Here you can find all of the information about ATI. This includes their GPU’s, as well as their parent company AMD and the related software like the Stream SDK. In addition, we have information regarding CrossFire and EyeFinity hardware systems.
You may also want to check out:
AMD annouced today the FireStream 9350 and 9370 graphics processing units (GPU) for general purpose computing (GPGPU). Both of the GPUs are scheduled to be release in the 3rd quarter of 2010.
The FireStream 9350 will come with 1,440 SIMD cores running at 700 MHz. The GPU will also come with 2 GB of GDDR5 graphics memory running at 1 GHz. The GPU supports OpenCL, DirectX 11 and OpenGL. The GPU will consume 150 Watts of power, and will cost $799.
The FireStream 9370 will come with 1,600 SIMD cores running at 825 MHz. The GPU will also come with 4 GB of GDDR5 graphics memory running at 1.15 GHz. The GPU supports OpenCL, DirectX 11 and OpenGL. The GPU will consume 225 Watts of power, and will cost $1,999.
via : AMD Press Release
Icrontic take the new ATI FirePro V8800 out for a spin against all of it’s brethren in the ATI and NVidia professional camps, and finds it worthy of praise. In particular, they like the Eyefinity feature for 4-displays on one card.
Perhaps an even greater addition to the FirePro V8800 is support for AMD’s multiple monitor fete, Eyefinity Multi-display. The GPU supports up to four display outputs at once via the four Displayport connectors. The GPU ships with Displayport to DVI converters to increase display compatibility out of the box. Having the ability to output to an array of four displays from one GPU is a definite killer app in the busy DCC industries. To an artist, it is invaluable to have that much screen real estate.
In the picture above, see the 3 monitors and a Wacom Cintiq all running on the one card with ease. Currently working in a lab where we’re trying to find a way to get Dual 30″ displays and a Cintiq working with NVidia cards, this is particularly appealing.
via ATI takes the workstation crown: FirePro V8800 reviewed « Icrontic Tech.
Now that NVidia finally has DX11 hardware available, Microsoft has the chance to compare how well the ATI RadeonHD cards stand up to the new GTX480/470. Specifically, they were looking for oddities in the DirectX11 implementation, and found one particularly interesting one in NVidia’s use of the “Feature Level” concept.
The ATI Radeon HD 5000 Series only provides one quality level per sample count, while the NVIDIA GeForce GTX 470/480 exposes a number of fine-grain quality levels per sample count. This highlighted a few UI bugs in some of the samples as well as DXUT/DXUT11 that were corrected in the June 2010 release. Be sure to test the behavior of any MSAA settings and quality levels in your DX10.x and DX11 programs on both vendor's hardware.
The guys at Geeks3d took both cards for a spin to get the actual results from the ‘CheckMultisampleQualityLevels’ Microsoft mentions, and sees the obvious difference.
Here are the details for a GTX 480 (with R257.15 drivers – Win7 64-bit):
Direct3D 10 - Adapter 0 - Description: NVIDIA GeForce GTX 480 Direct3D 10 - Adapter 0 - Dedicated video memory: 1503MB Direct3D 10 - Adapter 0 - vendorId: 10DE, deviceId: 06C0, revision: 00A3 Direct3D 10 - MSAA 2X supported with 3 quality levels Direct3D 10 - MSAA 4X supported with 17 quality levels Direct3D 10 - MSAA 8X supported with 33 quality levelsAnd here are the details for a HD 5870 (with Catalyst 10.5 – Win7 64-bit):
Direct3D 10 - Adapter 0 - Description: ATI Radeon HD 5800 Series Direct3D 10 - Adapter 0 - Dedicated video memory: 1014MB Direct3D 10 - Adapter 0 - vendorId: 1002, deviceId: 6898, revision: 0000 Direct3D 10 - MSAA 2X supported with 1 quality levels Direct3D 10 - MSAA 4X supported with 1 quality levels Direct3D 10 - MSAA 8X supported with 1 quality levels
Computex 2010 – AMD Booth Tour from Ryan Shrout on Vimeo.
While we have already had some interesting news come out of AMD this week, we couldn’t leave without stopping by the AMD booth to see what they had on display. While we of course see the normal wall of motherboards and wall of graphics cards, demos of Opteron processors and some of the new Vision-based notebooks, the killer spots were two sets of 5 portrait Samsung displays running HUGE Eyefinity configurations.
via : PC Perspective
What is the fastest laptop graphics you can buy? Well, from the image to the right you could easily guess that it is the Nvidia GeForce GTX 480M. However, no one seems to have one just yet for testing. Therefore, Anandtech has taken the AMD Mobility 5870 and the Nvidia GTX 285M and tested them against one another to see which is the fastest. The AVADirect Clevo W860CU laptop supports both types of graphics chips.
These laptops are configured absolutely identically except for a crucial difference: one is equipped with NVIDIA’s flagship GeForce GTX 285M, and the other with AMD’s flagship Mobility Radeon HD 5870. These are the fastest (current) single chip mobile graphics solutions from either vendor. NVIDIA just announced their upcoming GeForce GTX 480M, which will boost performance (and power consumption) and very likely take a clear lead over today’s combatants, but we’ll hold off on saying more about the 480M until we can actually get one for testing.
Who wins? You will have to read the article to see. Let’s just say that I was surprised at who the winner was.
via AVADirect’s Clevo W860CU: Mobility 5870 vs. GTX 285M @ AnandTech .
