off the proverbial cliff. Hard. So hard that it will likely be hard pressed to beat AMD’s mid-range card.

What does this mean in the end? Is it cheating? Is it ethical? Is Kepler/GK104 going to be worth the money? Will it beat AMD’s 7970? These are all subjective decisions for you to make. What software will Nvidia show off as benchmarks to promote Kepler’s performance? That list is a little narrower. What will happen to sites that dare to test software that is not ‘legitimately accelerated’? No idea, but history offers some clues. One thing you can say for sure is that the information released prior to and with the card is unlikely to be the whole story. Legitimacy, performance, honesty, and ethics are unlikely to resemble the official talking points, and the whole truth is likely to be hidden from prying eyes for very partisan reasons. Big grains of salt around this one people, be very skeptical of everything you hear, and take nothing at face value.

Source: semiaccurate.com

Graphics Cards Maker Leaks AMD Roadmap: New Radeon HD 7000 Incoming in February, March.

2012-02-04T06:29:00.000-08:00

A well-known maker of graphics cards has revealed its roadmap for the following months. The roadmap of AMD’s incoming new Radeon HD 7000-series graphics solutions brings no surprises: as expected, the new graphics cards with new architecture will become available in February - March timeframe.

Based on a slide from an alleged presentation by MicroStar International (MSI) (first published by CPU-World web-site), in mid-February AMD and its partners will release its Radeon HD 7700-series “Cape Verde” graphics processing units and in March they will unveil Radeon HD 7800-series “Pitcairn” GPUs. The dual-chip flagship Radeon HD 7990 “New Zealand” will be released a bit later, possibly when Nvidia Corp. launches its next-gen offering.

AMD’s Cape Verde, Pitcairn Are Approaching

A well-known maker of graphics cards has revealed its roadmap for the following months. The roadmap of AMD’s incoming new Radeon HD 7000-series graphics solutions brings no surprises: as expected, the new graphics cards with new architecture will become available in February - March timeframe.

Based on a slide from an alleged presentation by MicroStar International (MSI) (first published by CPU-World web-site), in mid-February AMD and its partners will release its Radeon HD 7700-series “Cape Verde” graphics processing units and in March they will unveil Radeon HD 7800-series “Pitcairn” GPUs. The dual-chip flagship Radeon HD 7990 “New Zealand” will be released a bit later, possibly when Nvidia Corp. launches its next-gen offering.

AMD’s code-named Cape Verde and Pitcairn chips will increase performance of mainstream and performance-mainstream products, respectively. Neither pricing, nor specifications, of AMD Radeon HD 7700- and 7800-series graphics cards are currently known.

While the slide was not made public officially, it is highly likely that it is generally correct. Unfortunately, based on some indicators, it looks like the roadmap is somewhat outdated. AMD originally wanted to launch its Radeon HD 7970 “Tahiti XT” graphics card on the 9th of January, 2012, at the Consumer Electronics Show and immediately start to sell the novelty. Unfortunately, the company decided to paper-launch the board on the 22nd of December, 2011. Therefore, other launch timeframes noted in the document may be incorrect.

AMD did not comment on the news-story.

Source: xbitlabs.com

AMD Starts Selling Second Radeon HD 7900-Series Graphics Card

2012-02-04T06:27:00.000-08:00

In a bid to further improve its competitive positions on the market of high-end graphics cards, Advanced Micro Devices on Wednesday began selling a cut-down version of its high-end code-named Tahiti graphics chip. The new Radeon HD 7950 promises slightly better performance compared to Nvidia GeForce GTX 580 at lower - $449 - price-point.

AMD Unleashes Radeon HD 7950 Graphics Board for $449

In a bid to further improve its competitive positions on the market of high-end graphics cards, Advanced Micro Devices on Wednesday began selling a cut-down version of its high-end code-named Tahiti graphics chip. The new Radeon HD 7950 promises slightly better performance compared to Nvidia GeForce GTX 580 at lower - $449 - price-point.

The new Radeon HD 7950 "Tahiti Pro" belongs to Southern Islands family of graphics processing units (GPUs) and is based on the so-called graphics core next (GCN) architecture. The model 7950 supports all the advantages that the GCN has to offer, including support for DirectX 11.1, OpenCL 1.2, ultra high-definition output support, optimized power consumption, and other new technologies.

Compute performance of the Radeon HD 7950 is substantially lower - by around 25% - compared to its older brother 7970. Still, thanks to new architecture and support for new features it is likely to become popular when at $449 price-point.

Numerous add-in-board partners of AMD have introduced their versions of the Radeon HD 7950 graphics cards with advanced cooling systems, printed-circuit boards design and so on.

Source: xbitlabs.com

Nvidia: We Expected More from AMD Radeon HD 7970

2012-01-26T19:11:00.000-08:00

Perhaps, the launch of the Radeon HD 7970 graphics card powered by Advanced Micro Devices’ latest Southern Islands/GCN (graphics core next) was not the most impressive one as it was a rush announcement two days ahead of Christmas. But AMD does have a next-gen GPU on the market, unlike its major rival Nvidia Corp., a clear advantage. But the latter claims that the model 7970 fails to impress.

“Honestly, we expected more from our competitor’s new architecture,” an unnamed official Nvidia initiative told NordicHardware web-site.

Nvidia itself has been astonishingly quiet about its next-generation The company has learnt its lesson when it unveiled code-named Fermi architecture about half a year ahead of actual release and would like to ensure instant availability of next-generation graphics cards after the formal launch, according to unofficial information.

Nvidia Bashes AMD’s Latest Graphics Architecture

Perhaps, the launch of the Radeon HD 7970 graphics card powered by Advanced Micro Devices’ latest Southern Islands/GCN (graphics core next) was not the most impressive one as it was a rush announcement two days ahead of Christmas. But AMD does have a next-gen GPU on the market, unlike its major rival Nvidia Corp., a clear advantage. But the latter claims that the model 7970 fails to impress.

“Honestly, we expected more from our competitor’s new architecture,” an unnamed official Nvidia initiative told NordicHardware web-site.

Nvidia itself has been astonishingly quiet about its next-generation The company has learnt its lesson when it unveiled code-named Fermi architecture about half a year ahead of actual release and would like to ensure instant availability of next-generation graphics cards after the formal launch, according to unofficial information. At present Nvidia considers scheduling the launch of code-named Kepler products on March or April. Officially, the company claims that its Kepler will be released in the first half of 2012.

While the launch of the AMD Radeon HD 7970 was rather weak in terms of publicity and in many other terms, it is clear that today, around a month after the formal product announcement, AMD has actual new-generation graphics cards on the market. Meanwhile, it will take another two or more months for Nvidia to start selling its next-gen GeForce GTX 780-series. Naturally, Nvidia is interested in slowing down sales of AMD’s code-named Tahiti graphics solution so that to satisfy delayed demand for high-end graphics boards later this year.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on. Entry-level chips may not get all the features that Kepler architecture will have to often. Production of Kepler chips was supposed to start in Q4 2011.

Source: xbitlabs.com

Overclockers reveals Radeon HD 7950 benchmark

2012-01-26T19:08:00.000-08:00

Online retailer Overclockers.co.uk have got their paws on a reference Radeon HD7950 Graphic Card and have been showing off the the price, launch date and benchmark results of the beast.

The site said that the Radeon HD7950 ran at at 810MHz Core / 5000MHz Memory, Packs 3GB GDDR5 Memory Buffer, 1792 Stream Processors, 32ROP’s, 112TMU’s and memory runs along a 384-bit wide interface. The first set of benchmark results were obtained at a resolution of 1680×1050, 4xAA and Normal Tessellation, and showed AMD 7950 Stock (810MHz Core / 5000MHz Memory) – 1379 Score, AMD 7950 Maxed on Overdrive (1025MHz Core / 6300MHz Memory) – 1689 Score, and AMD 7950 Maxed OC on Asus OC BIOS (1180MHz Core / 7200MHz Memory) – 1936 Score.

The clocks managed to reach a core clock of 1180Mhz and 7200Mhz on Memory which means that the card has huge overclocking potential. Overclockers.co.uk then pushed the card to run at 2560×1600 Resolution and it did pretty well and did not catch fire. The site said that it will sell for $449 and will not only outperform a GeForce GTX580 but also overclocks better. It is cooler too.

Apparently a 1.5GB Variant is also expected in March for $399. Overclockers.co.uk’s pricing goes, they’ll price the reference model at £350.

Source: fudzilla.com

Xbox 720 may have AMD HD6670 graphics

2012-01-26T19:04:00.000-08:00

Microsoft may still be making good money with Xbox 360 sales, but the next generation of console is already being developed with a launch said to be at the end of the year, probably around October or November. It should be around six times as powerful as the current model and is said to feature AMD HD6000 graphics inside.

IGN have had a source ‘close to the project’ and they say that it will be using the HD6670 with full support for Direct X 11, 3D and 1080p HD output, and even multidisplay capabilities.

The Radeon HD6670 costs around £60 in the United Kingdom and while it is considered a very mainstream solution in the PC sector, it is six times as powerful as the graphics in the Xbox 360, and is 20% faster than the upcoming Wii U console.

More Radeon HD 7950 benchmarks surface

2012-01-26T18:51:00.000-08:00

Following yesterday's GPU-Z screenshot and 3DMark 11 performance leak from Donanimhaber.com, Mydrivers has come up with a few additional HD 7950 benchmarks, unfortunately still only focused on 3DMark 11.

Unlike the one seen yesterday, the Radeon HD 7950 card at Mydrivers site is set at reference 800MHz for the GPU and 1250MHz for 3GB of GDDR5 memory as well as overclocked to 1030MHz for the GPU. The results of the card were compared to GTX 570, Radeon HD 6970 as well as Radeon HD 7970 cards.

Still only 3DMark 11

Following yesterday's GPU-Z screenshot and 3DMark 11 performance leak from Donanimhaber.com, Mydrivers has come up with a few additional HD 7950 benchmarks, unfortunately still only focused on 3DMark 11.

Unlike the one seen yesterday, the Radeon HD 7950 card at Mydrivers site is set at reference 800MHz for the GPU and 1250MHz for 3GB of GDDR5 memory as well as overclocked to 1030MHz for the GPU. The results of the card were compared to GTX 570, Radeon HD 6970 as well as Radeon HD 7970 cards.

As we always note, 3DMark 11 is a decent benchmark but we always like to see a card running some game titles before we draw any conclusion. The preview results in Performance 3DMark 11 show that the HD 7950 will end up competitive to both GTX 570 and probably even GTX 580 and have a decent overclocking potential to be as fast as the reference HD 7970.

Of course, we are about a week away from the official launch date set at 31st of January when we will probably see some of the first benchmarks and know for sure.

You can check out the results here.

Source: fudzilla.com

Nvidia Lowers Revenue Expectations Due to Hard Drives Shortages

2012-01-26T18:45:00.000-08:00

Nvidia Corp. on Wednesday announced that it expects revenue for the fourth quarter ending January 29, 2012, to be lower than the company's previous outlook provided back in October. The developer of multimedia chips blames shortages of hard disk drives as well as lowering demand towards Tegra 2 system-on-chips for the lowering revenue.

The global disk-drive shortage caused by the flooding in Thailand had more impact on the mainstream GPU segment than anticipated. Shipments by some PC OEMs were reduced. And the higher prices of disk-drives constrained some PC OEMs' ability to include a discrete GPU in their systems.

Nvidia Suffers from Tight Supply of Hard Drives, Lowering Demand for Tegra 2

Nvidia Corp. on Wednesday announced that it expects revenue for the fourth quarter ending January 29, 2012, to be lower than the company's previous outlook provided back in October. The developer of multimedia chips blames shortages of hard disk drives as well as lowering demand towards Tegra 2 system-on-chips for the lowering revenue.

The global disk-drive shortage caused by the flooding in Thailand had more impact on the mainstream GPU segment than anticipated. Shipments by some PC OEMs were reduced. And the higher prices of disk-drives constrained some PC OEMs' ability to include a discrete GPU in their systems.

Additionally, the Tegra 2 mobile business declined more rapidly than expected, ahead of devices based on the Tegra 3 processor ramping into production in the first quarter of calendar-year 2012.

Nvidia is not the first company to blame shortages of hard drives and their increased prices for lower-than-expected revenue. Intel Corp. also suffered as manufacturers of computers had to buy more affordable chips to compensate boosted HDD pricing. Dropping demand for Tegra 2 is rather surprising, but indicates that tablet makers are now gearing for a new breed of slates with four processing engines as well as advanced graphics adapter.

Nvidia's earnings are now expected to be $950 million, plus or minus 1% percent, compared with original expectations of $1066 million, plus or minus 2%, provided on November 10, 2011.

Source: xbitlabs.com

AMD Radeon HD 7950 listed in Europe

2012-01-26T18:29:00.001-08:00

We managed to score an early listing of the upcoming Powercolor Radeon HD 7950 PCS+ card that was listed by one retailer in Germany with a price tag of €459.

The Powercolor HD 7950 PCS+ 3072MB is factory overclocked to 880MHz while the 3072MB of GDDR5 memory ticks at 1250MHz (5GHz). The card should pack 1792 stream processors and of course feature the same 384-bit memory interface. Although the listing at Atelco.de didn't feature a picture of the card, we are pretty sure that it should end up with the same cooler that we had a chance to see recently.

Factory overclocked card priced at €459

We managed to score an early listing of the upcoming Powercolor Radeon HD 7950 PCS+ card that was listed by one retailer in Germany with a price tag of €459.

The Powercolor HD 7950 PCS+ 3072MB is factory overclocked to 880MHz while the 3072MB of GDDR5 memory ticks at 1250MHz (5GHz). The card should pack 1792 stream processors and of course feature the same 384-bit memory interface. Although the listing at Atelco.de didn't feature a picture of the card, we are pretty sure that it should end up with the same cooler that we had a chance to see recently.

Unfortunately for us, the link was taken off while we were writing up this news post, but as usual we already took a decent screenshot of the listing. Of course, Google tends to keep records, so you can see that the link was there.

Source: fudzilla.com

AMD Radeon HD 7950 detailed

2012-01-26T18:14:00.000-08:00

Although we are still over a week away from the official launch date, the guys from Donanimhaber.com scored some details regarding the upcoming Tahiti based Radeon HD 7950 card.

According to their info, the card should be running at 800MHz for the GPU and 1,250MHz (5GHz) for 3GB of GDDR5 memory paired up with the same 384-bit memory interface. In case you missed it, the Radeon HD 7950 packs 1792 stream processors, or 28 GCN compute units as AMD calls them these days.

In addition to the reference specs, Donanimhaber.com also scored 3DMark 11 Extreme and CPU-Z screenshot of the Radeon HD 7950 card running on 880MHz for the core. According to results, the card scores 2385 marks and still ends up faster than the GTX 580.

Core clock set at 800MHz

Although we are still over a week away from the official launch date, the guys from Donanimhaber.com scored some details regarding the upcoming Tahiti based Radeon HD 7950 card.

According to their info, the card should be running at 800MHz for the GPU and 1,250MHz (5GHz) for 3GB of GDDR5 memory paired up with the same 384-bit memory interface. In case you missed it, the Radeon HD 7950 packs 1792 stream processors, or 28 GCN compute units as AMD calls them these days.

In addition to the reference specs, Donanimhaber.com also scored 3DMark 11 Extreme and CPU-Z screenshot of the Radeon HD 7950 card running on 880MHz for the core. According to results, the card scores 2385 marks and still ends up faster than the GTX 580. As always, 3DMark 11 doesn't always reflect the actual real-world performance so we'll have to wait a bit longer to see how well it performs in various game titles.

As you already know, it looks like AMD gave a green light to its partners for non-reference cards since day one and there will be a lot of custom cooled and factory overclocked cards once it launched on January 31st. According to what we know, the price is still set at US $449.99.

You can check out more here.

Source: fudzilla.com

Exclusive: And the Nvidia Kepler/GK104 price is……..

2012-01-26T18:11:00.000-08:00

We hear that Nvidia (NASDAQ:NVDA) has sent out Kepler pricing to AIBs in the far east, or will once the New Year party dies down. A few green-tinged moles, we think it’s the New Year’s celebratory hair dye, tell SemiAccurate that the initial Kepler/GK104 cards will be priced around the $299 mark. This should tell you quite a bit about how large the silicon is, but not necessarily what it will be marketed as.

There are two caveats to this number that you should be aware of. First, this is the initial round of pricing, and as we know, they do change, volume, yields etc.

Initial number head out to AIBs as a New Year's gift.

We hear that Nvidia (NASDAQ:NVDA) has sent out Kepler pricing to AIBs in the far east, or will once the New Year party dies down. A few green-tinged moles, we think it’s the New Year’s celebratory hair dye, tell SemiAccurate that the initial Kepler/GK104 cards will be priced around the $299 mark. This should tell you quite a bit about how large the silicon is, but not necessarily what it will be marketed as.

There are two caveats to this number that you should be aware of. First, this is the initial round of pricing, and as we know, they do change, volume, yields etc. Second, it could well be a red herring to try and crater AMD 79xx sales, even given the number of sources that gave us the same number. Lets see how long it takes for the echo chamber to ‘independently confirm’ this one.

Source: semiaccurate.com

AMD Radeon HD 7970 now shipping from Newegg

2012-01-09T08:07:00.000-08:00

AMD's highly anticipated flagship 28nm single-GPU card, the Radeon HD 7970, has now officially become available from Newegg nearly twenty-four hours before its official launch.

The 28nm Tahiti-based GPU is available from six different add-in board partners - Sapphire, PowerColor, HIS, ASUS, XFX and Gigabyte - with a price range between $549 and $599 and approximately $8 dollar shipping within the continental United States.

We expect more AMD retail and online e-tail partners should have already received inventory and will begin listing the card for sale within the next twenty-four hour time period.

$549 - $599, high consumer demand

AMD's highly anticipated flagship 28nm single-GPU card, the Radeon HD 7970, has now officially become available from Newegg nearly twenty-four hours before its official launch.

The 28nm Tahiti-based GPU is available from six different add-in board partners - Sapphire, PowerColor, HIS, ASUS, XFX and Gigabyte - with a price range between $549 and $599 and approximately $8 dollar shipping within the continental United States.

We expect more AMD retail and online e-tail partners should have already received inventory and will begin listing the card for sale within the next twenty-four hour time period. Meanwhile, we expect Newegg's preliminary inventory to be purchased very quickly, as these cards are drawing large amounts of anticipation from PC hardware enthusiasts who value their overclocking potential and overall performance benefits.

Newegg's product listing for AMD Radeon HD 7970 models can be found here.

Source: fudzilla.com

Rest of AMD's OEM-only HD 7000 rebrands detailed

2012-01-09T08:01:00.000-08:00

After detailing the AMD Radeon HD 7670 OEM-only rebrand based on the same Turks GPU found in the HD 6670 card, the guys from Tomshardware.com managed to score details for the rest of the rebrand lineup. In addition to the HD 7670, the lineup includes the Radeon HD 7570, Radeon HD 7470/7450 as well as the entry-level Radeon HD 7350 card.

The Radeon HD 7570 is based on the same Turks GPU and with the same specs as the HD 6570, while the HD 7470 and the HD 7450 will be based on quite well known Caicos GPU with 160 stream processors, previously seen on the HD 6450 card. The last card in the lineup, the HD 7350, comes as a surprise as it is based on now quite old Evergreen family Cedar GPU, the same one that was happily ticking in the HD 5450 card.

Everything below HD 77xx

After detailing the AMD Radeon HD 7670 OEM-only rebrand based on the same Turks GPU found in the HD 6670 card, the guys from Tomshardware.com managed to score details for the rest of the rebrand lineup. In addition to the HD 7670, the lineup includes the Radeon HD 7570, Radeon HD 7470/7450 as well as the entry-level Radeon HD 7350 card.

The Radeon HD 7570 is based on the same Turks GPU and with the same specs as the HD 6570, while the HD 7470 and the HD 7450 will be based on quite well known Caicos GPU with 160 stream processors, previously seen on the HD 6450 card. The last card in the lineup, the HD 7350, comes as a surprise as it is based on now quite old Evergreen family Cedar GPU, the same one that was happily ticking in the HD 5450 card.

The good part of the story is that these are only meant for OEMs but then again rebrands are never popular. You can check out more details here.

