Subject: Graphics Cards | June 8, 2012 - 05:23 PM | Tim Verry
Tagged: tahiti, graphics, gpu, computex, binning, amd, 7970 ghz edition
AMD is having a string of successes with its 28nm 7000 series graphics cards. While it was dethroned by NVIDIA’s GTX 680, the AMD Radeon HD 7970 is easier to get a hold of. It certainly seems like the company is having a much easier time in manufacturing its GPUs compared to NVIDIA’s Kepler cards. AMD has been cranking out HD 7970s for a few months now and they have gotten the binning process down such that they are getting a good number of pieces of silicon that have a healthy bit of overhead over that of the 7970’s stock speeds.
And so enters Tahiti 2. Tahiti 2 represents GPU silicon that is binning not only for HD 7970 speeds but is able to push up the default clock speed while running with lower voltage. As a result, the GPUs are able to stay within the same TDP of current 7970 cards but run faster.
But how much faster? Well, SemiAccurate is reporting that AMD is seeing as much as a 20% clock speed improvement over current Radeon HD 7970 graphics cards. This means that cards are able to run at clock speeds up to approximately 1075MHz – quite a bit above the current reference clock speed of 925MHz!
The AMD 7970 3GB card. Expect Tahiti 2 to look exactly the same but run at higher clock speeds.
They are further reporting that, because the TDP has not changed, no cooler, PCB, or memory changes will be needed. This will make it that much easier for add in board partners to get the updated reference-based GPUs out as quickly as possible and with minimal cost increases (we hope). You can likely count on board partners capitalizing on the 1,000MHz+ speeds by branding the new cards “GHz Edition” much like the Radeon 7770 has enjoyed.
With 7970 chips having overhead and binning higher than needed, an updated and lower-power using refresh may also be in order for AMD’s 7950 “Tahiti Pro” graphics cards. Heck, maybe they can refresh the entire lineup with better binned silicon but keep the same clock speeds in order to reduce power consumption on all their cards.
Subject: Graphics Cards | June 7, 2012 - 06:40 AM | Tim Verry
Tagged: video, sapphire, radeon 7770, passive cooling, graphics card, gpu, computex
Not to be left out of the Computex news, graphics card manufacturer Sapphire Technology unveiled a passively cooled AMD Radeon 7770 graphics card running at reference clock speeds. Following the release of the company’s factory overclocked Vapor-X 7770, the new Sapphire HD 7770 Ultimate 1GB card is the first to sport a passive cooler – other vendors are going in the opposite direction by using custom (active) coolers to push up reference clockspeeds for factory overclocked cards.
What makes the sapphire card neat is that the company did not have to underclock the GPU or memory in order to make a passive cooler feasible. With this card, you will get a silent GPU with the same specs and speeds as the reference 7770 we recently reviewed. The card looks to take up about two PCI expansion slots and utilizes a horizontal stack of vertically aligned (if that makes sense?) aluminum fins connected to the GPU via four heatpipes. Because of the cooler, the card is about 25% longer than a reference card, so keep that in mind if you are considering this for a HTPC build using a tiny case.
Beyond the cooler, which is arguably the most important aspect of the card, the Saphhire 7770 Ultimate 1GB is nearly identical to AMD’s reference design. The only major change is that Sapphire had to move the GDDR5 memory chips to the opposite (top, when installed in the case) side of the PCB in order to accommodate the cooler. With that said, the video outputs on the graphics card are a small improvement over the reference design with an additional DVI port (thanks to not needing a full fan grill in the second PCI slot) bringing the total to two DVI ports, one full size HDMI, and one full size DisplayPort. Otherwise, the GPU is stock, running at 1GHz while the 1GB of GDDR5 memory is likely running at 1125 MHz (stock speeds). The Cape Verde-based graphics card contains 640 stream processors, 1.5 billion transistors, 1.28 Teraflops of compute performance, and a Texture fill rate of 40 giga-transfers per second (GT/s). The full specifications of the 7770 GPU core can be found in our review.
The MSRP of reference AMD HD 7770 cards is $159 but expect the Sapphire card to come in a bit above that number thanks to the custom cooler. You can find more photos of the passively cooled Sapphire GPU over at AnandTech who managed to snag some good shots of the card at the company’s Computex booth.