At ISC10, a new Top500 list was published and showed one major upset: China’s “Nebulae” supercomputer unseated long-time Top500 member Roadrunner for the #2 spot. The #1 spot remains the Cray “Jaguar” at Oak Ridge National Labs, with an impressive 1.7 PetaFlops of performance, but China has made an impressive showing.
The Nebulae computer clocked in at 1.27 Petaflops by combining Quad-Core Intel Xeon processors with NVidia Tesla C2050′s. While I can’t find concrete numbers anywhere, piecing together various news articles shows:
While the .5 probably means they have some nodes with Tesla’s and some without, it’s a surprising density.
Update 6/1/10: A reader pointed out that, unbeknownst to me, the Top500 list includes the count of GPU cores in the total core count. From his website, he theorizes:
Based on all this information, it is almost certain that Nebulae is built on 4640 nodes, where each node has two X5650 processors and one C2050 GPU, for a total of 9280 processors and 4640 GPUs:
4640 nodes * (12 processor cores + 14 SIMD units) = 120640 cores
Why so many? This comment from the EEtimes sums it up nicely:
The graphics processors are also relatively power efficient. The Nebulae system which uses 4,640 Tesla chips, consumes about 2.55 megawatts compared to about 7 MW for Jaguar.
So it made #2 on the Top500 list with only 1/3rd of the power. But that’s not all, China has another system in the Top10, the Tianhe-1 which couples Intel processors with ATI Radeon 4870′s at an impressive 1:1 ratio (2 CPU’s and 2 GPU’s in each node). Granted, they have only half the nodes of Nebulae at 71680 cores ( = 17920 CPU’s, or 8960 nodes) but that’s enough to put them in the Top 10. And China isn’t stopping there (from the Daily Pioneer):
The super computer named “Xingyun”, has been developed in Tianjin, and works at double the speed of “Tianhe-1″, the previous fastest machine in China.
The Tianhe-1 was developed by the National University of Defence Technology in October 2009, Li Jun, president of the Dawning Information Industry Co. Ltd., was quoted as saying by Xinhua.
“Its peak performance reaches nearly three quadrillion calculations per second, three times the peak speed of Tianhe-1,” Li said.
This new machine is expected to be delivered by the end of 2010, so China may have another upset in the next Top500 if ORNL doesn’t step up the game with Jaguar.
If you want to see the list, but don’t want to dig through the text, then check out this neat interactive Treemap that BBC published online.
Anandtech has posted a review of three overclocked Radeon HD 5870 graphics cards: Sapphire’s Radeon HD 5870 Toxic 2GB, MSI’s Radeon HD 5870 Lightning, and Gigabyte’s Radeon HD 5870 Super Overclock. These three cards are clearly in the high-end gaming category and all three boast such a price since they retail for around $500. How well do they perform? While they are all faster than the Radeon HD 5870 (of course), they all come slower than the Radeon HD 5970, which is to be expected. Also, don’t look to overclock them much as they are already right at the edge of what the graphics chip can handle.
Looking at the larger issues, we haven’t mentioned price too much up until now since the focus of this article is the roundup, but at $480-$500 for these cards it’s time to discuss the issue. Even when overclocked these cards are only at most 10% faster than a reference Radeon 5870 for a 20%-25% higher price. These are luxury cards that are not cost-effective, and at $500 are running right in to NVIDIA’s GeForce GTX 480.
Guru3d has posted a picture of the new Eyefinity12 graphics card from Powercooler. This graphics card is based on the AMD Radeon HD 5970 graphics card, and sports 12 mini-DisplayPorts. The graphics card will take up 3 slots. A normal Radeon HD 5970 would come with 2 GB of GDDR5 memory. This new version has 4 GB of GDDR5 memory.
Remember that the Radeon HD 5970 is comprised of two Cypress graphics processors connected by a PCI-Express bridge. A normal Radeon HD 5970 has a core clock of 750 MHz and a memory clock of 1000 MHz. This is slower than Radeon HD 5870 with a core clock of 850 MHz and a memory clock of 1200 MHz.
While this is an interesting technological achievement, I do not think that we will see this sell very many units. The first problem is that not many people will be able to invest in 12 new monitors that support DisplayPort inputs. Not to mention that a 12 monitor wall is huge. Second, I am not sure how well it will perform with the latest games. Third, is that the bezels start to get in the way.
AMD has just refreshed their ‘Vision’ technology and plans to bring AMD to 109 notebooks and 26 ultrathin notebooks this year. The technology boasts 8 hours of battery life, DDR3 memory, and DirectX10.1 support, putting it on par with the latest battery-conserving offerings from NVidia and Intel.
In testing with HQV 2.0, a VISION-based system with AMD integrated graphics achieved a video experience score almost 2x that of a comparable Intel-based system. Additionally, a VISION-based system with AMD discrete graphics scored 50% better than a comparable Intel-i5 based system with Nvidia discrete graphics.
via AMD Vision Refresh: Better, Faster, More – Amd vision 2010 – Gizmodo.
Microway, a systems integration specialized focusing on high-end GPGPU systems, has published a PDF slide deck showing the various capabilities and specifications on the higher end NVidia and ATI video cards. It includes details on memory structures, silicon design, core counts, and much more in easy tabulated forms.
Read the whitepaper here (Online with Google Viewer) and check out their website.
VizWorld.com Got an idea for a story? Drop us a line at [email protected] or other means.
Comments