Source: fudzilla.com

Radeon HD 7970 shows up at Amazon.de

2012-01-09T07:57:00.001-08:00

Just a day before it should officially show up in retail/e-tail, German/Austrian Amazon has jumped the gun and listed AMD's HD 7970 with a price set at €529.

According to our sources, the final price of the card for Europe is set €499. The FOB price is set at around US $480 but when you add tax and some retail margins you end up at around €499/US $549. Of course, Amazon was never one of the cheapest and you can expect at least some of the retailers/e-tailers to sell the card for €499.

The rest of the e-tail/retail should start to list the cards tomorrow and you can check out the Amazon listing here.

Listed at €529

Just a day before it should officially show up in retail/e-tail, German/Austrian Amazon has jumped the gun and listed AMD's HD 7970 with a price set at €529.

According to our sources, the final price of the card for Europe is set €499. The FOB price is set at around US $480 but when you add tax and some retail margins you end up at around €499/US $549. Of course, Amazon was never one of the cheapest and you can expect at least some of the retailers/e-tailers to sell the card for €499.

The rest of the e-tail/retail should start to list the cards tomorrow and you can check out the Amazon listing here.

Source: fudzilla.com

Close up on the Radeon 7000 series

2012-01-09T07:44:00.000-08:00

GPU manufacturers must be desperate for the present generation of consoles to fade away and die. With so much money available to publishers in the console space, there’s a reduced drive to create games that need much more than the weak-arsed specs found in a PlayStation 3 or XBox 360.

Still, we know that a quantum leap will happen at some stage – at which point we’ll all be reaching for ‘more’.

If you’d like to digest the 7000 series launch in no time at all, we have the video for you:-

AMD Readies “Sea Islands” Family of Graphics Processors

2012-01-09T07:31:00.001-08:00

The recent firings from Advanced Micro Devices apparently reduced not only marketing and PR personnel, but also engineers. Fired in November, 2011, a hardware designer from AMD has disclosed the code-name of the company’s next-generation of graphics processors – Sea Islands – over at his profile in LinkedIn social network.

Alexander Shternshain, a former MTS design engineer at AMD, who is looking for a new job now, has unveiled the name of AMD’s future graphics processors: Sea Islands. The new family will likely emerge in 2012 or in 2013 and will be most probably be made using 28nm fabrication process at Taiwan Semiconductor Manufacturing Company. According to VR-Zone web-site, Sea Islands name “directly refers to the chain of islands on the USA's Atlantic coast”.

AMD’s Next-Gen Graphics Chips Code-Named “Sea Islands”

The recent firings from Advanced Micro Devices apparently reduced not only marketing and PR personnel, but also engineers. Fired in November, 2011, a hardware designer from AMD has disclosed the code-name of the company’s next-generation of graphics processors – Sea Islands – over at his profile in LinkedIn social network.

Alexander Shternshain, a former MTS design engineer at AMD, who is looking for a new job now, has unveiled the name of AMD’s future graphics processors: Sea Islands. The new family will likely emerge in 2012 or in 2013 and will be most probably be made using 28nm fabrication process at Taiwan Semiconductor Manufacturing Company. According to VR-Zone web-site, Sea Islands name “directly refers to the chain of islands on the USA's Atlantic coast”.

Based on his LinkedIn profile, Mr. Shternshain was involved in design of the latest AMD families of graphic processors starting with Evergreen and Northern Islands (Radeon HD) GPU families, Fusion APUs (Ontario, Llano, Krishna), and the Southern Islands and Sea Islands GPU families.

Actual specifications or peculiarities of the Sea Islands product family remain unknown. What is natural to expect is further development of GCN [graphics core next] architecture with the addition of multimedia-related features, gneral purpose computing and graphics performance improvements. The Sea Islands lineup will likely continue to belong to DirectX 11.1 generation of GPUs.

AMD did not comment on the news-story.

Source: xbitlabs.com

Kepler to show its face in Q1 2012

2012-01-02T09:10:00.000-08:00

Multiple sources have confirmed that Kepler, Nvidia’s GT 600 generation, is coming in Q1 2012, but so far we don’t have any better timeframe than that.

We would not be surprised to see Kepler in some form during CES, next week in Las Vegas, at least as a behind-closed-doors presentation for select members of the press and investors.

Sources close to Nvidia and AIB partners are telling us that Kepler is faster than Radeon HD 7970, but of course you would expect them to claim that their upcoming product is better.

Without any doubt, AMD is the first with a successful 28nm launch and it will enjoy some time at the top of the market, at least a couple of months.

Cebitish

Multiple sources have confirmed that Kepler, Nvidia’s GT 600 generation, is coming in Q1 2012, but so far we don’t have any better timeframe than that.

We would not be surprised to see Kepler in some form during CES, next week in Las Vegas, at least as a behind-closed-doors presentation for select members of the press and investors.

Sources close to Nvidia and AIB partners are telling us that Kepler is faster than Radeon HD 7970, but of course you would expect them to claim that their upcoming product is better.

Without any doubt, AMD is the first with a successful 28nm launch and it will enjoy some time at the top of the market, at least a couple of months.

We will keep our eyes peeled for any new developments, but it seems we will have to wait a bit before anything even close to official from Nvidia.

Source: fudzilla.com

AMD Radeon HD 7970 hits 1.26GHz and P10259 in 3DMark 11

2012-01-02T09:09:00.000-08:00

It appears that our friends over at VR-Zone have been ringing in the New Year with the excitement of some exceptional GPU hardware benchmarks on fresh 28nm AMD Southern Islands silicon, particularly of the enthusiast type.

The site was able to successfully overclock their AMD Radeon HD 7970 reference board from 1125MHz core speed to a 1267MHz core speed with a minor voltage bump from 1.15v to 1.25 volts using publicly available software and a special GPU BIOS (which VR-Zone claims will be revealed on Monday, January 9th).

Using minor voltage mod and special BIOS

It appears that our friends over at VR-Zone have been ringing in the New Year with the excitement of some exceptional GPU hardware benchmarks on fresh 28nm AMD Southern Islands silicon, particularly of the enthusiast type.

The site was able to successfully overclock their AMD Radeon HD 7970 reference board from 1125MHz core speed to a 1267MHz core speed with a minor voltage bump from 1.15v to 1.25 volts using publicly available software and a special GPU BIOS (which VR-Zone claims will be revealed on Monday, January 9th).

We were impressed with the 3DMark 11 Performance Preset scores, which improved from P8227 on stock clocks to P10259 with the voltage adjustment. Nevertheless, 3DMark 11 Extreme Preset scores also improved from X2764 on stock clocks to X3490 with the voltage adjustment.

We look forward to reporting on AMD’s 28nm Tahiti GPUs in person at CES 2012 in just one week from now. In the meantime, more preliminary benchmarks and details can be found here.

Source: fudzilla.com

AMD HD 7950 won't show up on January 9th

2012-01-02T09:07:00.000-08:00

In case the HD 7970 has a too steep price for your taste and you were looking forward to the Tahiti Pro HD 7950, we are afraid that we have some bad news as, according to our sources, there is simply no way that this one will be available on January 9th.

The scheduled launch date set at January 9th still stands, according to some major players in the AIB business, but they won't have the card ready by the 9th of January. We guess that we might see a couple of reviews on that date, but these cards will probably come directly from AMD. Unfortunately, there are no precise details regarding the delay and partners simply have no explanation. It seems the HD 7950 might show up somewhere around the Chinese New Year, or February.

Don't hold your breath

In case the HD 7970 has a too steep price for your taste and you were looking forward to the Tahiti Pro HD 7950, we are afraid that we have some bad news as, according to our sources, there is simply no way that this one will be available on January 9th.

The scheduled launch date set at January 9th still stands, according to some major players in the AIB business, but they won't have the card ready by the 9th of January. We guess that we might see a couple of reviews on that date, but these cards will probably come directly from AMD. Unfortunately, there are no precise details regarding the delay and partners simply have no explanation. It seems the HD 7950 might show up somewhere around the Chinese New Year, or February.

Of course, the Tahiti XT HD 7970 should start selling on January 9th, but for now the cheaper HD 7950 won't.

Source: fudzilla.com

HD 7900 gets first water block

2012-01-02T09:04:00.001-08:00

Although the new HD 7970 won’t be available until January 9, a German company has already started taking preorders for the first water blocks designed for AMD’s Tahiti cards.

Aquacomputer’s Aquagratix 7900 should fit the HD 7970 and HD 7950, and we suspect it will also work on the HD 7890 once it shows up. Of course, it will only fit reference PCBs.

The cooler appears milled from a single slab of copper and it features two standard G1/4” threads. As for quality, well it’s made in Germany. Case closed.

Don’t forget das Kälteschutzmittel

Although the new HD 7970 won’t be available until January 9, a German company has already started taking preorders for the first water blocks designed for AMD’s Tahiti cards.

Aquacomputer’s Aquagratix 7900 should fit the HD 7970 and HD 7950, and we suspect it will also work on the HD 7890 once it shows up. Of course, it will only fit reference PCBs.

The cooler appears milled from a single slab of copper and it features two standard G1/4” threads. As for quality, well it’s made in Germany. Case closed.

Other than that there’s not much to report, other than the €89 price tag. The blocks should be available in three weeks.

Source: fudzilla.com

Nvidia GPU roadmap slips a year too

2012-01-02T08:58:00.000-08:00

It looks like Nvidia has slipped their GPU roadmap a year, but that is news to no one. The biggest shock of them all is that the company actually admitted it slipped, even if it is two years after the fact.

The short story is that Nvidia has once again admitted what SemiAccurate has been saying for literally years, their GPU roadmap had no chance of being released on the schedule they were promising. Without getting in to the gory details, the story is exactly as we have been telling our consulting clients, the Fermi debacle has waterfall effects on the entire roadmap, and that is finally starting to show up in official documents.

Analysis: Admitting the no longer deniable in Santa Clara

It looks like Nvidia has slipped their GPU roadmap a year, but that is news to no one. The biggest shock of them all is that the company actually admitted it slipped, even if it is two years after the fact.

The short story is that Nvidia has once again admitted what SemiAccurate has been saying for literally years, their GPU roadmap had no chance of being released on the schedule they were promising. Without getting in to the gory details, the story is exactly as we have been telling our consulting clients, the Fermi debacle has waterfall effects on the entire roadmap, and that is finally starting to show up in official documents.

If you recall, Nvidia doesn’t give out long term roadmaps until they are way behind, and want to ‘prove’ that they are not. Until the Fermi debacle, there was never any comment on future roadmaps, but to ‘prove’ that all was well, the company started tossing them out like candy at a parade. The problem now is when those roadmaps can’t be met because they were never realistic, they become just stock price adjustment tools for the unwary analyst.

When the silicon doesn’t match the roadmap, the silicon is renamed, and the roadmap stays intact. The most recent example of this is Tegra 4, a chip that was code named T40. T40 had zero chance of coming out on time due to executive meddling, micromanagement, and engineering through whim, so Tegra 3.5 was inserted. T40 is now Tegra 5, T50 is now Tegra 6, and T35 was always meant to be there. Executive face saved, roadmap intact, nothing to see here, move along folks. The silicon for the roadmaps is completely different, but the slides are reusable.

That brings us to the latest slip, the GPUs. Remember that Fermi was promised to rooms full of analysts for 2009, and Jen-Hsun himself waved faked cards around to ‘prove’ it. Until a few weeks ago, the GPU roadmap shown off by the company was still insisting that the GPU came out in 2009, even though it was not released until Q2/2010. That curious situation didn’t change the fact that the roadmaps still claimed otherwise for all of 2010 and most of 2011.

Those same roadmaps claimed that Kepler will be a 2011 part, even though the company directly said that it will not be. Luckily at GTC Asia in mid-December, Nvidia corrected that slide, two years late, to say the same thing we have been telling the financial community for two plus years. It now seems that Fermi was actually 2010, Kepler will not be 2011, and Maxwell went from 2013 to 2014. Who knew? SemiAccurate’s clients for one, but no one at Nvidia management seems to have been let in on the joke.

The short version is that their roadmap slipped a year. Kepler is 2012, and Project Denver/T50 which is based on Maxwell, is not, and never had a chance to be a late 2012 product, much less early 2013. Then again, it was never meant to be, unless you were an analyst talking privately to the company, they got different timelines for some reason. Since Denver is based on a core that has now officially slipped from 2013 to 2014, well, you can do that math yourself.

In the end, it is refreshing to see Nvidia finally admitting that there is a slip, even if it is in a backhanded and circuitous route. Their roadmaps slipped a year, but that isn’t a change as much as a semi-public admission of the no longer deniable truth. I suppose congratulations are in order for this milestone, lets hope the personal growth in Santa Clara continues.

Source: semiaccurate.com

More HD 7000 series specs, prices revealed

2012-01-02T08:56:00.000-08:00

Earlier this week Fudo dug up some HD 7000 details, and just in time to end the miserable year that was 2011 in style, we’ve come across a few more.

We already wrote about Tahiti and New Zealand, and we mentioned mid-range and entry-level cards based on Pitcairn and Cape Verde GPUs. Now we have a few more details and we seem to have missed a couple of interesting cards.

In addition to the HD 7900 series, it appears Tahiti will also appear as an LE version, branded HD 7890. It packs 1.5GB of memory, 1536 stream processors, 96 texture units, 32 ROPs and a 216GB/s memory bus. Intrigued? We are, as the SRP is $399, a full $150 less than the Tahiti XT. It should show up in March.

Full frontal nudity

Earlier this week Fudo dug up some HD 7000 details, and just in time to end the miserable year that was 2011 in style, we’ve come across a few more.

We already wrote about Tahiti and New Zealand, and we mentioned mid-range and entry-level cards based on Pitcairn and Cape Verde GPUs. Now we have a few more details and we seem to have missed a couple of interesting cards.

In addition to the HD 7900 series, it appears Tahiti will also appear as an LE version, branded HD 7890. It packs 1.5GB of memory, 1536 stream processors, 96 texture units, 32 ROPs and a 216GB/s memory bus. Intrigued? We are, as the SRP is $399, a full $150 less than the Tahiti XT. It should show up in March.

Pitcairn will also be available in LE flavor, but only in the second quarter of 2012, under the HD 7790 brand. It has 1GB of memory, 1152 stream processors, 72 texture units and 24 ROPs. The memory bandwidth is 144GB/s and the price stands at $199. In addition, AMD will offer Pitcair Pro, HD 7850 cards with either 1GB of 2GB of memory, for $219 and $249 respectively. The former should launch in March.

So, no now we have the last pieces of the puzzle and mind you, AMD has quite a nice lineup for 2012. The race is on, and Nvidia will have to pull out all the stops in 2012 to hold its ground. This is probably the last story I’ll do this year, so happy New Year folks, enjoy your new 28nm toys.

Source: fudzilla.com

AMD Radeon HD 7970 Game Benchmarks

2012-01-02T07:50:00.000-08:00

A collections of real-world gaming tests using AMD Radeon HD 7970 Game Benchmarks, listed alphabetically.

"The AMD Radeon HD 7970 3GB graphics card chews through DX11 game titles, overclocks like a beast and is very power efficient!" -legitreviews.com

Happy New Year GPU-WARS!

2011-12-31T06:48:00.000-08:00

Have a Blessed Year 2012 to Everyone!

Have a Blessed Year 2012 to Everyone!

Source:

AMD Radeon HD 7970 justifies $549 price tag

2011-12-24T10:36:00.000-08:00

AMD has decided to launch its newest Radeon HD 7970 flagship graphics card with a US $549.99 price tag and we already wrote about this price early this month. If you consider the performance of the HD 7970 than a US $550 price tag seems reasonable as the card outperforms the competition and claims the highest performing single-GPU card crown.

Early this month, one of our trusted sources whispered to us the US $549.99 price tag and although it did make sense, considering the rumoured performance, lack of competition, we honestly expected AMD to prove us wrong. I think we, and most of the consumers for that matter, simply got used to the fact that AMD was somehow always giving a bit more for the money and trying to beat Nvidia by launching at somewhat lower prices and delivering as much bang for buck as possible.

Pricey, but...

AMD has decided to launch its newest Radeon HD 7970 flagship graphics card with a US $549.99 price tag and we already wrote about this price early this month. If you consider the performance of the HD 7970 than a US $550 price tag seems reasonable as the card outperforms the competition and claims the highest performing single-GPU card crown.

Early this month, one of our trusted sources whispered to us the US $549.99 price tag and although it did make sense, considering the rumoured performance, lack of competition, we honestly expected AMD to prove us wrong. I think we, and most of the consumers for that matter, simply got used to the fact that AMD was somehow always giving a bit more for the money and trying to beat Nvidia by launching at somewhat lower prices and delivering as much bang for buck as possible.

Nvidia wasn't far off with performance/price with some of its own cards but AMD was always sort of better in that segment (HD 6950, HD 6850, HD 5850), while Nvidia was more of a "performance no matter the cost" company. Some might agree and some might not with the previous comparison, but that's just the way things go as some will always prefer one company and their products over the other.

Let's get back to the matter at hand and the interesting US $549.99 price tag. AMD is targeting Nvidia's GTX 580 with this card and has priced it accordingly. Some might argue that it should be compared against the GTX 580 3GB version but we were having a hard time to find a single review that had the HD 7970 pitted against the GTX 580 3GB version, and we don't even want to start talking about how well did GTX 580 3GB performed when compared to standard 1.5GB version, so we'll stick to the standard GTX 580.

The GTX 580 is currently priced at around US $500 in the US and around €400 in Europe, while the HD 7970 is priced at US $549.99 (hopefully if retailers/e-tailers don't jack up the price). In Europe, it should retail/e-tail at below €500. Strangely, we still haven't heard the exact EU price for the HD 7970 except for those few early listings that were simply insanely priced. In the best case scenario the HD 7970 should end up at around €420 (if you simply convert the currency) but that is never the case since Europe always gets worse deal.

Unfortunately, the time where AMD was pushing for best bang for buck is over and honestly we can't blame AMD, as Nvidia is late to the game. There is no competition, and they have the fastest single-GPU, so basically Nvidia takes the blame. The HD 7970 is simply an enthusiast card meant for high resolution/multi-display gaming and you get a lower consumption and better compute power as an extra. Those that were hoping for a US $449 or even US $399 card should wait for mid-February, AMD's HD 7800 series or even the HD 7950 that could end up pretty good considering the performance of the HD 7970.

Of course, there is always the good old HD 6970 that currently sells for around US $340 in the USA and around €280 in Europe, but that won't last for long as that one will probably disappear from the market soon. We've been hearing that some partners have already stopped ordering them.

We have already heard rumours that GTX 580 might not stay at US $500 and a price cut seems like a logical move for Nvidia. Back when HD 7970 US $549.99 price was rumoured we already heard that the card could be cheaper so it might mean that AMD has a bit more room for slight price adjustment if Nvidia makes the move.

In any case, it all depends on what are you after, a top-class, best performing single-GPU enthusiast graphics card for some high-resolution/multi-display gaming, a best bang for buck card, or a budget/cheaper solution that can push all your games at 1080p.

Source: fudzilla.com

Skyrim gets up to 25% performance boost via GeForce beta drivers

2011-12-24T10:26:00.000-08:00

Nvidia has unleashed its latest GeForce beta driver with a few updates, including -- and perhaps most significantly -- performance optimizations for Skyrim players. Version 290.53 should offer GeForce owners up to 25% more frames per second. Nvidia reports a 25% boost over last month's 290.36 beta drivers while playing with cards such as the GeForce GTX 560 at 1920x1200 on Ultra quality settings.

Download GeForce 290.53 beta
Desktop: Windows XP 32-bit | Windows XP 64-bit | Windows Vista/7 32-bit | Windows Vista/7 64-bit
Mobile: Windows Vista/7 32-bit | Windows Vista/7 64-bit

That's a best case scenario, the company admits. The 25% gain was recorded indoors, while outdoor scenarios only run up to 2-3% better. Your mileage will vary depending on your specific GPU, system configuration and in-game location. Nonetheless, it's hard to complain about a double-digit increase indoors considering much of your time is spent in smaller environments, be that caves or buildings.

Skyrim players can also look forward to improved Ambient Occlusion, an Nvidia Control Panel quality setting that enables more realistic shadows. The company added Ambient Occlusion support for Skyrim with 290.36, but 290.53's effects are "faster" and should eliminate the tacky shimmering users have reported. Nvidia has also updated the 3D Vision crosshair to better match Skyrim's default 2D pointer.