In case you missed it, our video review of the HD 7770 card is embedded below in which we show off the (7770 and 7750) card also show off several custom 7770 designs from MSI, XFX, and others. It should bring you up to speed on what the 7770 is and where it stands in terms of performance with other cards from AMD and NVIDIA.
Earlier this year I had the chance to take a look at the first ultrabook with discrete graphics, the Acer Aspire Timeline Ultra M3. My review was not particularly favorable, but the idea of placing discrete graphics in an ultrabook is both compelling and necessary. Intel’s low-voltage processors have difficulty with gaming when paired with the HD 4000 IGP and this flaw is difficult to excuse in products typically priced at $800 or above.
Four new ultrabooks with NVIDIA discrete GPUs have been unveiled to tackle the problem of gaming with a slim laptop. The list includes two laptops from Acer, two laptops from Gigabyte and one from ASUS.
The Gigabyte U2442N, which has a 14” 1600x900 display and a GeForce GT 650M GPU, is obviously the most powerful and the product that offers the most promising gaming experience on paper. Only the ASUS UX32 looks questionable. There’s no way that a GeForce GT 620M is going to handle gaming on a 1080p display.
Unfortunately, a closer look at the announcement suggests these product lines aren’t that exciting. The Gigabyte laptops have received a lot of positive attention, but Gigabyte has no meaningful presence in the North American laptop market and it’s nearly guaranteed the laptop won’t be popular on this side of the pond. The Acer M5-581TG appears to be an Ivy Bridge updated version of the Acer Aspire M3 that we reviwed – and did not like – while the M5-481TG is just a smaller version.
That leaves the ASUS UX32 and its GT 620M which, although likely quicker than Intel HD 4000, isn’t sufficient for serious gaming.
Hopefully NVIDIA will be able to bring discrete graphics to more products from larger manufacturers, but the fact so few companies have gone this route suggests there is some underlying reason. My personal guess? Heat. The Acer Aspire M3 became quite toasty during load. It’ll be interesting to see if the U244N has some design trick that makes the GT 650M manageable – or if Gigabyte, like Acer, doesn’t mind putting out a laptop with high exterior temperatures.
NVIDIA recently revised its notebook GPU lineup, but there was one part notably missing – the GTX 680M. The x80M part has been NVIDIA’s fastest mobile GPU in each generation for some time, so we knew that a GTX 680M was coming. We had only two questions. When? And what architecture will it be based on?
Now we have the answers to both questions. NVIDIA has pulled the wraps off its flagship component. The new GTX 680M is a Kepler component (unlike other high-end 600 series parts, which are Fermi) packing 1344 CUDA cores and up to 4GB of GDDR5.
The green team is laying out some big numbers in its press release by claiming that performance is up about 80% in comparison to the GTX 580M and 30% in comparison to the AMD Radeon 7970M. NVIDIA also says that the new part will play every game available today at 1080p with maximum in-game settings.
Other selling points include NVIDIA’s FXAA and TXAA, Adaptive V-Sync, Optimus, SLI, PhysX and 3D Vision. The company is clearly making a strong effort to distinguish itself from AMD not only with performance but also with features.
Five launch laptops were announced. They include the Alienware M17x and M18x, the MSI GT70 and the Clevo P150EM/P170EM. The Clevo units are the chassis used by companies like Maingear, Origin and other boutiques. Only the M18x has confirmed SLI support and only the M17x has confirmed 3D Vision support. Pricing has not been announced.
We will of course be looking to obtain a review unit so we can see if the GTX 680M is as mightly as claimed.
Subject: Graphics Cards | May 23, 2012 - 07:57 PM | Jeremy Hellstrom
Tagged: jon peddie, sales, gpu
Jon Peddie's newest report on the state of the graphics card market has arrived and while the news is not good it is nowhere near as bad as it could have been. The graphics card market had a very large hurdle to deal with over these last few quarters which is why the total market declining 0.8% from last quarter and a 3.38% decline from Q1 2011 is not terrible news. The impact came from the flooding in Thailand, which has been causing lowered sales for most of the PC market this year. With hard drives being in such short supply the number of systems that could be built by vendors dropped dramatically, those systems which were built were noticeably more expensive than before the flood as the price of hard drives doubled in some cases. With less systems being built and sold there was less demand for GPUs from the vendors, thankfully the industry has recovered from the shortage and we are seeing prices and supplies returning to their normal levels.