There's not much else to see for folks without Bethesda's latest RPG. Build 290.53 includes SLI profiles for Afterfall: InSanity, Trine 2 and WRC 2: FIA World Rally Championship 2011 along with new or updated 3D Vision profiles for Trine 2, Orcs Must Die!, Stronghold 3 and a few lesser-known titles. The beta driver also resolves triangular artifacting in Battlefield 3 and a default panel resolution-timing bug.

Source: techspot.com

nVidia slashes prices to help welcome Radeon 7970

2011-12-24T10:16:00.000-08:00

There are few things in the universe funnier, than watching a couple of multi-billion dollar giants clubbing each other around the head with product launches, marketing schemes and price reductions.

It not only provides entertainment, it also keeps the market healthy.

KitGuru’s done a little price comparison, following the launch of AMD’s Radeon 7970. Here’s what we discovered in the mass market.

They’ve been around for ages, with strong industry relationships, so when one of the major online resellers has their nVidia pricing trumped by a high street major – to the tune of almost 25%, then you know something’s up.

Have a look at these screen grabs, taken very soon after each other, and you’ll see what we mean:-

The card we checked first was the multi-award winning Asus GT550.

KitGuru says: Dropping around 25% on a mainstream product is proper fighting talk. Looks like Jen Hsun and his Christmas Elves are ready for a blood bath ahead of the launch of Kepler in 2012. Whatever happens, price cuts like this, on award winning Asus cards, will only benefit the end user. That’s the KitGuru reader that is.

Source: kitguru.net

Exclusive: A sneek peek at the AMD VISION 2012 offerings

2011-12-24T10:01:00.000-08:00

So here is the AMD VISION 2012 lineup, basically what’s new?

The easiest thing to spot would be the AMD A10 series APU, since it’s positioned above the A8 series (as 10 is bigger than 8, duh?), we could easily draw an equal sign between A10 series APU and “Trinity”, the codename for the next mainstream-class (even performance-class) APU offering from AMD. Trinity will have 2 “Piledriver” modules, or 4 “Piledriver” CPU cores, a 7xxx series GPU codenamed “Devastator” which, in theory, providing 25% more shader groups over VLIW5 offering with the same shader count, and hopefully gets 25% more performance if the shader groups perform around the same.

Fun game inside!

So here is the AMD VISION 2012 lineup, basically what’s new?

The easiest thing to spot would be the AMD A10 series APU, since it’s positioned above the A8 series (as 10 is bigger than 8, duh?), we could easily draw an equal sign between A10 series APU and “Trinity”, the codename for the next mainstream-class (even performance-class) APU offering from AMD. Trinity will have 2 “Piledriver” modules, or 4 “Piledriver” CPU cores, a 7xxx series GPU codenamed “Devastator” which, in theory, providing 25% more shader groups over VLIW5 offering with the same shader count, and hopefully gets 25% more performance if the shader groups perform around the same. It supports faster system memory speeds to provide more bandwidth available to both the CPU cores and the GPU. Okay…

And then there is an E1 series, not sure if it’s a renamed E series or not, but the targeted audiences should be the same. So they should be Bobcat-based offerings, with a few tweaks to the CPU cores here and there (two more cores is kind of a tweak, too), with a GPU codenamed “Scrapper” sitting on the same die.

So here comes the fun little game that I mentioned in the subtitle.

Complete the game by drawing straight lines to match the items from the left column to the right column, some of the items have been matched as example. The answer should probably be announced sometime between CES 2012 and Computex/AFDS 2012 if things goes well, no prizes will be given though, due to unforeseen circumstances.

So what do you think the E1/E2 series will be? Will it be based on the old “Llano” APU as previous rumor mills suggested, or it will be 32/28 nm offering, or it will be something else? One thing for sure is that AMD already has the answers. While 2012 will continue to be interesting to watch, too.

Source: semiaccurate.com

AMD unchains Tahiti Pro cards from day one

2011-12-24T09:57:00.000-08:00

A few weeks ago, we brought you a few more tidbits on the AMD (NYSE:AMD) HD7000 line, and there is one more important bit. It has to do with the second card to launch, the Tahiti Pro.

If you recall, the SI/GCN/HD7000 launch goes Tahiti XT, Tahiti Pro, Pitcairn XT, Pitcairn Pro, spaced about a month apart, starting in January. In that previous article we said prices are also about $500, $400, $300, and $200 respectively, but one of those is purposefully off by a bit to catch plagiarizers like ATI-Forum.de.

Back to the new news, and it concerns the Tahiti Pro card. Word has reached SemiAccurate that Tahiti Pro will be unconstrained to the normal reference designs.

Reference designs are for the weak

A few weeks ago, we brought you a few more tidbits on the AMD (NYSE:AMD) HD7000 line, and there is one more important bit. It has to do with the second card to launch, the Tahiti Pro.

If you recall, the SI/GCN/HD7000 launch goes Tahiti XT, Tahiti Pro, Pitcairn XT, Pitcairn Pro, spaced about a month apart, starting in January. In that previous article we said prices are also about $500, $400, $300, and $200 respectively, but one of those is purposefully off by a bit to catch plagiarizers like ATI-Forum.de.

Back to the new news, and it concerns the Tahiti Pro card. Word has reached SemiAccurate that Tahiti Pro will be unconstrained to the normal reference designs. If you recall, most GPU manufacturers will force AIBs to make cards based on the reference design for the first 3 months or so, and there are a variety of very good business reasons to do this.

Unfortunately, it leads to a problem where the reviews all are the same, mainly because all the cards are the same. The main difference between manufacturers comes down to what color the AIB decides to put on the chrome bikini of the girl with the big sword riding the mythical beast just below their logo. We are partial to Hafnium bikini’s on women riding giant Were-moles around here. Luckily, Tahiti Pro changes this.

Yes, we are hearing from multiple sources that the junior card of the two will be able to use non-reference designs from day one. Instead of having the same card over and over in head to head testing, you will actually see real differences. If you look at Max’s review of the 6850, you can see the Sapphire version is quite different from the reference design. It also runs significantly cooler, significantly quieter, and probably costs less to make. It is a win/win, and Tahiti Pro will get this treatment at the start. This qualifies as a ‘good thing’ for all involved.

Source: semiaccurate.com

Radeon HD 7000 is 384-bit

2011-12-07T06:30:00.000-08:00

On top of the new 28nm manufacturing process, courtesy of TSMC, AMD’s new HD 7000 series will also feature other exciting novelties.

We have learned that high end Southern Islands parts will feature a 384-bit memory interface. We‘ve already told you in that the memory of choice is GDDR5.

Naturally some mainstream and entry level parts will still feature a slower, cut down version of the memory engine, due to necessity for lower TDPs and production costs, but most of you know how it goes. Some parts end up with 384-bit memory, yet others, like mobile GPUs for example have to make do with 128-bit. Fewer bits translate into lower TDPs and cheaper cards.

High-end SKUs, of course

On top of the new 28nm manufacturing process, courtesy of TSMC, AMD’s new HD 7000 series will also feature other exciting novelties.

We have learned that high end Southern Islands parts will feature a 384-bit memory interface. We‘ve already told you in that the memory of choice is GDDR5.

Naturally some mainstream and entry level parts will still feature a slower, cut down version of the memory engine, due to necessity for lower TDPs and production costs, but most of you know how it goes. Some parts end up with 384-bit memory, yet others, like mobile GPUs for example have to make do with 128-bit. Fewer bits translate into lower TDPs and cheaper cards.

Our sources are still telling us that the official launch is early next year and only rebrands will make it to this year, not the actual 28nm parts.

Source: fudzilla.com

Nvidia refreshes mobile GPU lineup

2011-12-07T06:28:00.000-08:00

Nvidia rolled out a few new 6xx series mobile GPUs on Tuesday, but frankly there is not much to talk about. The GT 635M, GT630M and 610M are merely revamped 5xx series parts, basically a low-end stopgap until 28nm parts come online.

Even the clocks remained unchanged, so we are basically looking at GF106/108 and GF119 cores. The GT 635M is basically a GT 555M with a new sticker, the GT630 is the GT 540M, while the 520MX is now 610M.

The designations seem to indicate that Nvidia is leaving room for future 645 or 655 parts, probably based on new 28nm cores. There is still no word on any upcoming high-end models.

When we say refresh, we mean rebrand

Nvidia rolled out a few new 6xx series mobile GPUs on Tuesday, but frankly there is not much to talk about. The GT 635M, GT630M and 610M are merely revamped 5xx series parts, basically a low-end stopgap until 28nm parts come online.

Even the clocks remained unchanged, so we are basically looking at GF106/108 and GF119 cores. The GT 635M is basically a GT 555M with a new sticker, the GT630 is the GT 540M, while the 520MX is now 610M.

The designations seem to indicate that Nvidia is leaving room for future 645 or 655 parts, probably based on new 28nm cores. There is still no word on any upcoming high-end models.

Source: fudzilla.co

AMD May Soon Launch Radeon HD 6930

2011-12-07T06:07:00.000-08:00

According to Chinese site MyDrivers, AMD may be releasing a Radeon HD 6930 card, code-named Cayman CE, which will get 1280 stream processors and 80 texture units, down from 1408 and 88 on the Radeon HD 6950. The GPU clock will drop to 750 MHz and the memory will fall to 4800 MHz.

The board will be available with a 256-bit memory interface and a choice of 1 GB or 2 GB of GDDR5 memory. Expect the 6930 to be priced below $200.

In related news, Fudzilla also got its hands on the upcoming Radeon HD 7000 supposed pricing. The actual numbers should not be too surprising. The HD 7950 will apparently be debut with a target price of $449, while high-end HD 7970 cards should cost $549.

It appears that AMD is preparing a new variant of the Radeon 6000-series to compete with Nvidia's GTX 560 graphics card.

According to Chinese site MyDrivers, AMD may be releasing a Radeon HD 6930 card, code-named Cayman CE, which will get 1280 stream processors and 80 texture units, down from 1408 and 88 on the Radeon HD 6950. The GPU clock will drop to 750 MHz and the memory will fall to 4800 MHz.

The board will be available with a 256-bit memory interface and a choice of 1 GB or 2 GB of GDDR5 memory. Expect the 6930 to be priced below $200.

In related news, Fudzilla also got its hands on the upcoming Radeon HD 7000 supposed pricing. The actual numbers should not be too surprising. The HD 7950 will apparently be debut with a target price of $449, while high-end HD 7970 cards should cost $549.

Source: tomshardware.com

GTX 560 448 officially launches on Nov 29th

2011-11-27T08:29:00.000-08:00

Nvidia and Zotac are not happy about the GTX 560 448 picture and specs we posted yesterday and now we got some more info about the card.

The chip behind the card is a castrated GTX 570 and the card supports three-way SLI, has a 320-bit memory bus and 1280MB of memory as we said in the previous post.

Performance wise, it will come just slightly short of GTX 570 but we could not find the price for it, although it looks like it will cost about €250. We will confirm this as soon as the price gets fixed.

Asus triple DCII cooler too

Nvidia and Zotac are not happy about the GTX 560 448 picture and specs we posted yesterday and now we got some more info about the card.

The chip behind the card is a castrated GTX 570 and the card supports three-way SLI, has a 320-bit memory bus and 1280MB of memory as we said in the previous post.

Performance wise, it will come just slightly short of GTX 570 but we could not find the price for it, although it looks like it will cost about €250. We will confirm this as soon as the price gets fixed.

The launch date is November 29th, again a Tuesday launch. Asus also has a 448-core card on the way, probably the rest of the gang as well, but we have heard that Asus comes with a triple slot DCII cooler, the same one that they use on the Asus 570 DCII card.

Source:

AMD Catalyst 11.11a boosts Skyrim performance, more

2011-11-27T08:25:00.000-08:00

AMD has followed up on their 11.11 release from just a few days ago. The newly relased Catalyst 11.11a "performance" driver addresses a number of performance issues as well as a couple of bugs found in 11.11.

Rage and Skyrim get most of the love from 11.11a, but Batman Arkham City and Battlefield 3 were also singled out with bug fixes. The small number of improvements listed here are in addition to a long list of changes made in 11.11. AMD has since pulled 11.11 from their downloads section and replaced it with 11.11a.

Batman Arkham City
• Improves DirectX 11 performance for single GPU and CrossFireX configurations

Rage
• Delivers AMD CrossfireX performance scaling
• Resolves a number of image / stability issues seen with the title:
• Fix geometry corruption, sometimes seen in Bash TV entrance
• Fix issues with Low-memory conditions on 32bit systems.
• Fix issue with extreme corruption with missing textures on 32bit systems.
• Fix memory leaks when deleting/reusing sync objects.
• Fix hitching and pausing, especially noticeable on some Quad Core systems when doing races and Stanley Express runs.
• Fix some missing shadows

Elder Scrolls Skyrim
• Improves performance 2-7% on single GPU configurations
• Resolve corruption seen when enabling Multi-Sample Anti-Aliasing on the AMD Radeon HD 6970 Series
• Disables CrossFireX (to resolve negative scaling and image quality issues seen when CrossFireX is enabled)
• AMD is continuing to work on a CrossFireX solution for Skyrim – as soon as it’s ready we’ll make it publicly available

Battlefield 3
• Resolves intermittent corruption seen when playing the game at specific camera angles

Noteworthy improvements from the previous 11.11 release include Flash 11 hardware acceleration, various operating system fixes, improved multi-monitor handling and a multitude of game fixes for titles like Dragon Age 2, Shogun 2, Far Cry 2, Rage, Homefront, Dead Space 2, Bulletstorm, Civilization V and more.

The 11.11a driver package can be downloaded for both 32 and 64-bit versions of Windows XP and Vista/7. AMD recommends this release for Radeon 5000 and 6000 HD owners only, however, it should install for most not-so-older chipsets as well. If you have problems getting it installed on your aging board, modded versions are already floating around the web.

Most laptops sporting a Radeon 5000 or 6000 HD series chip are compatible with 11.11a as well, although support will vary from model to model. Toshiba, Panasonic and some Sony laptops are called out specifically for their incompatibility.

Source: techspot.com

AMD Radeon 7000M to come in December

2011-11-27T08:19:00.000-08:00

According to a report at Computerbase.de, AMD's entire Radeon HD 7000M series should be launched in December and it is a quite an extensive list of SKUs ranging from low-end Seymour to high-end Wimbledon GPUs.

The upcoming AMD London series lineup features a total of five different GPUs that will be available in an extensive number of SKUs depending on the clock, amount and type of memory. The low-end part is based on the Seymour GPU that will be avialable in couple of Pro, XT and XTX SKUs variations and clocked between 600 and 800MHz and paried up with 512MB/1GB 64-bit DDR3 (in one case GDDR5 memory). The low-end offer will bear the Radeon HD 7400M series name.

Full London series lineup

According to a report at Computerbase.de, AMD's entire Radeon HD 7000M series should be launched in December and it is a quite an extensive list of SKUs ranging from low-end Seymour to high-end Wimbledon GPUs.

The upcoming AMD London series lineup features a total of five different GPUs that will be available in an extensive number of SKUs depending on the clock, amount and type of memory. The low-end part is based on the Seymour GPU that will be avialable in couple of Pro, XT and XTX SKUs variations and clocked between 600 and 800MHz and paried up with 512MB/1GB 64-bit DDR3 (in one case GDDR5 memory). The low-end offer will bear the Radeon HD 7400M series name.

The mid-range offer is based on the Thames GPU and it is quite an extensive one as it features a total of ten different SKUs. It will bear the Radeon HD 7600M and Radeon HD 7500M series name. The lineup starts with the Thames LE and Thames LP GPUs clocked at 450MHz and paired up with either 128-bit or 64-bit 1GB of DDR3/GDDR5 memory.

The top part of this mid-range offer is represented by Thames Pro and Thames XT GPUs clocked at 600MHz and paired up with 1GB of 128bit GDDR5/DDR3 memory. These four SKUs will be quite easily differentiated as the Thames Pro will carry Radeon HD 7570M (DDR3) and HD 7590M (GDDR5) name while the Thames XT will be known as the HD 7670M and the Radeon HD 7690M.

The high-end of the offer starts with Chelsea GPU that will be available in XT and Pro variation with clocks ranging from 450-550 to 600-700MHz. These should feature 2GB of GDDR5 memory paired up with a 128-bit memory interface and will be known under Radeon HD 7700M name. The next in line is the Heathrow, known as the Radeon HD 7800M series. It will feature the same amount of memory, higher clocks and probably more stream processors when compared to the Chelsea GPU.

The real high-end will be known under the Radeon HD 7900M brand and represented by two Wimbledon GPU variations. The Wimbledon Pro should be clocked at somewhere between 650 and 700MHz and paried up with 2GB of GDDR5 256-bit memory, while the XT variant should end up at between 700 and 750MHz with up to 4GB of GDDR5 256-bit memory.

Of course, you should take these informations with a pinch of salt due to such amount of SKUs and interesting rumoured branding scheme. It is important to note that all GPUs, except Seymour, should be based on the 28nm manufacturing process. The Seymour on the other hand is simply a rebrand of the Radeon HD 6400M series.

Of course, you can expect a similar lineup from Nvidia under the Geforce 600M series lineup and we might see them at CES 2012 in January with the launch date probably towards March and Ivy Bridge.

Source: fudzilla.com

More bits on HD7000/Southern Islands/GCN leak

2011-11-19T21:31:00.000-08:00

It looks like AMD (NYSE:AMD) is planning on launching desktop HD7000 GPUs in January, and SemiAccurate just got a few more bits about them. There isn’t much new, January launch for Tahiti XT, followed by Tahiti Pro a month later, then Pitcairn XT in March, Pro in April.

The cards themselves are known as GCN (Graphics Core Next), Southern Islands, or HD7000, depending on who you talk to. All of them are made on TSMC’s 28nm process, and are cousins to the laptop lines we told you about earlier. Pricing is tentatively set for around $500 for the top Tahiti part, $400 for the Pro, $300 for Pitcairn XT, $200 for the Pro, all subject to much change and our minor but patented SPMOODT(TM)(R)(C)(P) (Source Protecting Minor Obfuscation Of Data Technology).

nVidia’s quick move to 28nm secures laptop business for 2012

2011-11-19T21:14:00.000-08:00

The weakened processing capability of the iPad, combined with incredible battery life, has shown the work where the future lies. Laptop manufacturers need to be able to show extensive battery life – alongside a full feature set.

While the Tegra 3 chip is looking to be a bit of a darling in the ultra-portable/entry-level space, nVidia sees Kepler technology winning back some of the mainstream laptop market.

As KitGuru regulars know, the UK market is now almost 3:1 in favour of laptops over desktops. Once a 1.3 million unit business, it’s possible that the UK market for desktop graphics cards could dip below 600,000 for the first time in 2012.

With all of the growth coming in the mobile sector, performance per watt will become key. AMD’s DX11 launch ensured a huge amount of 3D success in both the desktop and mobile space, but nVidia has been fighting back – while Intel pushes hard to improve its own offering at the same time.

So what is being claimed? Well, nVidia’s move to the 500 series was a marked improvement over the original Fermi 400 series. For the Kepler/600 series, they have targeted themselves to achieve a 300% improvement in performance per watt. Laudable ambition – but only time (and the KitGuru Lab) will be able to tell if it has succeeded.

Either way, the consumer is going to be a winner, with the kind of mobile solution being offered at Christmas 2012 looking certain to be streets ahead of December 2011.

Source: kitguru.net

Volume shipments of 28nm AMD graphics in 2012

2011-11-05T12:50:00.000-07:00

We spoke with a few knowledgeable people and the final take on 28nm graphics is that both AMD and Nvidia could launch some in 2011, but even if they manage, the yields and availability will be nothing short of catastrophic.

TSMC simply didn't get its act together with 28nm transition, which looks more challenging than the company originally anticipated. Globalfoundries have the same issues as 28nm seems to be a pretty tough nut to crack.

AMD is still considering whether they should launch new graphics in late 2012 to cheer up investors, as it would made them look good, but if this happens, make sure to buy one the first day as they will sell out like hot cakes.

Late 2011 launch still an option

We spoke with a few knowledgeable people and the final take on 28nm graphics is that both AMD and Nvidia could launch some in 2011, but even if they manage, the yields and availability will be nothing short of catastrophic.

TSMC simply didn't get its act together with 28nm transition, which looks more challenging than the company originally anticipated. Globalfoundries have the same issues as 28nm seems to be a pretty tough nut to crack.

AMD is still considering whether they should launch new graphics in late 2012 to cheer up investors, as it would made them look good, but if this happens, make sure to buy one the first day as they will sell out like hot cakes. They will be better than current 40nm chips that are with us for years now, but at this time we cannot give any conclusive advice telling you who will end up faster, AMD or Nvidia.

Nvidia's new chips are almost certainly destined for 2012 launch, closer to Ivy Bridge, which means that Nvidia should be ready with its Kepler generation by March / April time. By that time yields and production should get better, accent is on should, as 28nm transition was supposed to be smooth, and guess what, it is proving anything but smooth.

Games like new Call of Duty, Battlefield 3 are definitely helping both notebook and desktop graphics sales. However, having the new generation ready in late Q1, possibly even in early Q2 2012, doesn't really help many people who like to do their gaming and upgrades between Thanksgiving and New Year Eve. Come springtime, many people have other priorities on their mind.

Source: fudzilla.com

Report: AMD's Next-Gen GPU to Boost Performance by Two Times, Will Be Delayed

2011-11-05T12:45:00.000-07:00

Although Advanced Micro Devices said for a number of times that its first graphics processing units (GPUs) made using 28nm process technology will be released in 2011. A new report, however, suggests that the company's high-performance GPU made at 28nm node will only become available in the first quarter of 2012. The good news is that it will deliver two times higher performance compared to existing solutions.

As it appears, AMD this year will not be able to please PC gamers with a new high-end graphics solution as its most powerful Southern Island chip will only become available in the first quarter of 2012, according to Rage3D web-site.

AMD's High-Performance Next-Generation Radeon HD 7000-Series GPU Set to Be Unveiled in Q1 2012 - Report

Although Advanced Micro Devices said for a number of times that its first graphics processing units (GPUs) made using 28nm process technology will be released in 2011. A new report, however, suggests that the company's high-performance GPU made at 28nm node will only become available in the first quarter of 2012. The good news is that it will deliver two times higher performance compared to existing solutions.

As it appears, AMD this year will not be able to please PC gamers with a new high-end graphics solution as its most powerful Southern Island chip will only become available in the first quarter of 2012, according to Rage3D web-site. Unofficial sources reportedly indicate that the problem is on the production side and Taiwan Semiconductor Manufacturing Company is unable to ramp up the 28nm GPUs in volume in Q4 2011.

Fortunately, the new high-end single GPU offering is projected to bring performance levels close to the Radeon HD 6990 dual-GPU flagship card, which means that performance enthusiasts will be interested to buy it even in the first quarter, a slow season for graphics cards.

It remains to be seen whether AMD Radeon HD 7000-series will rely on VLIW4 architecture (Cayman-like), will sport a new graphics/compute architecture or something hybrid. It will be interesting to find out whether AMD will implement unified address space for CPU and GPU cores and/or other enhancements. What is for sure is that the Southern Islands will belong to DirectX 11-generation hardware as well as sport a new video processing engine.

Source: xbitlabs.com

AMD rolls out new FirePro professional graphics card

2011-11-05T12:29:00.000-07:00

AMD has added yet another graphics card to its professional FirePro lineup, the entry-level V4900. Featuring a single slot design, dual-link DVI and two DisplayPort 1.2 outputs the new FirePro will be available in select Dell, Fujitsu and HP workstations and as a standalone product.

Spec wise, the new card features 480 stream processors, has 1GB of GDDR5 memory paired up with a 128-bit memory interface and supports up to six displays via DVI and two DisplayPort outputs, all thanks to Eyefinity tech. The new FirePro V9400 features support for DirectX 11, OpenGL and Shader Model 5.0 and has a maximum TDP of 75W.

The FirePro V4900 with 480 stream processors

AMD has added yet another graphics card to its professional FirePro lineup, the entry-level V4900. Featuring a single slot design, dual-link DVI and two DisplayPort 1.2 outputs the new FirePro will be available in select Dell, Fujitsu and HP workstations and as a standalone product.

Spec wise, the new card features 480 stream processors, has 1GB of GDDR5 memory paired up with a 128-bit memory interface and supports up to six displays via DVI and two DisplayPort outputs, all thanks to Eyefinity tech. The new FirePro V9400 features support for DirectX 11, OpenGL and Shader Model 5.0 and has a maximum TDP of 75W.

As noted, the card will be available in select workstations but will also be sold as a standalone product with a $189 price tag.

Source: fudzilla.com

AMD rolls out new 11.10 Catalyst driver

2011-11-05T11:33:00.000-07:00

AMD has rolled out the latest incarnation of the Catalyst driver, the Catalyst 11.10. Although it doesn't feature many performance improvements, other than those for Rage and Battlefield 3 that were already available, the new driver brings a few new enhancements for those handful of enthusiasts who are using AMD Eyefinity, as well as some other AMD VECC/CCC enhancements.

The new driver supports all the Radeon HD cards since the Radeon HD 2000 series as well as Radeon 3000 and 4000 series chipset. The new Catalyst pack features AMD display driver v8.902, AMD's HydraVision driver, Southbridge/IXP driver and AMD Catalyst Control Center v8.920 as well as AMD Vision Engine Control Center v8.902.

AMD Eyefinity and VECC/CCC enhancements

AMD has rolled out the latest incarnation of the Catalyst driver, the Catalyst 11.10. Although it doesn't feature many performance improvements, other than those for Rage and Battlefield 3 that were already available, the new driver brings a few new enhancements for those handful of enthusiasts who are using AMD Eyefinity, as well as some other AMD VECC/CCC enhancements.

The new driver supports all the Radeon HD cards since the Radeon HD 2000 series as well as Radeon 3000 and 4000 series chipset. The new Catalyst pack features AMD display driver v8.902, AMD's HydraVision driver, Southbridge/IXP driver and AMD Catalyst Control Center v8.920 as well as AMD Vision Engine Control Center v8.902.

The new driver doesn't bring any new performance improvements but does fix some issues found under Windows 7, Vista or XP OS and game titles like Mafia II, Lost Planet 2, F1 2010, Dirt 2 and other applications. The Catalyst pack does include some new features like the AMD Eyefinity enhancements including support for Eyefinity 5x1 portrait and landscape display configurations, increased maximum resolution to 16000x16000 on AMD's Radeon HD 6000 series and DX11 applications, and ability to set bezel compensation when using sets of displays that have mismatched pixel densities.

The AMD VECC/CCC enhancements include the improvements in the AMD Overdrive and Information Center parts of the the VECC/CCC.

As always, the driver is available for Windows 7, Vista and XP in both 32 and 64-bit versions. You can find it here.

Source: fudzilla.com

Powercolor's Devil 13 HD 6970 goes full monty

2011-10-29T19:25:00.000-07:00

We managed to get our hands on more details regarding Powercolor's upcoming Devil 13 HD 6970 card. Touted as a card that will feature top notch components, custom PCB and custom dual slot cooler, we now managed to get some more details regarding those features.

As noted earlier, the card has a 12-phase design packed on a black custom PCB. The feature list start with dual BIOS button located on the I/O part of the card as well as voltage measurement points. The PCB will include UltraHigh Current power bead-UHB that have reduced energy loss, twice longer lifespan and 15% lower temperature as well as Proadlizer chip and DirectFET that should all make this card a potential overclocking beast.

Naked PCB pictured, features detailed

We managed to get our hands on more details regarding Powercolor's upcoming Devil 13 HD 6970 card. Touted as a card that will feature top notch components, custom PCB and custom dual slot cooler, we now managed to get some more details regarding those features.

As noted earlier, the card has a 12-phase design packed on a black custom PCB. The feature list start with dual BIOS button located on the I/O part of the card as well as voltage measurement points. The PCB will include UltraHigh Current power bead-UHB that have reduced energy loss, twice longer lifespan and 15% lower temperature as well as Proadlizer chip and DirectFET that should all make this card a potential overclocking beast. Another neat feature is the LED lightning signal that shows the working status of the GPU.

Spec wise, the new Devil 13 HD 6970 card still has 1536 stream processors and works at 960MHz for the GPU and 5700MHz for 2GB of GDDR5 memory paired up with a 256-bit memory interface. These are the performance clocks and will be flashed into second BIOS while the normal one takes the card down to 880MHz for GPU and 5500MHz for memory.

The card is still expected to have that large dual-slot cooler with 92mm fans and four 8mm copper heatpipes. Of course, all these top-notch components and features come with a premium price tag, so the new Devil 13 HD 6970 is currently listed in Europe at €428.

Source: fudzilla.com

EVGA rolls out a new GTX 580

2011-10-29T19:21:00.000-07:00

EVGA has added yet another graphics card to its GTX 500 series lineup, the GTX 580 Batman Arkham City edition.

The new GTX 580 is pretty much a standard EVGA GTX 580 Superclocked slapped with Batman: Arkham City sticker and bundled with that game. The card still has 512 CUDA cores and works at 797MHz for the GPU and 1594MHz for shaders. The new GTX 580 has 1536MB of GDDR5 memory clocked at 4050MHz and paired up with a 384-bit memory interface. The card still has those standard two DVI and mini HDMI outputs.

With Batman: Arkham City game

EVGA has added yet another graphics card to its GTX 500 series lineup, the GTX 580 Batman Arkham City edition.

The new GTX 580 is pretty much a standard EVGA GTX 580 Superclocked slapped with Batman: Arkham City sticker and bundled with that game. The card still has 512 CUDA cores and works at 797MHz for the GPU and 1594MHz for shaders. The new GTX 580 has 1536MB of GDDR5 memory clocked at 4050MHz and paired up with a 384-bit memory interface. The card still has those standard two DVI and mini HDMI outputs.

The card is currently available at Newegg.com with a price set at US $529.99.

Source: fudzilla.com

Nvidia Preps GF110-based GeForce GTX 560 Ti‎ Upgrade

2011-10-29T19:15:00.000-07:00

This new Ti will, if VR-Zone is right, be based on the GF110 chip, which makes it a close relative to the 570 and 580 models. It will also mean a drop of the current GF114.

The publication reports that the new 560 Ti will essentially be a 570 with 14 (and one disabled) streaming multiprocessor with a total of 448 cores, up from the current 384. It is more than likely that the clock speed will be below that of the 570. Expect to see the 320-bit memory interface with 56 TMUs and 40 ROPs.

The likely performance gain is pure speculation at this time, but I will make an educated guess that the card will be consuming more power than the 170 W of the current 560 Ti and move closer to the 219 W of the 570.

Nvidia is apparently planning to release an updated and upgraded version of its GTX 560 Ti graphics card reference model.

This new Ti will, if VR-Zone is right, be based on the GF110 chip, which makes it a close relative to the 570 and 580 models. It will also mean a drop of the current GF114.

The publication reports that the new 560 Ti will essentially be a 570 with 14 (and one disabled) streaming multiprocessor with a total of 448 cores, up from the current 384. It is more than likely that the clock speed will be below that of the 570. Expect to see the 320-bit memory interface with 56 TMUs and 40 ROPs.

The likely performance gain is pure speculation at this time, but I will make an educated guess that the card will be consuming more power than the 170 W of the current 560 Ti and move closer to the 219 W of the 570.

Current 560 Ti graphics cards sell for street prices in the $220 to $270 range. An upgrade could breath more life into the card, which should help Nvidia move more chips during the upcoming Christmas season.

Source: tomshardware.com

Sapphire launches HD 6970 Battlefield 3 Edition

2011-10-29T19:02:00.000-07:00

Sapphire has let loose yet another card that will be a part of its HD 6900 series lineup, the Sapphire Radeon HD 6970 Battlefield 3 Special Edition. Obviously bundled with a Battlefield 3 title, the new card will feature both Sapphire's Flex technology and its praised Vapor-X cooler.

The new HD 6970 from Sapphire still has 1536 Stream Processors and works at 880MHz for the GPU and 5500MHz for 2GB of GDDR5 memory paired up with a 256-bit memory interface. In order to make this card a bit more interesting, Sapphire has also included the dual-BIOS switch that pushes the GPU clock up to 930MHz.

Vapor-X and FleX on the same card

Sapphire has let loose yet another card that will be a part of its HD 6900 series lineup, the Sapphire Radeon HD 6970 Battlefield 3 Special Edition. Obviously bundled with a Battlefield 3 title, the new card will feature both Sapphire's Flex technology and its praised Vapor-X cooler.

The new HD 6970 from Sapphire still has 1536 Stream Processors and works at 880MHz for the GPU and 5500MHz for 2GB of GDDR5 memory paired up with a 256-bit memory interface. In order to make this card a bit more interesting, Sapphire has also included the dual-BIOS switch that pushes the GPU clock up to 930MHz.

As noted, the card features Sapphire's FleX technology, which means that you'll be able to hook it up to three DVI monitors (third via supplied HDMI to DVI cable) for some Eyefinity fun, and all without those special adapters. This time around, Sapphire has decided to use its, quite well famous, dual-slot Vapor-X cooler that should keep the card well cooled inside mostly any case.

The card features two DVI, HDMI and two mini Displayport outputs and, as noted, will be shipped bundled with Dice's Battlefield 3 title. Unfortunately, the price or the actual availability date haven't been mentioned, but we'll keep an eye out for this one.

Source: fudzilla.com

Nvidia Unveils 3D Vision 2: Stereoscopic 3D Images Get Brighten

2011-10-19T16:19:00.000-07:00

Nvidia Corp. has introduced a new version of its stereoscopic 3D (S3D) technology which improves image quality of S3D visuals. The 3D Vision 2 technology sports new larger active-shutter glasses as well as requires next-generation S3D monitors that support Nvidia 3D LightBoost capability.

Nvidia 3D Vision 2 glasses provide deeper immersion in games through lenses that are 20% larger than those in first-generation glasses, resulting in a wider viewing area and increased external light blocking.

Nvidia Improves Quality of Stereo-3D Images with 3D Vision 2

Nvidia Corp. has introduced a new version of its stereoscopic 3D (S3D) technology which improves image quality of S3D visuals. The 3D Vision 2 technology sports new larger active-shutter glasses as well as requires next-generation S3D monitors that support Nvidia 3D LightBoost capability.

Nvidia 3D Vision 2 glasses provide deeper immersion in games through lenses that are 20% larger than those in first-generation glasses, resulting in a wider viewing area and increased external light blocking. In addition, Nvidia 3D Vision 2 glasses are made of soft composite materials for a more comfortable fit with gaming headphones. The new glasses utilize advanced active-shutter and IR wireless technologies.

Nvidia 3D Vision 2 glasses are fully backwards-compatible with all existing Nvidia 3D Vision Ready content and supporting products, including more than 70 different 3D Vision monitors, notebooks and projectors, video applications, and cameras.

The Nvidia 3D Vision 2 glasses kit, which includes one pair of 3D Vision 2 glasses and a wireless USB IR emitter, will be available worldwide in October for $149. Extra Nvidia 3D Vision 2 glasses are available for $99

Nvidia 3D LightBoost is a new display technology that delivers up to two times brighter 3D images than existing 3D solutions and improved color quality. It also increases environmental lighting, making gaming keyboards and mice more visible, and reducing 3D ghosting.

The first Nvidia 3D LightBoost-certified desktop display is the Asus VG278H, a 27" LED full HD (1920x1080) monitor with 120Hz refresh rates and 2ms response times, featuring a 3D HDMI 1.4 input, as well as a dual-link DVI input for full HD 1080p 3D gaming. Expected to be available by the end of October 2011, the Asustek VG278H includes Nvidia 3D Vision 2 glasses and is priced at $699.

Additional Nvidia 3D LightBoost-compatible displays from Acer and BenQ are expected to ship in the coming months. Nvidia 3D LightBoost compatible notebooks include several Toshiba models - Satellite P770/P775, Dynabook Satellite T572, Dynabook T572, and the Qosmio X770/X775 - with others expected in the coming months.

Source: xbitlabs.com

Report: First AMD 28nm GPU Due in December

2011-10-19T16:07:00.000-07:00

Industry sources told Germany's heise online that AMD is targeting December 6 as a launch date.

There was no information which architecture these new GPUs will be based on, but there appears to be reason to believe that AMD may be unveiling performance desktop boards, which will become available in very limited quantities in December, heise said. Due to 28 nm transition problems, it may take some time until there is enough supply to meet demand.

If heise's sources are correct, then we assume that the announcement in December will be about AMD's Southern Island GPU family, which is expected to be called the Radeon HD 7000 series. We are expecting these GPUs to use Rambus' XDR2 DRAM memory instead of GDDR5, MIMD instructions instead of VLIW, support for PowerTune as well as x86 addressing and Partially Resident Textures.

AMD may be unveiling its first 28 nm graphics chips before the end of the year.

Industry sources told Germany's heise online that AMD is targeting December 6 as a launch date.

There was no information which architecture these new GPUs will be based on, but there appears to be reason to believe that AMD may be unveiling performance desktop boards, which will become available in very limited quantities in December, heise said. Due to 28 nm transition problems, it may take some time until there is enough supply to meet demand.

If heise's sources are correct, then we assume that the announcement in December will be about AMD's Southern Island GPU family, which is expected to be called the Radeon HD 7000 series. We are expecting these GPUs to use Rambus' XDR2 DRAM memory instead of GDDR5, MIMD instructions instead of VLIW, support for PowerTune as well as x86 addressing and Partially Resident Textures.

TSMC reportedly increased its pricing for 28nm products as it noticed substantially higher demand than the company initially expected. Xbit labs speculated that TSMC's capacity for 28nm wafer production is only 7000 to 10,000 units during the current fourth quarter.

This isn't the first time we've heard about the Radeon HD 7000 series though.

Source: tomshardware.com

AMD Releases New Preview Driver for Battlefield 3, Rage

2011-10-08T18:13:00.000-07:00

Just in time for the weekend, AMD has unleashed an updated preview driver (version 2) of Catalyst 11.10.

Those playing the Battlefield 3 Open Beta or the newly released Rage with AMD Radeon HD 6000 and HD 5000 will like the features of this preview driver:

New Radeon HD drivers for your weekend gaming sessions.

Just in time for the weekend, AMD has unleashed an updated preview driver (version 2) of Catalyst 11.10.

Those playing the Battlefield 3 Open Beta or the newly released Rage with AMD Radeon HD 6000 and HD 5000 will like the features of this preview driver:

•Improves performance in Battlefield 3 Open Beta release for both non-Anti-Aliasing and application enabled Anti-Aliasing cases on the AMD Radeon™ HD 6000 and AMD Radeon HD 5000 series for single GPU configurations.
•Improves performance in Battlefield 3 Open Beta release for both non-Anti-Aliasing and application enabled Anti-Aliasing cases on AMD CrossFire™ configurations using the AMD Radeon HD 6000 and AMD Radeon HD 5000 series of products.
•Improves performance in Rage on single GPU and AMD CrossFire configurations using the AMD Radeon HD 6000, AMD Radeon HD 5000 Series and AMD Radeon HD 4000 series of products.
•Resolves intermittent crashing seen loading levels in Rage
•Resolves flickering of NPCs in Rage
•Enables automatic Vsync for Rage
•Enables support for AMD Eyefinity 5x1 display (portrait and landscape) configurations using the AMD Radeon HD 6000 and AMD Radeon HD 5000 series of products.
•AMD Vision Engine Control Center: User Interface enhancements have been implemented for the AMD CrossfireX™, GPU AMD Overdrive™ and Information Center pages.

Pick up the new download here – and stay tuned for a new look at AMD Eyefinity 5x1.

Source: tomshardware.com

AMD Demos 28nm Graphics Solutions Again

2011-10-08T18:09:00.000-07:00

Advanced Micro Devices has demonstrated a working example of its 28nm graphics processing unit known as the Radeon HD 7000. Unfortunately, the company did not unveil any information about revenue shipments of its latest products.

At Fusion 2011, AMD demonstrated its next generation graphics processor, based on the 28nm process technology. The demonstration was delivered by Matt Skynner, the head of AMD’s graphics unit, as part of his keynote titled, “Enabling the Best Visual Experience.” Skynner demonstrated a notebook-based version of AMD’s 28 nm next-generation graphics processor delivering “a smooth, high-resolution game experience” while playing Bioware’s popular role-playing title, Dragon Age 2, the game that was demonstrated earlier this week at the smartphone Apple iPhone 4S presentation.

AMD Shows Off Functioning Radeon HD 7000 for Second Time

Advanced Micro Devices has demonstrated a working example of its 28nm graphics processing unit known as the Radeon HD 7000. Unfortunately, the company did not unveil any information about revenue shipments of its latest products.