When you break it down by company, only AMD saw growth from last quarter, though at a 0.3% increase it was not so much growth as simply holding their ground. NVIDIA has stopped reporting on their IGP sales which, believe it or not, still sell in Asia and so saw a drop of 4.5% from last quarter. Some of that decline will be due to the change in reporting but the lack of Kepler stock has certainly hurt their sales as well. Intel saw a decline of 1.3% from the previous quarter again likely due to the influence of the hard drive shortage reducing the number of systems which were sold.
When you look at only discrete cards, the sales increased 2.7 % from the last quarter but were down 11% from this time last year, thanks to the GPU now present on both AMD and Intel processors. With Llano and Trinity as well as SandyBridge and Ivy Bridge we saw the arrival of onboard graphics which provided enough horsepower that many casual users no longer need a discrete GPU. Previous generations of IGPs and onboard graphics cores struggled to play HD video without stuttering and they were essentially useless if they were called upon to power even casual games. The new generations of processors can not only handle HD video but are quite capable of light gaming duties. They also made possible tablets and extremely small laptops, aka Sleekbooks and Ultrabooks, which provided good enough performance for many users and these small form factors have little space for discrete GPUs. As both AMD and Intel's processors have a graphics core they count towards the total graphics card market share which is good news for them but not for NVIDIA who count on add in card sales exclusively. On the plus side, when you examine add in card sales for laptops alone, NVIDIA actually saw a gain of 5%.
Subject: Graphics Cards | May 21, 2012 - 10:10 PM | Tim Verry
Tagged: recall, nvidia, kepler, graphics cards, gpu
Editor's Note: We are getting a lot of flak for posting this story today, telling us that we are "giving the site credibility", referring to the Pnosker site that first started the recall rumor, simply by posting about it on our site. Even though our post by Tim states to "take the leak with a grain of salt" and that "these GPUS go through rigorous testing and certification", some people think we were in the wrong to post about this.
So let me be perfectly clear - the recall referenced in the story below is almost assuredly complete and utter BULLSHIT.
According to Pnosker, NVIDIA is allegedly looking into recalling all Kepler based, 600-series graphics cards. Such a recall would affect users that have purchased GTX 670, GTX 680, and GTX 690 GPUs. The website has stated that their source has indicated that the graphics cards will possibly be recalled because the chips suffer from performance degradation after prolonged periods of heavy usage.
While their source has reportedly been correct in the past, the author cautions readers to take the leak with a grain of salt. Other websites that have picked up on this have mentioned that these GPUs go through rigorous testing and certification processes before getting to the market, so this rumor does not have much ground to stand on. Another reason to take this report with a shaker-full of salt is that if there was such a defect in the Kepler GPU, it would be more likely to completely fail rather than continue working with degraded performance.
This rumor is likely just that: a rumor. Why such a rumor was started is unknown but your Kepler graphics card purchases are probably safe from performance degradation, though they may not get as high of a boost clock as other users’ cards.
UPDATE @ 7:30pm ET: To quote from NVIDIA PR - "There is no truth to this rumor."
Subject: Graphics Cards | May 17, 2012 - 02:40 AM | Tim Verry
Tagged: nvidia, gtx 680m, gpu, mobile, kepler
Videocardz.com managed to get their hands on some rumored details about an upcoming NVIDIA mobile graphics card–the GTX 680M. According to rumors, the mobile chip will be launched at Computex 2012 in Taiwan next month.
There aren’t many details about the mobile chip, but it is set up to be a scaled down version of it’s Kepler based GTX 680 desktop counterpart. The GTX 680M will have approximately half as many CUDA cores at either 744 or 768 cores depending on the source. Either way, the card keeps the same 256-bit memory interface and can support SLI configurations. In addition, the 680M will be able to have up to 4GB of GDDR5 memory. Reportedly, it can use as much as 100 Watts of power.