At Fusion 2011, AMD demonstrated its next generation graphics processor, based on the 28nm process technology. The demonstration was delivered by Matt Skynner, the head of AMD’s graphics unit, as part of his keynote titled, “Enabling the Best Visual Experience.” Skynner demonstrated a notebook-based version of AMD’s 28 nm next-generation graphics processor delivering “a smooth, high-resolution game experience” while playing Bioware’s popular role-playing title, Dragon Age 2, the game that was demonstrated earlier this week at the smartphone Apple iPhone 4S presentation.

“AMD strives to be at the forefront of every key inflection point in graphics technology,as demonstrated by our leadership in everything from process node transitions, to adoption of the latest graphics memory. Our pace-setting transition to the 28nm process node, coupled with new innovations in our underlying graphics architecture, is already generating excitement among the ODM community here in Taipei this week,” said Mr. Skinner.

The main reason why AMD essentially pre-announces its 28nm GPUs is because it wants to secure contracts for notebooks coming in sometimes in Q1, or most probably Q2, of 2012. We do not know AMD’s performance in securing BTS [back to school] contracts, hence, the actuial success remains to be determined by analysts like Jon Peddie.

The company remained tight-lipped about specifications and architectural peculiarities of the demonstrated Southern Islands graphics processor. Everything that is known at present is that the company's 28nm GPUs are already fully functional and the company is ready to formally introduce them.

The 28nm generation of AMD's graphics processors will be rather broad. In fact, it is rumoured that even within Southern Islands family there will be chips with VLIW4 architecture as well as more progressive so-called GCN (graphics core next) architecture.

Source: xbitlabs.com

AMD rolls out its Rage performance driver

2011-10-08T17:43:00.000-07:00

Following the release of Id Software's first person shooter RAGE, AMD has rolled out its newest performance driver made for those that want a little extra frames on their Radeon HD 6000 and HD 5000 series graphics cards.

The driver is simply called the Catalyst Rage performance driver and should bring significant performance gains for single GPU configurations on the Radeon HD 6000 and HD 5000 series cards. The new driver includes the latest OpenGL driver component for Rage performance optimizations.

For Radeon HD 6000 and 5000 series

Following the release of Id Software's first person shooter RAGE, AMD has rolled out its newest performance driver made for those that want a little extra frames on their Radeon HD 6000 and HD 5000 series graphics cards.

The driver is simply called the Catalyst Rage performance driver and should bring significant performance gains for single GPU configurations on the Radeon HD 6000 and HD 5000 series cards. The new driver includes the latest OpenGL driver component for Rage performance optimizations.

The new driver can be found here.

Source: fudzilla.com

Nvidia Begins to Include GeForce 6-Series Support into Drivers

2011-10-08T17:29:00.001-07:00

Nvidia Corp.’s latest beta drivers include support of what can possibly the company’s next-generation graphics processing units (GPUs) made using 28nm process technology. The actual chips are projected to officially emerge next year, but Nvidia may easily be tailoring its software for the new hardware.

The latest unreleased Nvidia ForceWare 285-series drivers contain mentioning of GeForce 610M and GeForce GT 630M graphics adapters for notebooks, according to a news-story by VR-Zone web-site. It is believed that the new GPUs belong to GK100-series chips that are based on the code-named Kepler micro-architecture.

Nvidia said earlier this year that it plans to test-drive its next-generation Kepler graphics processing unit (GPU) this year and introduce the new chips commercially in 2012.

Nvidia’s Latest Drivers Potentially Include 28nm GPUs Support

Nvidia Corp.’s latest beta drivers include support of what can possibly the company’s next-generation graphics processing units (GPUs) made using 28nm process technology. The actual chips are projected to officially emerge next year, but Nvidia may easily be tailoring its software for the new hardware.

The latest unreleased Nvidia ForceWare 285-series drivers contain mentioning of GeForce 610M and GeForce GT 630M graphics adapters for notebooks, according to a news-story by VR-Zone web-site. It is believed that the new GPUs belong to GK100-series chips that are based on the code-named Kepler micro-architecture.

Nvidia said earlier this year that it plans to test-drive its next-generation Kepler graphics processing unit (GPU) this year and introduce the new chips commercially in 2012.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on. Entry-level chips may not get all the features that Kepler architecture will have to often.

"[Our experience with 28nm] is looking really good, it is looking much better than our experience with 40nm. It is just a comprehensive, across-the-board engagement between TSMC and ourselves making sure that we are ready for production ramp when the time comes. So I feel really good about 28nm," said Jen-Hsun Huang, chief executive officer of Nvidia.

Source: xbitlabs.com

AMD releases new version of Catalyst 11.9 package

2011-10-08T17:07:00.000-07:00

AMD has made the final version of the Catalyst 11.9 drivers available for Windows users. As in the previous version of the package, most desktop and laptop DirectX 10 and 11 GPUs are supported, but Windows 7 users need Service Pack 1 to be installed.

The new drivers enable 3D support via DisplayPort and AMD tells us that the Samsung 750 and 950 displays are specifically supported. This time there are no particular performance enhancements listed for any games, though hanging and crashing issues in several games have been resolved.

Available for 32-bit and 64-bit Windows 7, Vista, and XP

AMD has made the final version of the Catalyst 11.9 drivers available for Windows users. As in the previous version of the package, most desktop and laptop DirectX 10 and 11 GPUs are supported, but Windows 7 users need Service Pack 1 to be installed.

The new drivers enable 3D support via DisplayPort and AMD tells us that the Samsung 750 and 950 displays are specifically supported. This time there are no particular performance enhancements listed for any games, though hanging and crashing issues in several games have been resolved.

The new drivers can be downloaded from AMD's web site, through the Catalyst software's automatic updater, or via Steam. The final version of the 11.10 drivers will replace the 11.9 drivers when they're made available next month.

Source: fudzilla.com

Nvidia 28nm mobile production in December

2011-10-08T16:58:00.000-07:00

Most of the new Nvidia 28nm mobile parts are Kepler based but some lower end chips will have its Fermi roots and more importantly - they won’t ship until 2012. The exact date is still uncertain, but we expect to see new mobile parts with the introduction of Ivy Bridge 22nm architecture.

The aim is for the Maho Bay platform to get its 28nm Nvidia chips and continue the good transition of Optimus graphics. The possible launch timeframe is going to be set by Intel, and we hear it will happen either in March or April of 2012.

The production of mobile chips starts in December at the latest, which means that these chips have been taped out for a while now. However, 28nm is currently not doing as great as some companies would make you think, as the yields are not good.

To launch later

Most of the new Nvidia 28nm mobile parts are Kepler based but some lower end chips will have its Fermi roots and more importantly - they won’t ship until 2012. The exact date is still uncertain, but we expect to see new mobile parts with the introduction of Ivy Bridge 22nm architecture.

The aim is for the Maho Bay platform to get its 28nm Nvidia chips and continue the good transition of Optimus graphics. The possible launch timeframe is going to be set by Intel, and we hear it will happen either in March or April of 2012.

The production of mobile chips starts in December at the latest, which means that these chips have been taped out for a while now. However, 28nm is currently not doing as great as some companies would make you think, as the yields are not good.

Kepler desktop version should also come in 2012, probably around the same time.

Source: fudzilla.com

AMD Radeon HD 7000 slips to next year

2011-10-08T16:56:00.000-07:00

We all know that 28nm from AMD has been demonstrated and that AMD is doing quite well with the chip, but it looks like it’s the fab that is holding things down. Reportedly, both TSMC and Globalfoundries have serious issues with mass production of 28nm chips and the yields are simply not there.

Our sources are saying that to their knowledge, AMD's Southern islands 28nm are going to ship next year, most likely in early Q1 2012 and this goes for both desktop and mobile chips. Professional graphics based on 28nm has a chance to ship in this year, but this is something that will be decided at a later date.

Nvidia has already said that they will ship in 2012, meaning that the new architecture codenamed Kepler should start shipping next year. The company also went on record and said that Kepler 28nm production starts this year, but it refered to actual chip production, with no comment on the actual market availability or retail / OEM products.

Professional graphics may launch in 2011

We all know that 28nm from AMD has been demonstrated and that AMD is doing quite well with the chip, but it looks like it’s the fab that is holding things down. Reportedly, both TSMC and Globalfoundries have serious issues with mass production of 28nm chips and the yields are simply not there.

Our sources are saying that to their knowledge, AMD's Southern islands 28nm are going to ship next year, most likely in early Q1 2012 and this goes for both desktop and mobile chips. Professional graphics based on 28nm has a chance to ship in this year, but this is something that will be decided at a later date.

Nvidia has already said that they will ship in 2012, meaning that the new architecture codenamed Kepler should start shipping next year. The company also went on record and said that Kepler 28nm production starts this year, but it refered to actual chip production, with no comment on the actual market availability or retail / OEM products.

AMD might actually launch at some point in Q4 2011 but with current yields, we would expect that the product would be sold out and literary nonexistent all the way to Q1 2012 when they expect to ramp up production. AMD did this last year with Northern Islands, Radeon HD 6000 series, and it was almost impossible to get the card throughout Q4 2010. The big issue is, once you launch the new generation, sales of the previous one take a serious hit.

Source: fudzilla.com

AMD Radeon Gets Battlefield 3 Ready, Too

2011-09-27T08:01:00.000-07:00

Just hours after NVIDIA released its Battlefield 3 optimised GeForce driver suite, AMD is ready with a similar piece of software of its own, the AMD Catalyst 11.10 Preview. Preview drivers are essentially BETA hotfix drivers. The new Catalyst software suite makes your ATI/AMD Radeon graphics card (that meets minimum/recommended requirements) optimized for Battlefield3. Its changelog follows:

•Improves performance in Battlefield 3 Open Beta release for both non-Anti-Aliasing, and application enabled Anti-Aliasing cases on the AMD Radeon HD 6000 and AMD Radeon HD 5000 Series for single GPU configurations
•Enables support for Eyefinity 5x1 display (portrait and landscape) configurations using the AMD Radeon HD 6000 and AMD Radeon HD 5000 Series
•Vision Engine Control Center: User Interface enhancements have been implemented for the CrossfireX, GPU Overdrive, and Information Center pages.

Nvidia updates GPU drivers ahead of Battlefield 3 beta

2011-09-27T07:49:00.000-07:00

As you may know, the hotly anticipated Battlefield 3 beta opens this week, and the folks at Nvidia are all over it. They've just released a new set of beta drivers, version 285.38, that should make a nice matched set with the BF3 beta. As usual, tailored drivers look to be the best way to go. Nvidia claims these betas "increase performance in the Battlefield 3 beta by up to 38% compared to the 285.27 drivers." They also include an SLI profile for BF3. You can download the drivers right here.

AMD is apparently preparing a BF3-oriented graphics driver update, as well, although we don't have a clear sense of the time frame. We'll keep an eye out for its release.

AMD unveils Radeon E6460, roadmap reveals 2012 desktop plans

2011-09-27T07:45:00.000-07:00

AMD has unveiled a new embedded discrete graphics processor with double the 3D graphics-crunching power as its E2400 GPU. The entry-level Radeon E6460 GPU "sets a new bar" for features and performance among its kind and AMD supposedly plans to push the chip for many generations to come, with five years of planned supply availability.

Although the E6460 can't compete with the power of the company's flagship E6760 (it has a third of the shaders and half the memory interface), it's plenty sufficient for everyday multimedia tasks. AMD says the part is ideal for casino gaming machines, digital signage, kiosks, point-of-sale systems and other entertainment and infotainment devices.

It features the third-gen unified video decoder (UVD3) with support for H.264, VC-1, MPEG-2/4, DivX/Xvid, Dual HD, Blu-ray and stereo 3D, as well as DirectX11 graphics. It can also drive up to four displays via Eyefinity -- that's two less than the E6760 -- and it supports AMD's Accelerated Parallel Processing (APP) for entry-level GPGPU applications.

"The AMD Radeon E6460 GPU provides the high value, small footprint, and long supply demanded by embedded customers," said Richard Jaenicke. "With a footprint-compatible subset of the high-performance AMD E6760 GPU, the new AMD Radeon E6460 GPU enables broadly scalable graphics and multimedia performance in a single board design."

In less official news, a "leaked" roadmap published by DonanimHaber suggests that AMD will launch fresh desktop components by the third quarter of 2012 (this is separate from the mobile-oriented leak earlier today). As early as the second quarter of next year, the company will replace its existing A-Series Llano APUs with a chip codenamed "Trinity."

Trinity parts will contain the next-generation x86-64 "Piledriver" architecture along with the upcoming Radeon HD 7000 series graphics. Piledriver-based chips will also flood the company's FX desktop series, displacing the long-awaited "Bulldozer" architecture, which is expected to emerge inside FX processors codenamed "Zambezi" sometime this year.

Source: techspot.com

Sapphire rolls out another HD 6870

2011-09-18T05:49:00.000-07:00

Sapphire has updated it graphics card lineup with yet another card, the HD 6870 Dirt3 Special Edition. As the name suggests, the new card comes with a voucher code for Dirt3 download, features a dual fan cooler and has a factory overclock.

As far as the spec go, the new Radeon HD 6870 Drit3 Edition card features 1120 stream processors and 56 texture processing units. It comes with a slight factory overclock to 920MHz for the GPU and 4200MHz for its 1GB of GDDR5 memory on a 256-bit memory interface.

With Dirt 3 and dual fan cooler

Sapphire has updated it graphics card lineup with yet another card, the HD 6870 Dirt3 Special Edition. As the name suggests, the new card comes with a voucher code for Dirt3 download, features a dual fan cooler and has a factory overclock.

As far as the spec go, the new Radeon HD 6870 Drit3 Edition card features 1120 stream processors and 56 texture processing units. It comes with a slight factory overclock to 920MHz for the GPU and 4200MHz for its 1GB of GDDR5 memory on a 256-bit memory interface.

The card features a dual-slot dual fan cooler, two DVIs, two HDMIs and two mini DisplayPort outputs. Unfortunately, there was no announcement regarding the price or the actual relase date but we are sure that this one will show up in retail/e-tail pretty soon. The price should be somewhere around €160.

Source: fudzilla.com

AMD launches new professional graphics cards

2011-09-15T12:32:00.000-07:00

AMD, backed by broadcast technology outfits, has just launched its FirePro SDI-Link and AMD FirePro V7900 SDI Professional Graphics Cards.

The cards are aimed to address real-time and GPU-accelerated broadcast pipelines, which require serial digital interface (SDI) input and output and are backed by AJA Video Systems, Blackmagic Design, Bluefish444, Deltacast, DVS, and Matrox. AMD FirePro SDI-Link allows fully featured SDI- and GPU-based cards with ultra-low latency between select AMD professional graphics cards and third party SDI input/output products.

It means that integrators in the broadcast and real-time video production market can use GPU acceleration. Systems built around AMD FirePro SDI-Lin can solve the need for flexible SDI input and output capabilities, and the requirement for real-time GPU-SDI communication in an integrated way.

Teams up with broadcast tech outfits

AMD, backed by broadcast technology outfits, has just launched its FirePro SDI-Link and AMD FirePro V7900 SDI Professional Graphics Cards.

The cards are aimed to address real-time and GPU-accelerated broadcast pipelines, which require serial digital interface (SDI) input and output and are backed by AJA Video Systems, Blackmagic Design, Bluefish444, Deltacast, DVS, and Matrox. AMD FirePro SDI-Link allows fully featured SDI- and GPU-based cards with ultra-low latency between select AMD professional graphics cards and third party SDI input/output products.

It means that integrators in the broadcast and real-time video production market can use GPU acceleration. Systems built around AMD FirePro SDI-Lin can solve the need for flexible SDI input and output capabilities, and the requirement for real-time GPU-SDI communication in an integrated way.

The new DirectGMA technology that allows SDI I/O products and select professional graphics cards such as the new AMD FirePro V7900 SDI to communicate directly over PCI Express (PCIe), customers can build solutions for virtual sets, professional video production/post-production, on-air motion graphics, on-set 3D VFX, and other scenarios. It manages to do this using APIs found in the AMD FirePro SDI-Link software developer kit.

The AMD FirePro V7900 SDI professional graphics card is slated for availability in October 2011 for $2,499.

Source: fudzilla.com

AMD Demonstrates Radeon HD 7000 "Southern Islands" at Event [PHOTOS ADDED]

2011-09-15T12:17:00.000-07:00

At its own media event in San Francisco, California, Advanced Micro Devices has showcased a system featuring its next-generation AMD Radeon HD 7000-series graphics processing unit (GPU) code-named Southern Islands. The demonstration is a proof-of-concept that AMD is on-track to deliver its first graphics chips made using 28nm process technology.

"Today, we provided an early look at our upcoming 28nm next-gen notebook discrete GPU, driving Codemasters’ cutting-edge driving simulator, Dirt 3, demonstrating that we have working 28nm technology in house and already delivering a great gaming experience," a statement by AMD reads.

AMD Shows Off Working Sample of Next-Generation 28nm Graphics Processor

At its own media event in San Francisco, California, Advanced Micro Devices has showcased a system featuring its next-generation AMD Radeon HD 7000-series graphics processing unit (GPU) code-named Southern Islands. The demonstration is a proof-of-concept that AMD is on-track to deliver its first graphics chips made using 28nm process technology.

"Today, we provided an early look at our upcoming 28nm next-gen notebook discrete GPU, driving Codemasters’ cutting-edge driving simulator, Dirt 3, demonstrating that we have working 28nm technology in house and already delivering a great gaming experience," a statement by AMD reads.

David Cummings, director of technology management at GPU division showcasing AMD's in-house working 28nm next-gen notebook discrete GPU.

The company remained tight-lipped about specifications and architectural peculiarities of the demonstrated Southern Islands graphics processor. Everything that is known at present is that the company's 28nm GPUs are already fully functional and the company is ready to formally introduce them.

Unfortunately, given the rumours that Taiwan Semiconductor Manufacturing Company has issues with 28nm production ramp and recently even hiked the quotes on 28nm wafers, it is unclear whether AMD will be able to release its new GPUs in volume, or those will be limited edition products.

The 28nm generation of AMD's graphics processors will be rather broad. In fact, it is rumoured that even within Southern Islands family there will be chips with VLIW4 architecture as well as more progressive so-called GCN (graphics core next) architecture.

Source: xbitlabs.com

GeForce 285.27 beta boosts speed in Black Ops, Crysis 2, more

2011-09-15T12:09:00.000-07:00

Nvidia has published its latest band of beta graphics drivers, giving GeForce owners a performance boost over last month's WHQL-certified release. Folks running the 280.26 drivers with GeForce 400 or 500-series cards stand to gain a 5 to 13% boost in a dozen or more PC games, including Call of Duty: Black Ops, Civilization V, Crysis 2, Dragon Age 2, F1 2010, Just Cause 2, Mafia 2, Metro 2033, S.T.A.L.K.E.R.: Call of Pripyat, and StarCraft II.

GeForce beta 285.27 also contains numerous 3D Vision-related changes, such as new game profiles for Aion, Diablo 3, Dead Rising 2: Off The Record, From Dust, Dirt 3 and others. It introduces support for viewing 3DVisionLive.com and YouTube 3D with SLI-enabled PCs, and it makes windowed mode the default viewing option for the 3D Vision Photo Viewer. You'll find the full release notes here (PDF) or just view one of our download pages:

Desktop: Windows XP 32-bit | Windows XP 64-bit | Windows Vista/7 32-bit | Windows Vista/7 64-bit
Mobile: Windows Vista/7 32-bit | Windows Vista/7 64-bit

There are four bug fixes for Windows XP 32- and 64-bit: two pertaining to single-GPU configurations and two for multi-GPU setups. The first set are minor software glitches with Nvidia Update, while the latter pair seem more noteworthy. One caused a BSOD when playing Adobe Flash content at HD resolutions and then switching to fullscreen mode, while the other caused lousy performance in The Witcher 2: Assassins of Kings when using SLI.

Source: techspot.com

Nvidia 28nm mobile to beat Ivy Bridge

2011-09-12T11:39:00.001-07:00

Ivy Bridge will once again set Intel graphics standards at much higher level, but this does not mean that Intel is going to win the performance crown without a fight. Intel still competes with Nvidia and AMD with entry level GPUs that still have ability to beat Intel's on CPU graphics.

Nvidia's mobile and notebook boss, Rene Haas, has confirmed that despite the fact that Ivy Bridge gets better graphics, Nvidia is preparing something even better. He said that in current generation, Nvidia ended up some 20 to 30 percent faster than Sandy Bridge graphics, and that this trend will continue in the future.

Mobile boss confirms

Ivy Bridge will once again set Intel graphics standards at much higher level, but this does not mean that Intel is going to win the performance crown without a fight. Intel still competes with Nvidia and AMD with entry level GPUs that still have ability to beat Intel's on CPU graphics.

Nvidia's mobile and notebook boss, Rene Haas, has confirmed that despite the fact that Ivy Bridge gets better graphics, Nvidia is preparing something even better. He said that in current generation, Nvidia ended up some 20 to 30 percent faster than Sandy Bridge graphics, and that this trend will continue in the future.

You can expect that next generation graphics core for notebooks from Nvidia ends up at least 20 to 30 percent faster from Ivy Bridge, and in some cases much more / much less. Nvidia also reminded us that over time Intel gets quite good in optimizing its graphics for benchmarks, which is quite an interesting remark from a company that just got a saucy settlement and money from mighty Intel.

Ivy Bridge is expected to show up in March and April depending on the version while we expect to see Nvidia'a new notebook chips before then.

Source: fudzilla.com

Nvidia believes 3D gaming makes a difference

2011-09-12T11:18:00.000-07:00

If you happen to live in a developed country and can afford a new 46+ inch TV, it’s highly likely that you will be going after LED TV with 3D capabilities, as prices of new TVs with LED panels went down dramatically over the past year or so.

Many might be reluctant to buy a new TV due to a possible second economic downturn, but all the proud owners of 3D TVs now have the chance to plug a notebook or Geforce desktop in their 3D TV and use Nvidia's 3D vision.

This is now possible via HDMI 1.4 connector and cable and the beauty is that you can use the glasses that you get with your TV.

3D TV gaming now possible

If you happen to live in a developed country and can afford a new 46+ inch TV, it’s highly likely that you will be going after LED TV with 3D capabilities, as prices of new TVs with LED panels went down dramatically over the past year or so.

Many might be reluctant to buy a new TV due to a possible second economic downturn, but all the proud owners of 3D TVs now have the chance to plug a notebook or Geforce desktop in their 3D TV and use Nvidia's 3D vision.

This is now possible via HDMI 1.4 connector and cable and the beauty is that you can use the glasses that you get with your TV. This is what Ujesh Desai Nvidia's Vice President of event marketing was very enthusiastic about. We were told that ex-ATI chipset expert Phil Eisler who runs 3D Vision for Nvidia is doing a good job.

Of course, $99 USD wired glasses for 3D monitors definitely help the cause and 3D monitors are now widespread and don’t cost that much than 60Hz monitors with the same resolution.

There should be some 500 games that support 3D vision today, and we can tell you that it looks quite good until it makes you dizzy. To be fair it doesn’t happen to everybody and it doesn’t happen every time, but we know quite a few people who got dizzy after Avatar 3D and games produce even worse effect to these individuals.

This is one of the key factors that gives PC gaming an edge and the fact that DirectX 10 and 11 games do look better than the console ones. This Xmas might be all about PC games, as 2012 is the game of new Xbox and Playstation that will get equipped with latest greatest 28nm DirectX 11 graphics, reportedly both from AMD.

Source: fudzilla.com

nVidia goes on recruitment drive to counter Radeon HD7000

2011-09-08T21:23:00.000-07:00

Right now, nVidia is holding its own with the latest iterations of Fermi against the Radeon 6000 series of cards. While both sides can boast some in-gaming benchmark wins at various price points, it’s about as level a playing field as it has been for quite some time.

nVidia has been spreading its mobile wings successfully with Tegra, while AMD is causing a mild headache for Intel with its Fusion-class processors.

But the status quo is about to be upset, big time, with a huge number of launches expected over the weeks leading up to magic Santa’s drop down the world’s chimneys.

While AMD prepares its Black Edition Fusion chips to attack Intel at the i3 low-end and Bulldozer chips to go up against the 2500/2600k parts, Intel themselves are preparing to raise the (already totally unchallenged) bar in the high-end performance space, by blowing triple-channel X58 away with quad-channel X79.

While nVidia is happy with the performance of the latest Fermi parts, and knows that it has Kepler coming up shortly – it has one area of inadequacy. It’s numbers are too low. Not only is it pushing the ’5′ against the ’6′ when GeForce takes on Radeon, it is also in the ‘hundreds’, while Radeon is in the ‘thousands’.

As AMD prepares to move the bar from ’6000′ to ’7000′, nVidia is feeling the pressure to respond. It has done so by moving to recruit a key (wo)man into its PR organisation. Someone with blackbelt spinning capability, ready to block suggestions that Fermi is feeble and challenge the idea that 7000 offers anything like 1,000 improvements over the 6000. Here’s the full job spec, in case you or a friend fancy applying.

Source: kitguru.net

28nm Kepler production this year

2011-09-08T21:17:00.000-07:00

We met Ujesh Desai, VP of event marketing at Nvidia, the chap that used to run the Geforce group at Nvidia. We had a pleasant chat and touched many subjects, including PC gaming, 3D vision, notebooks and the future of desktop market.

Ujesh said that 2011 was an exciting year in desktop and that as we have stated many times, Nvidia’s notebook did great. Nvidia PR representative Ken Brown confirmed the recent news that 28nm based Kepler starts production this year, but as Nvidia said before, the roadmap that Jensen showed back in January, is Nvidia’s production roadmap, not their partners.

Nvidia VP official, products in 2012

We met Ujesh Desai, VP of event marketing at Nvidia, the chap that used to run the Geforce group at Nvidia. We had a pleasant chat and touched many subjects, including PC gaming, 3D vision, notebooks and the future of desktop market.

Ujesh said that 2011 was an exciting year in desktop and that as we have stated many times, Nvidia’s notebook did great. Nvidia PR representative Ken Brown confirmed the recent news that 28nm based Kepler starts production this year, but as Nvidia said before, the roadmap that Jensen showed back in January, is Nvidia’s production roadmap, not their partners.

Despite the fact that Kepler production in 28nm might start very soon, in Q4 2011, actual products based on new chips will only start showing in 2012, probably early 2012. Both chaps confirmed production in late 2011 and products in 2012.

Of course, Nvidia didn't want to comment about any specific details and which chip comes first, desktop or notebook, but they are working on them all, as well as on in house new Quadro cards based on the new chip.

Source: fudzilla.com

External graphics still happening

2011-09-05T10:23:00.000-07:00

Rene Haas, general manager of Nvidia’s notebook division, was kind enough to offer us a cup of Starbuck’s coffee at Nvidia’s headquarters, and talk about some of the company’s mobile plans.

He was energized with the success of his group, as Nvidia got more than 200 design wins on Intel’s Sandy Bridge platform. Last time we met Rene, almost a year to the day, he mentioned external graphics and hinted that Nvidia might go down that road, a road less travelled.

He didn’t mention much more than that, he merely said that Nvidia was still working on it, so stay tuned for an update. Thunderbolt connectivity was mentioned in the conversation, giving us the impression that this might the connection of a choice, but this was not confirmed by Nvidia’s mobile boss.

Nvidia mobile boss claims

Rene Haas, general manager of Nvidia’s notebook division, was kind enough to offer us a cup of Starbuck’s coffee at Nvidia’s headquarters, and talk about some of the company’s mobile plans.

He was energized with the success of his group, as Nvidia got more than 200 design wins on Intel’s Sandy Bridge platform. Last time we met Rene, almost a year to the day, he mentioned external graphics and hinted that Nvidia might go down that road, a road less travelled.

He didn’t mention much more than that, he merely said that Nvidia was still working on it, so stay tuned for an update. Thunderbolt connectivity was mentioned in the conversation, giving us the impression that this might the connection of a choice, but this was not confirmed by Nvidia’s mobile boss.

We got a feeling that we might see this external graphics materializing in 2012 and with 28nm Nvidia mobile chips, but this is not something we got confirmed or denied at this time.

The bottom line is, external graphics with Nvidia chips inside are happening, something that might be the future for a number of Nvidia add-in-board partners who might end up selling external graphics, as the desktop graphics market dwindles. EVGA, Gainward, Asus, MSI, XFX, TGT and other usual suspects might resort to selling notebook external cards.

This sounds like a decent model for the future and one of the ways AIBs could maintain their revenues at reasonable levels. Also, bear in mind that AMD pioneered external graphics three years ago with XGP and it is still in the game. Sony’s Vaio Z, introduced last May, offers optional AMD external graphics via Thunderbolt.

Source: fudzilla.com

Nvidia Warns About Counterfeit GeForce Graphics Chips

2011-08-31T12:00:00.000-07:00

Nvidia Corp. has warned its customers to buy graphics processors, other chips and graphics cards only from “authorized” partners. Apparently, products sold on the grey market may be re-branded and sold under labels of more expensive products.

"Reseller who do not purchase from authorized partners run the risk of putting substandard products in their customer's hands. That can only lead to higher returns rates and damaged relationships with consumers," said Fiona Faulkner, senior director for sales for GeForce EMEAI, reports MicroScope web-site.

Nvidia Advices Partners to Buy Graphics Chips and Cards from Authorized Dealers

Nvidia Corp. has warned its customers to buy graphics processors, other chips and graphics cards only from “authorized” partners. Apparently, products sold on the grey market may be re-branded and sold under labels of more expensive products.

"Reseller who do not purchase from authorized partners run the risk of putting substandard products in their customer's hands. That can only lead to higher returns rates and damaged relationships with consumers," said Fiona Faulkner, senior director for sales for GeForce EMEAI, reports MicroScope web-site.

Remarking of graphics processors and even central processing units is not something unusual and occurred for many times in the past. For example, it is not hard to remark a chip into a more expensive model that runs on a higher clock-speed and then sell it at a higher price. The same can be done with a fully-made graphics card. In many cases, such “rebranding” may not be noticed quickly.

From time to time manufacturers of graphics cards and other hardware buy more graphics processing units that they need and then resell them. Chips and graphics cards sold by companies who do not directly buy from Nvidia or its reseller partners can be remarked in order to maximize their pricing.

While Nvidia does not elaborate, it is likely that a batch of counterfeit Nvidia GeForce graphics boards has emerged on the market.

Source: xbitlabs.com

28nm graphics dubbed Radeon HD 7000, official

2011-08-31T11:58:00.000-07:00

Globalfoundries have just showed a slide confirming next generation Radeon HD 7000 graphics will be used in upcoming Trinity parts, codenamed Comal and Virgo.

We believe that this settles the score, HD 7000 is the official designation for Southern Islands, AMD’s 28nm next generation graphics. The current schedule still indicates a launch in later part of September, just before the end of the quarter, to make the investors a bit happier. This should take place a year after the Northern Island Radeon HD 6000 launch.

Slide shown to public

Globalfoundries have just showed a slide confirming next generation Radeon HD 7000 graphics will be used in upcoming Trinity parts, codenamed Comal and Virgo.

We believe that this settles the score, HD 7000 is the official designation for Southern Islands, AMD’s 28nm next generation graphics. The current schedule still indicates a launch in later part of September, just before the end of the quarter, to make the investors a bit happier. This should take place a year after the Northern Island Radeon HD 6000 launch.

So the Radeon HD 7000 series moniker looks legitimate, and we do have a picture to back it up. It is now certain that AMD comes first with its 28nm chips, before Nvidia's Kepler that has reportedly been pushed back for early 2012. It was interesting that despite close partnership with Globalfoundries and the fact that they have 28nm, it still decided to manufacture its chip at TSMC, as TSMC has much more experience with highly complex graphics chips.

AMD simply didn’t have money or the practical need to support two different 28nm chip designs, one for TSMC and one for Globalfoundries.

Source: fudzilla.com

Nvidia hopes to get Apple back

2011-08-22T11:32:00.000-07:00

Nvidia was never too happy over losing Apple as a GPU customer but this time around AMD won Apple's heart with a nice box of chocolates and a heartfelt song.

Apple is definitely not a small player and more importantly, most people know Apple and its products better than any other computer and gadget companies. This is what it is important to get it back, despite the information that you cannot make much money out of GPU deals with Apple. Apple makes the rules and usually they are not so good for your revenues.

Need the right GPU on time

Nvidia was never too happy over losing Apple as a GPU customer but this time around AMD won Apple's heart with a nice box of chocolates and a heartfelt song.

Apple is definitely not a small player and more importantly, most people know Apple and its products better than any other computer and gadget companies. This is what it is important to get it back, despite the information that you cannot make much money out of GPU deals with Apple. Apple makes the rules and usually they are not so good for your revenues.

Jenen was asked about Nvidia’s possible return to Apple computers and he said that in order to do that, you need to have a great GPU and have it on time. Apple has its own operating agenda and it doesn’t care about the GPU refresh cycle. In case you meet these requests you might be winning the next round.

So far AMD GPUs are in most of new Apple computers and the rest is reserved for Intel. With the right GPU at the right time, Nvidia believes that it has a good chance but Nvidia was specific that they don’t have anything to announce at this time. However, AMD is probably hoping to have the “right GPU at the right time” as well.

Source: fudzilla.com

Ex-Cray Exec to Co-Define the Future of Nvidia GPUs, Tesla Compute Solutions

2011-08-22T11:30:00.000-07:00

Nvidia Corp. said recently it would appoint Steve Scott, a longtime Cray executive, to the position of chief technology officer of Tesla business unit to help spearhead the company's high-performance computing (HPC) initiative. The appointment underlines the importance of Tesla products and HPC technologies in general.

As chief technology officer (CTO) for Nvidia’s Tesla business unit, Steve Scott will be responsible for the Tesla roadmap and architecture. Tesla is rapidly becoming a fundamental technology in accelerated high performance computing and is expected to be the cornerstone in the race to exascale supercomputers towards the end of the decade.

Nvidia Hires Ex-Cray Executive to Boost HPC Business

Nvidia Corp. said recently it would appoint Steve Scott, a longtime Cray executive, to the position of chief technology officer of Tesla business unit to help spearhead the company's high-performance computing (HPC) initiative. The appointment underlines the importance of Tesla products and HPC technologies in general.

As chief technology officer (CTO) for Nvidia’s Tesla business unit, Steve Scott will be responsible for the Tesla roadmap and architecture. Tesla is rapidly becoming a fundamental technology in accelerated high performance computing and is expected to be the cornerstone in the race to exascale supercomputers towards the end of the decade.

Mr. Scott, age 45, served 19 years at Cray, including the last six as senior vice president and CTO, with responsibility for defining Cray's technology and system architecture roadmap. He holds 27 U.S. patents in the areas of interconnection networks, processor micro architecture, cache coherence, synchronization mechanisms and scalable parallel architectures.

"There are few people on the planet that have Steve's deep system level understanding of high performance computing. Steve's decision to join Nvidia is a resounding endorsement that GPU accelerated computing is the future of HPC. He will play a central role in architecting the world's most powerful supercomputers," said Bill Dally, Nvidia’s chief scientist.

It is remarkable that starting both Mr. Dally and Mr. Scott will define Nvidia’s graphics processors roadmap, but neither of them are graphics specialists. William Dally has specialized in parallel computing, whereas Steve Scott has specialized on development of HPC system architectures. Given Mr. Scott’s specialization, it is clear that he will be one of the key specialists in development of Echelon supercomputing project that Nvidia is developing with DARPA.

Source: xbitlabs.com/

AMD Quietly Debuts Graphics Processors for All-in-One Desktops

2011-08-22T11:27:00.000-07:00

Advanced Micro Devices has quietly started to sell a lineup of AMD Radeon HD 6000A designed specifically for all-in-one (AIO) desktops, such as Apple iMac, which require relatively low power consumption and thermals, but which do not need ultra-cool chips that are installed into mobile computers.

At present AMD offers four products in the Radeon HD 6000A lineup: models 6670A, 6650A and 6550A based on Turks graphics processing unit (GPU) with 480 stream processors, 24 texture units and 8/32 color/z render operating units as well as 6450A powered by Caicos chip with 160 stream processors, 8 texture units, 4/16 color/z ROPs.

AMD Radeon HD 6000A Targets AIO Desktops

Advanced Micro Devices has quietly started to sell a lineup of AMD Radeon HD 6000A designed specifically for all-in-one (AIO) desktops, such as Apple iMac, which require relatively low power consumption and thermals, but which do not need ultra-cool chips that are installed into mobile computers.

At present AMD offers four products in the Radeon HD 6000A lineup: models 6670A, 6650A and 6550A based on Turks graphics processing unit (GPU) with 480 stream processors, 24 texture units and 8/32 color/z render operating units as well as 6450A powered by Caicos chip with 160 stream processors, 8 texture units, 4/16 color/z ROPs. Based on specs and clock-speeds of the new products, they should offer similar levels of performance as appropriate add-in desktop cards.

The new graphics processors require less power than normal discrete graphics cards for desktops, but still consume more than fully-fledged mobile GPUs and may not support certain power management technologies.

"Depending on the form-factor of the target product, ASICs are defined and qualified differently, thus the use of an A or M at the end of the series number to distinguish an All-in-One (A) or mobile (M) part from that of a desktop part. OEMs immediately know that a part designated as AMD Radeon HD 6000A has been validated and specified for the unique needs of All-in-One designs, including thermals, power requirements, and feature support," said Dave Erskine, a spokesman for AMD.

AMD ships Radeon HD 6000A graphics processing units as MXM modules as well as separate processors and memory chips.

The fact that AMD decided to introduce a separate lineup of graphics processors for AIO desktops proves that the popularity of such systems is growing and end-users demand higher performance graphics solutions.

Source: xbitlabs.com

Nvidia reports strong Q2 earnings

2011-08-22T11:17:00.000-07:00

Nvidia posted its Q2 figures and the numbers looks pretty promising despite the seasonal downturn in desktop GPU shipments and a weaker than expected showing in the Tegra market.

The company reported $151.6 million earnings, or 25 cents a share, while non-GAAP net income was $193.5 million, with 32 cents per share and revenue of $1.02 billion. The figures are largely in line with Wall Street’s expectations, but Nvidia shares dipped Friday despite the positive report. Console royalties and Intel licensing revenues made up a significant chunk of Nvidia's Q2 earnings, but both of these revenue sources will dwindle in the future.

Tegra business remains flat despite the hype

Nvidia posted its Q2 figures and the numbers looks pretty promising despite the seasonal downturn in desktop GPU shipments and a weaker than expected showing in the Tegra market.

The company reported $151.6 million earnings, or 25 cents a share, while non-GAAP net income was $193.5 million, with 32 cents per share and revenue of $1.02 billion. The figures are largely in line with Wall Street’s expectations, but Nvidia shares dipped Friday despite the positive report. Console royalties and Intel licensing revenues made up a significant chunk of Nvidia's Q2 earnings, but both of these revenue sources will dwindle in the future.

Analysts see less than impressive figures in the Tegra business as a reason for concern in the long run. Although Nvidia is touting Tegra as the next big thing in its product portfolio, sales are still low relative to other market segments and growth has been slow for months.

"Investors just assumed the product was going to ramp of its own accord. In fact, Nvidia is really beholden to its (manufacturing) customers," said MKM Partners analyst Daniel Berenbaum. Berenbaum pinned the blame on the management, saying that Tegra ramp plans did not appear realistic.

In spite of Tegra issues, Nvidia remains upbeat and the company hopes to post stronger results in Q3, with revenue in the $1.06 and $1.08 billion range.

Source: fudzilla.com

Nvidia: We Are Well Prepared for 28nm Process Technology

2011-08-13T20:24:00.000-07:00

Nvidia Corp. said it had learnt the lessons with 40nm process technology and would not repeat its mistakes with the 28nm fabrication process. The company spent time on learning the peculiarities of the 28nm manufacturing technology from Taiwan Semiconductor Manufacturing Company and already has working 28nm chips.

"We are far better prepared for 28nm than we were for 40nm. Because we took it so much more seriously. We were successful on so many different nodes for so long that we all collectively, as an industry, forgot how hard it is. So, one of the things that we did this time around was to set up an entire organization that is dedicated to advanced nodes.

Nvidia: We Have Working 28nm Silicon

Nvidia Corp. said it had learnt the lessons with 40nm process technology and would not repeat its mistakes with the 28nm fabrication process. The company spent time on learning the peculiarities of the 28nm manufacturing technology from Taiwan Semiconductor Manufacturing Company and already has working 28nm chips.