When paired with an Intel Core i7 3720QM processor, the GPU was able to get a score of 4,905 points in 3DMark 11’s Performance present benchmark. It is supposed to be as much as 37 percent faster than the GTX 670M, which is not surprising considering that chip has only 336 CUDA cores and is clocked at 598 MHz (no word yet on what the GTX 680M will be clocked at).
No matter what the GTX 680M turns out to be, you can bet it will only be found in the highest end gaming notebooks where performance is more important than battery life. Until then, feel free to brush up on your Kepler architecture knowledge by visiting our GTX 680 (desktop) review.
Subject: Processors | May 16, 2012 - 06:29 PM | Jeremy Hellstrom
Tagged: trinity, radeon, igp, gpu, APU, amd. A10-4600M
AMD's A10-4600M APU has finally arrived, showing off an enhanced Piledriver core and a new Northern Islands based graphics core. This is a big step up from Llano in terms of general processing power but not a huge improvement over Bulldozer chips, though the raised clock speed does help it in general tasks. Unfortunately the AMD still chip lags far behind the performance of Intel's mobile i5 processors and while the graphics are certainly more powerful on Trinity they still aren't up to an impressive level of performance. The Tech Report liked the high end A10-4600M but think that Trinity's low power chips are really going to shine in inexpensive ultraportable machines.
You can also check out Matt's review of Llano in a reference laptop from AMD for more information.
"AMD has pulled the curtains back on Trinity, its next-generation APU, which features new Piledriver CPU cores and Northern Islands-derived integrated graphics. Join us as we outline Trinity's architecture and run it through a whole host of benchmarks, from old staples to OpenCL-accelerated apps and "inside the second" gaming tests."
Here are some more Processor articles from around the web:
- AMD Launches New Trinity APU @ TechwareLabs
- The AMD Trinity Review (A10-4600M): A New Hope @ AnandTech
- AMD A10 'Trinity' APU review @ Hardware.Info
- AMD Launches New 2012 A-Series APU (Trinity) @ Bjorn3d
- AMD Trinity Preview @ Neoseeker
- AMD Trinity A10-4600M APU Review: Jumping the Shark? @ VR-Zone
- AMD Trinity: Going Mobile with a New APU @ Hardware Canucks
AMD’s position is not enviable. Though they’re the only large competitor to Intel in the market for x86 processors, the company is dwarfed by the Giant of Santa Clara. As a resident of Portland, I can’t forget this fact. Intel offices are strewn across the landscape of the western suburbs, most of them at least four times larger than any office I’ve worked at.
Despite the long odds, AMD is set in this course for now and has no choice but to soldier on. And so we have today’s reference platform, a laptop powered by AMD’s latest mobile processor, codenamed Trinity. These processors, like the older Llano models, will be sold as the AMD A-Series. This might lead you to think that it’s simply another minor update, but that’s not the case.
Llano was released around the same time as Bulldozer, but it did not use Bulldozer cores. Instead it used yet another update of Stars, which is a mobile incarnation of Phenom II, which was of course an improvement upon the original Phenom. The “new” Llano APU in fact was equipped with some rather old processor cores. This showed in the performance of the mobile Llano products. They simply could not keep up with Sandy Bridge’s more modern cores.
Bulldozer isn’t coming to mobile with Trinity, either. Instead we’re receiving Piledriver. AMD has effectively skipped the first iteration of its new Bulldozer architecture and moved straight on to the second. Piledriver includes the third generation of AMD’s Turbo Core and promises “up to 29%” better processor performance than last year’s Llano-based A-Series.
That’s a significant improvement, should it turn out to be correct. Is it true, and will it be enough to catch up to Intel?
It seems like MSI has some cheaper Kepler based NVIDIA graphics cards coming soon if this photo from Malaysian retailer Cycom turns out to be legitimate. Spotted by Lowyat user Chapree, the photo appears to be an MSI GTX 670 graphics card. Further, the card appeared on the company’s website at a price of 1399 Malaysian ringgits which translates to just under $462 USD.
Interestingly, the box contains a typo for Displayport in the form of “DispalyPort.” While that may indicate a fake card, it doesn’t totally rule it out either. The NVIDIA GTX 680 is the only available Kepler card (if you can find one to buy that is!), and many users are clamoring for some cheaper options. Here’s hoping they are coming sooner rather than later!