"We are far better prepared for 28nm than we were for 40nm. Because we took it so much more seriously. We were successful on so many different nodes for so long that we all collectively, as an industry, forgot how hard it is. So, one of the things that we did this time around was to set up an entire organization that is dedicated to advanced nodes. We have had many, many tests chips run on 28nm, we have working silicon," said Jen-Hsun Huang, chief executive officer of Nvidia.

TSMC's 40nm manufacturing process was plagued with low yields conditioned by problems with the technology itself, manufacturing issues and design errors. Nvidia started to develop its 40nm lineup without knowing about the potential issues and when it ran into a set of unexpected problems it had to delay launch of its key-products based on Fermi architecture. With 28nm process technology, the company seems to be better prepared.

"[Our experience with 28nm] is looking really good, it is looking much better than our experience with 40nm. It is just a comprehensive, across-the-board engagement between TSMC and ourselves making sure that we are ready for production ramp when the time comes. So I feel really good about 28nm," said Jen-Hsun Huang.

Nvidia said earlier this month that it plans to test-drive its next-generation Kepler graphics processing unit (GPU) this year and introduce the new chips commercially in 2012.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on.

Source: .xbitlabs.com

28nm Radeons this year

2011-08-13T19:53:00.000-07:00

Last CES we had a chance to talk to some people in know, and we’ve learned that 28nm graphics chips for AMD should be coming this year. What caught our eye was that AMD intern CEO Thomas Seifert revealed that AMD already has working samples of 28nm chips. This comes as little surprise as the retail cards are expected at some point in Q4 2011.

So, Southern Islands should be coming this year in 28nm. Back at CES our sources were quite sure that TSMC is the only graphics provider for 28nm chips and that Globalfoundries was left behind simply as it was too expensive to develop a GPU for two fabs.

Globalfoundries might be included

Last CES we had a chance to talk to some people in know, and we’ve learned that 28nm graphics chips for AMD should be coming this year. What caught our eye was that AMD intern CEO Thomas Seifert revealed that AMD already has working samples of 28nm chips. This comes as little surprise as the retail cards are expected at some point in Q4 2011.

So, Southern Islands should be coming this year in 28nm. Back at CES our sources were quite sure that TSMC is the only graphics provider for 28nm chips and that Globalfoundries was left behind simply as it was too expensive to develop a GPU for two fabs.

Simply browsing over to Globalfoundries website we have found the story that you can read here, and the story talks about 28nm graphics chips, conviniently posted at Globalfoundies website. This gives us an idea that Globalfoundries might be involved in 28nm graphics after all.

Globalfoundries might get some action on 28nm Brazos successors, such as Deccan and Krishna, Wichita and Hondo APU’s, but this was also never confirmed. Brazos was currently being produced only at TSMC.

Globalfoundries annual forum is happening on August 30th so we might get some chance to find out more about its involvement in 28nm bulk GPUs and APUs manufacturing.

Source: fudzilla.com

Nvidia releases Geforce 280.26 WHQL driver

2011-08-13T19:31:00.000-07:00

Nvidia has released its newest WHQL-certified Geforce driver, version 280.26. It appears that the new driver brings all the updates previously seen on the 280.19 beta driver and a new 3D Vision component.

As it was the case with previous drivers, the new one supports Geforce 6, 7, 8, 9, 100, 200, 300, 400 and 500 series graphics cards as well as Nvidia's ION GPUs. It features PhysX software v9.10.0514, HD Audio v1.2.23.3 driver and enables Nvidia SLI on AMD's SLI-certified 990FX, 990X and 970 chipset motherboards.

The new 3D Vision part

Nvidia has released its newest WHQL-certified Geforce driver, version 280.26. It appears that the new driver brings all the updates previously seen on the 280.19 beta driver and a new 3D Vision component.

As it was the case with previous drivers, the new one supports Geforce 6, 7, 8, 9, 100, 200, 300, 400 and 500 series graphics cards as well as Nvidia's ION GPUs. It features PhysX software v9.10.0514, HD Audio v1.2.23.3 driver and enables Nvidia SLI on AMD's SLI-certified 990FX, 990X and 970 chipset motherboards.

Apart from bunch of 3D Vision profiles for multiple game titles, additional support for 3D Vision compatible devices, support for windowed mode in 3D Vision player, support for 3DVisionLive.com and Youtube 3D the new driver also comes with a new 3D Vision Controller driver that is now included with all GPU driver installations.

Unfortunately there hasn't been any talk regarding bug fixes or performance improvement.

You can find the new driver here.

Source: fudzilla.com

Nvidia and AMD lose GPU market share

2011-08-13T19:29:00.000-07:00

Graphics giants Nvidia and AMD have lost ground to Intel last quarter, mainly thanks to strong sales of Intel CPU with integrated graphics.

According to Jon Peddie Research Intel controlled more than 60 percent of the graphics market in Q2 and both AMD and Nvidia failed to capitalize on surprisingly healthy market. Despite the fact that demand continue to grow in Q2, contrary to seasonal trends, neither AMD nor Nvidia capitalized on the surge.

The figures paint a rather bleak picture for graphics companies. AMD ended the quarter with a 21.2 percent share, down from 24.7 percent in Q1 and 24.7 percent in Q2 2010. Nvidia did even worse, with a 17.5 percent share compared to 19.9 percent in Q1 and that 21.4 percent last year.

Intel extends lead

Graphics giants Nvidia and AMD have lost ground to Intel last quarter, mainly thanks to strong sales of Intel CPU with integrated graphics.

According to Jon Peddie Research Intel controlled more than 60 percent of the graphics market in Q2 and both AMD and Nvidia failed to capitalize on surprisingly healthy market. Despite the fact that demand continue to grow in Q2, contrary to seasonal trends, neither AMD nor Nvidia capitalized on the surge.

The figures paint a rather bleak picture for graphics companies. AMD ended the quarter with a 21.2 percent share, down from 24.7 percent in Q1 and 24.7 percent in Q2 2010. Nvidia did even worse, with a 17.5 percent share compared to 19.9 percent in Q1 and that 21.4 percent last year.

However, Intel’s share grew to 60.7 percent, up from 54.7 percent in Q1 and 52.9 percent in the second quarter of 2010. Year on year, Intel’s share grew by 14.7 percent, while AMD and Nvidia shed 14.2 percent and 18.4 percent respectively.

Source: fudzilla.com

Nvidia Signs On Cray Exec As New Tesla CTO

2011-08-13T19:26:00.000-07:00

Wednesday Nvidia said that longtime Cray Inc. executive Steve Scott will be signing on as chief technology officer (CTO) for Nvidia's Tesla business unit. According to the company, Scott will be responsible for the overall Tesla roadmap and architecture.

"There are few people on the planet that have Steve's deep system level understanding of high performance computing," said Bill Dally, Nvidia's chief scientist. "Steve's decision to join Nvidia is a resounding endorsement that GPU accelerated computing is the future of HPC. He will play a central role in architecting the world's most powerful supercomputers."

Cray Inc. executive Steve Scott will be joining Nvidia's Tesla business unit as CTO.

Wednesday Nvidia said that longtime Cray Inc. executive Steve Scott will be signing on as chief technology officer (CTO) for Nvidia's Tesla business unit. According to the company, Scott will be responsible for the overall Tesla roadmap and architecture.

"There are few people on the planet that have Steve's deep system level understanding of high performance computing," said Bill Dally, Nvidia's chief scientist. "Steve's decision to join Nvidia is a resounding endorsement that GPU accelerated computing is the future of HPC. He will play a central role in architecting the world's most powerful supercomputers."

The 45-year-old ex-Cray executive spent the last six years of his nineteen-year tenure as Cray Inc.'s senior vice president and CTO, and was responsible for defining the supercomputer company's technology and system architecture roadmap. He also currently holds 27 U.S. patents in the areas of interconnection networks, processor micro architecture, cache coherence, synchronization mechanisms and scalable parallel architectures... AKA a big win for Nvidia's patent portfolio.

In addition to his achievements at Cray Inc., Scott was the recipient of the 2005 ACM Maurice Wilkes Award and the 2005 IEEE Seymour Cray Computer Engineering Award.

Wednesday Nvidia did not indicate exactly when Scott will take the Tesla CTO throne. "Tesla is rapidly becoming a fundamental technology in accelerated high performance computing and is expected to be the cornerstone in the race to exascale," the company said in a press release.

Source: tomshardware.com

Nvidia Denies Plans to Release Kepler GPU in 2011

2011-08-07T08:35:00.000-07:00

Nvidia Corp. on Thursday clarified its plans regarding the next-generation Kepler graphics processing units (GPUs) and their release timeframes. Apparently, the company is on schedule to receive the early silicon of Kepler from its manufacturing partner later in 2011, but the commercial launch of the product is scheduled to occur only in 2012.

"Although we will have early silicon this year, Kepler-based products are actually scheduled to go into production in 2012. We wanted to clarify this so people wouldn’t expect product to be available this year," said Ken Brown, a spokesman for Nvidia, in an email statement.

Nvidia: Kepler-Based Products to Ship in 2012

Nvidia Corp. on Thursday clarified its plans regarding the next-generation Kepler graphics processing units (GPUs) and their release timeframes. Apparently, the company is on schedule to receive the early silicon of Kepler from its manufacturing partner later in 2011, but the commercial launch of the product is scheduled to occur only in 2012.

"Although we will have early silicon this year, Kepler-based products are actually scheduled to go into production in 2012. We wanted to clarify this so people wouldn’t expect product to be available this year," said Ken Brown, a spokesman for Nvidia, in an email statement.

Chris Malachowsky, senior vice president of research and a co-founder of Nvidia said at a recent event that Nvidia would start shipping its next-generation graphics processing units code-named Kepler by the end of the year. The company did not say that the new chips will actually become available commercially though.

Kepler is Nvidia's next-generation graphics processor architecture that is projected to bring considerable performance improvements and will likely make the GPU more flexible in terms of programmability, which will speed up development of applications that take advantage of GPGPU (general purpose processing on GPU) technologies. Some of the technologies that Nvidia promised to introduce in Kepler and Maxwell (the architecture that will succeed Kepler) include virtual memory space (which will allow CPUs and GPUs to use the "unified" virtual memory), pre-emption, enhance the ability of GPU to autonomously process the data without the help of CPU and so on.

The new chip is projected to be made using 28nm process technology. Many believe that Taiwan Semiconductor Manufacturing Company, which makes chips for Nvidia, AMD and many others, will not be able to supply enough 28nm products this calendar year.

Nvidia's Kepler family of products, which will likely get GeForce 600-series name in the consumer market segment, will not only power Nvidia's mid-term future products, but will also help Nvidia to boost sales of its desktop discrete graphics cards. In Q2 2011 shipments of discrete graphics boards for desktops were down 15%, according to some analysts.

Source: xbitlabs.com

Sapphire Launches Radeon HD 6950 TOXIC Edition, Fastest HD 6950 Yet

2011-08-07T08:30:00.000-07:00

SAPPHIRE Technology – the world leading supplier of graphics solutions, mainboards and other PC products - has just announced the SAPPHIRE HD 6950 TOXIC Edition, the fastest production card in its class.

The SAPPHIRE HD 6950 TOXIC Edition ships with clock speeds of 880MHz for the core and 1300Mhz (5.2 Gb/sec effective) for the memory making it the fastest production HD 6950 on the market. This has been achieved by building the card on a high specification PCB with digital VRMs for excellent power regulation and 8 + 6 pin power connectors. It has a high performance cooler featuring a Vapor Chamber, blower fan and a new perforated backplate to increase airflow out of the system case.

Like the standard HD 6950, the SAPPHIRE HD 6950 TOXIC Edition boasts true DX 11 capability and the powerful configuration of 1408 stream processors and 88 texture processing units. A Dual BIOS feature allows enthusiasts to experiment with alternative BIOS profiles and settings. Performance can be further enhanced with SAPPHIRE TriXX – the company’s dedicated overclocking tool which allows memory and core clock speeds and voltages to be modified and monitored and is available free to download.

With its two DVI, one HDMI and two mini-DisplayPort outputs, the SAPPHIRE HD 6950 TOXIC Edition supports a wide range of monitor configurations, including AMD Eyefinity with up to four screens.

The SAPPHIRE HD 6950 TOXIC Edition has the latest Gen3 on-board hardware UVD (Unified Video decoder), considerably reducing CPU load and delivering smooth decoding of Blu-ray and other video content for both VC-1 and H.264 codecs, as well as Mpeg files. The UVD can decode two 1080p HD video streams simultaneously and display HD video in high quality with Windows Aero mode enabled. The card has HDMI 1.4a which also supports Stereoscopic 3D output.

The SAPPHIRE HD 6950 family is supported by AMD's DirectX 11 WHQL certified graphics driver which delivers support for all of the key DirectX 11 level features required for new gaming experiences and acceleration of the latest generation of browsers, transcoding and other high performance applications (AMD APP acceleration).

Source: techpowerup.com

Conclusion: Point of View / TGT GTX 580 3GB Ultra Charged review

2011-08-07T08:13:00.000-07:00

We tested several games at 2560x1600 but we needed more than 1.5GB of memory only in rare cases. Although 3GB of memory sounds enticing, anyone who's less than hard-core gamer would do just fine with the reference GTX 580 1.5GB version. PoV/TGT aren't targeting uninformed customers who consider RAM to be the measure of GPU power, but rather informed gamers who like to know that they will never be limited by available memory. If you’re looking to use the card for scientific purposes and tap into CUDA, then you’ll probably love the 3GB version – it will allow for twice as much data in the memory.

Several Nvidia's partners have already launched 3GB versions of the GTX 580, which means that the demand exists. While there are only few games that can put the 3GB to use, we're hoping to see more by the year's end.

Our testing revealed that 2560x1600 gaming will suffer from GPU speed limitations more often than memory limitations. The 3GB version is also useful for SLI. If you use two reference GTX 580 cards with SLI, you have 2x 1.5GB and not 1x 3.0GB. With 2x 3.0GB versions, you have both the performance and the memory to support very high resolutions. Of course the 3GB version is also very useful for games which support multiple monitors.

Investing in future is not always an easy decision to make, especially if the ‘future’ costs €540 here. PoV/TGT GTX 580 UC card with 1.5GB goes for about €90 less.

We would've liked a better cooler on the GTX 580 3GB Ultra Charged. Compared to the reference version, our today's test sample was much louder and hotter.

The card packs a nice punch, there's no doubt about it, but unless you're an enthusiast or know exactly why you need 3GB of memory, we'd recommend you to stick with the more affordable, reference GTX 580 with 1.5GB of memory.

Source: fudzilla.com

Nvidia rolls out Geforce 280.19 beta driver

2011-07-31T10:33:00.000-07:00

Following AMD's release of its newest WHQL drivers, Nvidia also rolled out its new Geforce 280.19 beta driver pack. Although most of the updates are aimed at new 3D Vision profiles and support for 3D hardware, Nvidia noted that users should use these for Battlefield 3 Alpha in order to eliminate in-game glitches.

The new driver brings support for 3D Vision projectors like Acer X1111, BenQ W710ST, and NEC NP-V300W as well as new 3D Vision profiles for bunch of games including Alice Madness Returns, LOTRO: Shadows of Angmar, Mount and Blade, Warhammer 40k: Space Marine and much more. It also updates 3D Vision profiles for Crysis 2, Deep Black and Super Street Fighter IV: Arcade Edition.

Bunch of 3D Vision profiles

Following AMD's release of its newest WHQL drivers, Nvidia also rolled out its new Geforce 280.19 beta driver pack. Although most of the updates are aimed at new 3D Vision profiles and support for 3D hardware, Nvidia noted that users should use these for Battlefield 3 Alpha in order to eliminate in-game glitches.

The new driver brings support for 3D Vision projectors like Acer X1111, BenQ W710ST, and NEC NP-V300W as well as new 3D Vision profiles for bunch of games including Alice Madness Returns, LOTRO: Shadows of Angmar, Mount and Blade, Warhammer 40k: Space Marine and much more. It also updates 3D Vision profiles for Crysis 2, Deep Black and Super Street Fighter IV: Arcade Edition.

The bug fixing in the new driver inludes some fixes in Crysis 2 and other gmaes in on Geforce 500 series cards and some other glitches when using a single GPU, as well as a bunch of similar fixes on multi-GPU configurations.

You can check out the full list of fixes and download the driver here.

Source: fudzilla.com

AMD rolls out Catalyst 11.7 WHQL driver

2011-07-31T10:24:00.000-07:00

As expected, AMD has rolled out its latest monthly Catalyst release, the Catalyst 11.7 WHQL. The new driver supports all the card from Radeon HD 2000 days and although it doesn't bring any new features or impressive performance gains it does fix a rather large amount of bugs found in previous versions of Catalyst. In addition to the new 11.7 WHQL version, AMD has also rolled out the new preview driver as well, dubbed Catalyst 11.8 Preview.

As noted, the Catalyst 11.7 WHQL brings a rather large list of fixes including mouse cursor lag, some problems with 11.6 driver and HDMI/DP displays, 720p DivX problems on Windows Media Center and Windows Media Player, some problems with various games in both single and Crossfire configurations, problems with various applications including PowerDVD, VLC Player, WinDVD 10 and much more.

Followed by 11.8 64-bit preview one

As expected, AMD has rolled out its latest monthly Catalyst release, the Catalyst 11.7 WHQL. The new driver supports all the card from Radeon HD 2000 days and although it doesn't bring any new features or impressive performance gains it does fix a rather large amount of bugs found in previous versions of Catalyst. In addition to the new 11.7 WHQL version, AMD has also rolled out the new preview driver as well, dubbed Catalyst 11.8 Preview.

As noted, the Catalyst 11.7 WHQL brings a rather large list of fixes including mouse cursor lag, some problems with 11.6 driver and HDMI/DP displays, 720p DivX problems on Windows Media Center and Windows Media Player, some problems with various games in both single and Crossfire configurations, problems with various applications including PowerDVD, VLC Player, WinDVD 10 and much more.

On the other hand, the 11.8 Preview driver, available only for Windows 7 64-bit brings a couple of rather interesting features and pefromance updates. The most important one is the HD3D technology support for 3D monitors via DisplayPort. The list of perfromance updates includes Crysis 2 by up to 10 percent on Radeon HD 5000 and 6000 series cards as well as up to eight percent in Fear 3 under DirectX 11 with the same series cards.

The last is a performance improvement of up to 30 percent when AMD's Morphological Anti-Aliasing (MLAA) is enabled via Catalyst Control Center on Radeon HD 6000 and HD 5000 series cards.

Catalyst 11.7 WHQL can be found here while the 11.8 Preview one is here.

Source: fudzilla.com

AMD admits APU will hurt discrete GPU sales

2011-07-31T10:08:00.000-07:00

During its recent conference call with financial analyst, AMD's Thomas Seifert finally admited that APUs will indeed hurt the sales of dedicated graphics cards, at least the low-end ones.

Thomas Seifert, interim chief executive officer at AMD noted that, in the long run, some parts of the business will be cannibalized and the low-end discrete GPUs will be replaced with Fusion-type products. Of course, as far as AMD is concerned, it is not a bad thing as AMD wins both ways.

At least the entry-level ones

During its recent conference call with financial analyst, AMD's Thomas Seifert finally admited that APUs will indeed hurt the sales of dedicated graphics cards, at least the low-end ones.

Thomas Seifert, interim chief executive officer at AMD noted that, in the long run, some parts of the business will be cannibalized and the low-end discrete GPUs will be replaced with Fusion-type products. Of course, as far as AMD is concerned, it is not a bad thing as AMD wins both ways.

Of course, this won't happen over night and AMD will still continue to offer dedicated GPUs, even in low-end part of the market as it simply has an entire Intel based motherboard market that can still benefit even from a low-end discrete GPU.

Source: fudzilla.com

AMD Expects to Be First GPU Designer with 28nm Graphics Chips

2011-07-31T10:06:00.000-07:00

Advanced Micro Devices remains very optimistic about its ability to ship graphics processing units (GPUs) made using 28nm process technology this year. The company believes that 28nm node will not only provide it an opportunity to offer advanced graphics chips "shortly", but also to get lower manufacturing costs as a result of work with both of its foundry partners.

AMD already has working samples of GPUs produced using 28nm fabrication process and expects to release a number of chips made at 28nm node in calendar 2011.

AMD Vows to Beat Nvidia with First 28nm GPUs on the Market

Advanced Micro Devices remains very optimistic about its ability to ship graphics processing units (GPUs) made using 28nm process technology this year. The company believes that 28nm node will not only provide it an opportunity to offer advanced graphics chips "shortly", but also to get lower manufacturing costs as a result of work with both of its foundry partners.

AMD already has working samples of GPUs produced using 28nm fabrication process and expects to release a number of chips made at 28nm node in calendar 2011. In fact, AMD vows to be the first designer of graphics chips with 28nm-based offerings on the market, which means that the firm wants to again beat Nvidia with time-to-market with leading-edge fabrication process, just like it did with 90nm, 80nm, 65nm, 55nm and 40nm nodes.

"We have working silicon in-house and remain on-track to deliver the first members of what we expect will be another industry-leading GPU family to market later this year. We expect to be at the forefront of the GPU industry's transition to 28nm," said Thomas Seifert, interim chief executive officer of AMD, during the most recent conference call with financial analysts.

Among the natural advantages that 28nm process technologies with high-k metal gate (HKMG) will bring (e.g., ability to increase performance and improve feature-set while maintaining or lowering power consumption or costs), AMD expects that it will be able to manage risks and costs if required since 28nm fabrication process will be offered by both Globalfoundries as well as Taiwan Semiconductor Manufacturing Company.

"At the 28nm node, all of our products will be based on bulk process technology, providing increased flexibility to work across our two committed and valued partner. [...] [With the introduction of 28nm process technologies], our flexibility to manage risk across the foundry partner ecosystem that we have has significantly increased," said Mr. Seifert.

Even though both Globalfoundries and TSMC both offer 28nm fabrication processes, they are hugely different and AMD will not be able dynamically re-assign manufacturing of particular chips from one foundry to another. Still, thanks to the fact that the two contract makers of semiconductors by definition have higher capacities than one, the ramp of new product family will likely occur faster than typically.

Although AMD did not reveal when the first GPUs made at 28nm node become available it implied that they may emerge on the market earlier than one might expect.

There is the 28nm product coming. That is a great opportunity to grab share again, just like we did with the industry with the DirectX 11; you saw what that did to our revenue. There is another opportunity coming up again here shortly," said Rick Bergman, senior vice president and general manager of products group at AMD.

Later during this year AMD is projected to release its first graphics processing units made using 28nm process technology code-named Southern Islands. The SI family is projected to support DirectX 11 functionality along with architectural improvements and some new features

Source: xbitlabs.com

Nvidia shares slide amidst strong competition

2011-07-31T10:03:00.000-07:00

Analysts are starting to voice concerns about Nvidia’s performance in the coming quarter, as the outfit sees slower demand for GPUs and more competition in the ARM market.

Shares dipped 1.43 percent on Monday, but they are still trading around the $15 mark, slightly higher than last week. Nvidia peaked in January and February, with shares trading at over $25. However, a year ago, in August, Nvidia was trading at under $10, so it seems the rollercoaster drive is coming to an end.

Analysts pull the brakes

Analysts are starting to voice concerns about Nvidia’s performance in the coming quarter, as the outfit sees slower demand for GPUs and more competition in the ARM market.

Shares dipped 1.43 percent on Monday, but they are still trading around the $15 mark, slightly higher than last week. Nvidia peaked in January and February, with shares trading at over $25. However, a year ago, in August, Nvidia was trading at under $10, so it seems the rollercoaster drive is coming to an end.

UBS analyst Urche Orji lowered his estimated for the company’s stock from $20 to $16.5 on Monday. He cited slow graphics sales reported by AMD last week and stronger competition in the ARM market. Orji notes that Google has already chosen Texas Instruments chips for the Ice Cream Sandwich reference design, while both TI and Qualcomm have announced chips that could outperform Nvidia’s Kal-El in most applications.

Source: fudzilla.com

AMD: Xbox 720 to Have Avatar-level Graphics

2011-07-19T10:34:00.000-07:00

We've already heard talk of DirectX 11 and an E3 2012 launch for the newest Xbox console and now AMD is adding that the console will be capable of producing graphics with Avatar-levels of detail. Referring to the smash-hit movie that put 3D on the map, Examiner cites director of ISV relationships at AMD, Neal Robison, as saying Xbox fans have a lot to be excited about.

Aside from stunning graphics, Robison is quoted as saying that the next-generation console from Microsoft will boast A.I. and physics capabilities that will allow for every pedestrian in a games like Grand Theft Auto or Saints Row 'to have a totally individual mentality'. That means no more of the mob mentality that sees everyone just running away screaming, each character will do something different.

We've been hearing snippets of information about the Xbox 720 over the last few weeks and now AMD has put its two cents in.

We've already heard talk of DirectX 11 and an E3 2012 launch for the newest Xbox console and now AMD is adding that the console will be capable of producing graphics with Avatar-levels of detail. Referring to the smash-hit movie that put 3D on the map, Examiner cites director of ISV relationships at AMD, Neal Robison, as saying Xbox fans have a lot to be excited about.

Aside from stunning graphics, Robison is quoted as saying that the next-generation console from Microsoft will boast A.I. and physics capabilities that will allow for every pedestrian in a games like Grand Theft Auto or Saints Row 'to have a totally individual mentality'. That means no more of the mob mentality that sees everyone just running away screaming, each character will do something different.

The Examiner is using August's issue of Official Xbox Magazine as a source for the information, but AMD has not confirmed that it is working on the Xbox 720. Similarly, Microsoft has not confirmed any details regarding the new system. Elsewhere in the market, Nintendo just introduced the next generation Wii, the Wii U, which is scheduled for launch next year. Sony is also rumored to be working on the PS4, which is also expected in 2012.

Source: tomshardware.com

AMD HD 6990M not much faster than GTX 580M

2011-07-19T10:27:00.000-07:00

We stumbled upon a rather intriguing post over at PCPerspective regarding recently released AMD HD 6990M performance slides as it appears that HD 6990M is not that much faster than the GTX 580M.

In case you missed it, the recently introduced AMD Radeon HD 6990M is based on Barts XT architecture and has nothing to do with its desktop dual-GPU counterpart. It is a single GPU solution that has 1120 stream processors, GPU clock set at 715MHz and up to 1.6TFlops of computing power. The GPU is paired with 2GB of GDDR5 memory clocked at 900MHz which is enough for over 115GB/s of memory bandwidth.

Apparently, AMD contacted pcper.com to let them know that benchmark results need to be changed and that new one show that the difference between the HD 6990M and the HD 6970M, and the difference between Radeon HD 6990M and GTX 580M is not that high. You can check out the article here.

AMD expected less, but priced it right

We stumbled upon a rather intriguing post over at PCPerspective regarding recently released AMD HD 6990M performance slides as it appears that HD 6990M is not that much faster than the GTX 580M.

In case you missed it, the recently introduced AMD Radeon HD 6990M is based on Barts XT architecture and has nothing to do with its desktop dual-GPU counterpart. It is a single GPU solution that has 1120 stream processors, GPU clock set at 715MHz and up to 1.6TFlops of computing power. The GPU is paired with 2GB of GDDR5 memory clocked at 900MHz which is enough for over 115GB/s of memory bandwidth.

Apparently, AMD contacted pcper.com to let them know that benchmark results need to be changed and that new one show that the difference between the HD 6990M and the HD 6970M, and the difference between Radeon HD 6990M and GTX 580M is not that high. You can check out the article here.

After asking around, our birdies have told us that it appear that AMD was using a desktop configuration based on the HD 6870 card clocked at the same clocks as the HD 6990M for those slides and have expected the results to be identical, but unfortunately they aren't. This time around AMD's inteligence service didn't do its homework well.

The bottom line is that the HD 6990M is still faster than the GTX 580M and it all comes down to the price, which is, by the way, still on AMD's side. We checked out some prices and AMD's HD 6990M is around US $300 cheaper than the GTX 580M at least that's what we have seen on the Origin site and its EON17-S notebook and Alienware site and its M17x notebook.

So this round still goes to AMD. Of course, we are still waiting to see the "correct" slides that AMD is talking about.

Source: fudzilla.com

All Three Next-Gen Consoles Sporting AMD GPUs?

2011-07-19T10:19:00.000-07:00

Although AMD already announced that Nintendo is using a custom AMD Radeon RV770 GPU in the upcoming Wii U console, there's now talk that AMD GPUs will also be used in the next PlayStation and Xbox consoles. The news is based on a rumor stemming from E3 2011 claiming that it's literally a "done deal" with Microsoft and Sony. Unsurprisingly, this is the first we've heard of this particular GPU tale... there wasn't any mention of AMD hardware in a gaming console outside the Wii U, public or in private, during or after the show.

On the CPU front, we already know that IBM is providing the 45-nm multi-core "heart" of the Nintendo Wii U. As for the third generation Xbox, IBM is also supposedly lined up with a possible Cell processor.

Will all three next-generation consoles feature IBM processors and AMD GPUs?

Although AMD already announced that Nintendo is using a custom AMD Radeon RV770 GPU in the upcoming Wii U console, there's now talk that AMD GPUs will also be used in the next PlayStation and Xbox consoles. The news is based on a rumor stemming from E3 2011 claiming that it's literally a "done deal" with Microsoft and Sony. Unsurprisingly, this is the first we've heard of this particular GPU tale... there wasn't any mention of AMD hardware in a gaming console outside the Wii U, public or in private, during or after the show.

On the CPU front, we already know that IBM is providing the 45-nm multi-core "heart" of the Nintendo Wii U. As for the third generation Xbox, IBM is also supposedly lined up with a possible Cell processor. Current rumors indicate that Microsoft will reveal the console next year at E3 2012, but there's indication that the company may hold off an additional year or two due to the success of the Kinect. This little tidbit doesn't make much sense given that the next console will likely support both Xbox 360 games and Kinect anyway.

As for Sony's effort, reports on the PlayStation 4 processor are a bit more murky than the other two, ranging from an AMD APU (Bulldozer) to a newer updated 32-nm IBM Cell processor. The new Sony console is also expected to make an appearance next year at E3 2012, so we expect to hear something more solid before then. Again, like the Xbox, it is all mere rumor until we start getting closer to the next show and tidbits of more information are "leaked" out.

Honestly, if the Big Three manufactured consoles with IBM processors and AMD GPUs across the board, development would seemingly be a lot easier for those creating multi-platform titles. Even more, if all three integrated native support for a mouse and keyboard, there wouldn't be such a wide gap between PC and console gaming. The OnLive service and its mini-console, which is about the size of a Nintendo DS, has already proven that mouse/keyboard and gamepad control setups can coexist on one device.

Just recently OnLive's VP of Engineering Joe Bentley said that the company is now talking with Microsoft and Sony about bringing its tech to the consoles. "There are OnLive guys chatting [with Sony and MS], but we'll see where it goes," he said. "But it would absolutely work, we're ready to work with everybody. Our controller is a hybrid between a PS3 controller and an Xbox controller. It's all compatible, it would just work."

Source: tomsguide.com

AMD Catalyst 11.6b hotfix driver is out

2011-07-09T20:08:00.000-07:00

AMD released its Catalyst driver 11.6b hotfix today and it comes with improved support for Llano APUs and its Radeon HD 6550D and 6530D integrated graphics.

The hotfix will resolve issues related to blue screens when connecting HDMI and DisplayPort displays to Radeon HD graphics in AMD Catalyst 11.6 Furthermore, it will also address screen flickering in idle mode and corrupt mouse pointer issues.

Includes Llano A8 and A6 support

AMD released its Catalyst driver 11.6b hotfix today and it comes with improved support for Llano APUs and its Radeon HD 6550D and 6530D integrated graphics.

The hotfix will resolve issues related to blue screens when connecting HDMI and DisplayPort displays to Radeon HD graphics in AMD Catalyst 11.6 Furthermore, it will also address screen flickering in idle mode and corrupt mouse pointer issues.

You can find out more and download the hotfix here.

Source: fudzilla.com

EVGA shows GeForce GTX 580 Classified

2011-07-09T20:06:00.000-07:00

EVGA’s Classified series usually means business and the latest sneak peek at Geforce GTX 580 Classified suggests it’ll be the case once more.

The company showed the card on its forums and although it didn’t reveal much, pictures say enough for now. Power circuitry is top of the line and the card has a dedicated 6-pin power connector for GDDR5 memory, voltage read points, voltage indicator LEDs, dual-BIOS, OC mode switch, etc.

Saucy

EVGA’s Classified series usually means business and the latest sneak peek at Geforce GTX 580 Classified suggests it’ll be the case once more.

The company showed the card on its forums and although it didn’t reveal much, pictures say enough for now. Power circuitry is top of the line and the card has a dedicated 6-pin power connector for GDDR5 memory, voltage read points, voltage indicator LEDs, dual-BIOS, OC mode switch, etc.

It is powered via two 8-pin power connectors coupled with the aforementioned 6-pin one. The company hasn’t revealed much, meaning we’ll have to wait more for actual specs, but don’t expect it to be slow, or cheap for that matter.

Source: fudzilla.com

Geforce GTX 580M is not enough for DX11 Crysis 2

2011-07-09T20:03:00.000-07:00

The DirectX 11 patch for Crysis 2 is finally out and we have been playing with that in our lab for a few days now.

In addition to the patch, we had a chance to see one of the most powerful notebooks on planet earth, Alienware MX18 a massive notebook with a lot of graphics muscle and CPU power. The “notebook” is powered by the recently launched Geforce GTX 580 mobile chip and you can imagine our surprise when we learned that this chip is not enough to run Crysis 2 at 1920x1080 at more than 20 FPS. It was stuttering in DirectX 11 and it was going really smooth in DirectX 9 mode.

Alienware MX18 at 1920x1080

The DirectX 11 patch for Crysis 2 is finally out and we have been playing with that in our lab for a few days now.

In addition to the patch, we had a chance to see one of the most powerful notebooks on planet earth, Alienware MX18 a massive notebook with a lot of graphics muscle and CPU power. The “notebook” is powered by the recently launched Geforce GTX 580 mobile chip and you can imagine our surprise when we learned that this chip is not enough to run Crysis 2 at 1920x1080 at more than 20 FPS. It was stuttering in DirectX 11 and it was going really smooth in DirectX 9 mode.

It turns out that for Crysis 2 at this resolution and DirectX 11 settings and everything at highest possible, you need at least two of these cards. To put things in perspective, Alienware US charges $1,200 for two Geforce GTX 580M with 2GB GDDR5 each and you can imagine that this is a lot of money.

The base price of the rig with Geforce GTX 460M graphics starts at $2,149 and jumps to a shocking $3,349 when you add these two cards. In the German market MX18 prices start at €1999 and two Geforce GTX 580M cards are still not available. The UK the model starts at £1699 again misses the two GTX 580 M that might show up shortly as an option.

So Crysis 2 for notebook users at HD and with DirectX 11 will be a thing for extremely rich folks and for the rest we think a desktop is a much better idea.

Source: fudzilla.com

Nvidia Provides Free Copy of Duke Nukem Forever with 3D Vision Kit

2011-07-02T19:44:00.000-07:00

Nvidia Corp. this week announced that it would bundle its 3D Vision wireless stereo-3D glasses kit with a copy of Duke Nukem Forever video game. The offering will be active for a limited amount of time only.

Starting this week, and for a limited time, Nvidia 3D Vision wireless glasses kits include a free copy of Duke Nukem Forever. Gamers in the North America can purchase the 3D Vision Duke Nukem Forever special edition for $149 in electronic stores like TigerDirect.com, Newegg.com and NCIX. Otherwise, one can look for a 3D Vision Duke Nukem Forever special edition bundle at a local retail store.

Nvidia Gives Away Duke Nukem Forever for Free

Nvidia Corp. this week announced that it would bundle its 3D Vision wireless stereo-3D glasses kit with a copy of Duke Nukem Forever video game. The offering will be active for a limited amount of time only.

Starting this week, and for a limited time, Nvidia 3D Vision wireless glasses kits include a free copy of Duke Nukem Forever. Gamers in the North America can purchase the 3D Vision Duke Nukem Forever special edition for $149 in electronic stores like TigerDirect.com, Newegg.com and NCIX. Otherwise, one can look for a 3D Vision Duke Nukem Forever special edition bundle at a local retail store.

"Duke Nukem looks better than ever. Once limited only to MS-DOS graphics, today Duke bursts out of the screen with full 3D Vision support, Nvidia Surround support and HD graphics. Imagine hunting down aliens at up to 5760*1080 resolution in a completely immersive experience," said Andrew Fear, a senior product manager of consumer stereoscopic 3D products at Nvidia Corp.

Nvidia 3D Vision kit, which includes a pair of stereo-3D glasses (with IR receiver) and an IR emitter now costs $149.

In order to take advantage of Nvidia 3D Vision, which allows to build a stereo-3D-supporting gaming or multimedia system, one should not only acquire Nvidia-proprietary S3D glasses, but also a display or an HDTV certified for usage with Nvidia 3D Vision hardware and of course an appropriate high-performance GeForce graphics card. At present there is a significant number of full-HD displays and HDTVs supporting the 3D Vision technology.

Source: xbitlabs.com

Nvidia Boasts New Laptop GPU With DX11 Crysis 2

2011-07-02T19:30:00.001-07:00

Nvidia now stakes claim to the title of the world's fastest notebook graphics processing unit with the GeForce GTX 580M. How fast are they? In SLI, they're fast enough to play Crysis 2 at max settings in 1080p with DirectX 11 effects and high-resolution textures on (check out the video below for proof of that).

GeForce GTX 580M Crysis 2 SLI

The first notebook PC to feature the GeForce GTX 580M is the Alienware M18x with the option to go SLI.

Yes, it plays very well.

Nvidia now stakes claim to the title of the world's fastest notebook graphics processing unit with the GeForce GTX 580M. How fast are they? In SLI, they're fast enough to play Crysis 2 at max settings in 1080p with DirectX 11 effects and high-resolution textures on (check out the video below for proof of that).

GeForce GTX 580M Crysis 2 SLI

The first notebook PC to feature the GeForce GTX 580M is the Alienware M18x with the option to go SLI. The Alienware M17x will offer a bit more flexibility with the GeForce GTX 580M along with Nvidia Optimus technology and will deliver 5 hours of battery life when using the integrated graphics instead of the discrete GPU.

Nvidia also introduced the GeForce GTX 570M GPU, which offers 20 percent faster performance than the previous generation product it replaces.

Source: tomshardware.com

Powercolor officially rolls out dual HD 6870 X2

2011-06-25T23:46:00.000-07:00

Previously seen and introduced at Computex in Taipei, the Powercolor Radeon HD 6870 X2 graphics card has now been officially announced. Packing two 40nm Barts XT GPUs sharing the same PCB, the new HD 6870 X2 comes with a custom dual slot cooler, 2GB of GDDR5 memory and a total of 2240 Stream Processors.

As noted, the new card packs a total of 2240 Stream Processors, or 1120 per GPU and works at 900MHz for each GPU and 4200MHz for a total of 2GB of GDDR5 memory paired up with a dual 256-bit memory interface. Since we are talking about a custom card, Powercolor used a "Platinum Power Kit" on the PCB that includes 13-phase VRM, ferrite Core Chokes and DrMos.

Two Barts XT on the same PCB

Previously seen and introduced at Computex in Taipei, the Powercolor Radeon HD 6870 X2 graphics card has now been officially announced. Packing two 40nm Barts XT GPUs sharing the same PCB, the new HD 6870 X2 comes with a custom dual slot cooler, 2GB of GDDR5 memory and a total of 2240 Stream Processors.

As noted, the new card packs a total of 2240 Stream Processors, or 1120 per GPU and works at 900MHz for each GPU and 4200MHz for a total of 2GB of GDDR5 memory paired up with a dual 256-bit memory interface. Since we are talking about a custom card, Powercolor used a "Platinum Power Kit" on the PCB that includes 13-phase VRM, ferrite Core Chokes and DrMos.

The cooler that will keep the temps of both GPUs uses two slots and features six heatpipes with Heatpipe Direct Touch technology as well as two fans to keep the card as cool as possible. The card needs two 8-pin PCI-Express power connectors and features two DisplayPort 1.2, HDMI 1.4 and two DVI outputs.

The latest info was that it should be priced at around US $449 and should end up about 15 percent faster than the GTX 580.

Source: fudzilla.com