E3 2005 - Day 2: More Details Emerge on Console GPUs
by Anand Lal Shimpi on May 19, 2005 1:24 PM EST- Posted in
- Trade Shows
More Detail on the Xbox 360 GPU
ATI has been working on the Xbox 360 GPU for approximately two years, and it has been developed independently of any PC GPU. So despite what you may have heard elsewhere, the Xbox 360 GPU is not based on ATI's R5xx architecture.
Unlike any of their current-gen desktop GPUs, the 360 GPU supports FP32 from start to finish (as opposed to the current FP24 spec that ATI has implemented). Full FP32 support puts this aspect of the 360 GPU on par with NVIDIA's RSX.
ATI was very light on details of their pipeline implementation on the 360's GPU, but we were able to get some more clarification on some items. Each of the 48 shader pipelines is able to process two shader operations per cycle (one scalar and one vector), offering a total of 96 shader ops per cycle across the entire array. Remember that because the GPU implements a Unified Shader Architecture, each of these pipelines features execution units that can operate on either pixel or vertex shader instructions.
Both consoles are built on a 90nm process, and thus ATI's GPU is also built on a 90nm process at TSMC. ATI isn't talking transistor counts just yet, but given that the chip has a full 10MB of DRAM on it, we'd expect the chip to be fairly large.
One thing that ATI did shed some light on is that the Xbox 360 GPU is actually a multi-die design, referring to it as a parent-daughter die relationship. Because the GPU's die is so big, ATI had to split it into two separate die on the same package - connected by a "very wide" bus operating at 2GHz.
The daughter die is where the 10MB of embedded DRAM resides, but there is also a great deal of logic on the daughter die alongside the memory. The daughter die features 192 floating point units that are responsible for a lot of the work in sampling for AA among other things.
Remember the 256GB/s bandwidth figure from earlier? It turns out that that's not how much bandwidth is between the parent and daughter die, but rather the bandwidth available to this array of 192 floating point units on the daughter die itself. Clever use of words, no?
Because of the extremely large amount of bandwidth available both between the parent and daughter die as well as between the embedded DRAM and its FPUs, multi-sample AA is essentially free at 720p and 1080p in the Xbox 360. If you're wondering why Microsoft is insisting that all games will have AA enabled, this is why.
ATI did clarify that although Microsoft isn't targetting 1080p (1920 x 1080) as a resolution for games, their GPU would be able to handle the resolution with 4X AA enabled at no performance penalty.
ATI has also implemented a number of intelligent algorithms on the daughter die to handle situations where you need more memory than the 10MB of DRAM on-die. The daughter die has the ability to split the frame into two sections if the frame itself can't fit into the embedded memory. A z-pass is done to determine the location of all of the pixels of the screen and the daughter die then fetches only what is going to be a part of the scene that is being drawn at that particular time.
On the physical side, unlike ATI's Flipper GPU in the Gamecube, the 360 GPU does not use 1T-SRAM for its on-die memory. The memory on-die is actually DRAM. By using regular DRAM on-die, latencies are higher than SRAM or 1T-SRAM but costs should be kept to a minimum thanks to a smaller die than either of the aforementioned technologies.
Remember that in addition to functioning as a GPU, ATI's chip must also function as a memory controller for the 3-core PPC CPU in the Xbox 360. The memory controller services both the GPU and the CPU's needs, and as we mentioned before the controller is 256-bits wide and interfaces to 512MB of unified GDDR3 memory running at 700MHz. The memory controller resides on the parent die.
Facebook
Twitter
RSS
22 Comments
View All Comments
Shinei - Thursday, May 19, 2005 - link
Jarred, I thought ATI made the XBox 2 GPU specifically for the console, and wasn't incorporating any of its features into the R520? I'm not sure I agree that spending most of your R&D on a "dead-end" GPU is the best tactic; nVidia's approach of optimizing an existing desktop GPU architecture seems to be the more efficient way to spend R&D capital.It also allows nVidia to take any lessons learned from the PS3 GPU and add/modify them when they finally release the G70 (hopefully with fully functional PureVideo, not just "sort of functional" PureVideo--I'm paying for the transistor real estate in price and heat, I better be able to use it this time!)...
JarredWalton - Thursday, May 19, 2005 - link
Low Roller - I wouldn't put too much stock in that figure for the X360 GPU. The way the chip is designed (split in two pieces), I wouldn't be surprised to find that one piece is 150 million and the other is maybe 150 to 200 million.My DRAM knowledge is a bit fuzzy, and the "Embedded DRAM" is something we don't have specifics on, but 10MB of RAM represents 83,886,080 bits, and best case scenario you're using 1 transistor per bit. SRAM uses 6, and perhaps DRAM is 2? 2 transistors per bit would already put just the embedded RAM at 167,772,160 transistors. Heh. 150 million is WAY too small, no matter what IGN says.
As a separate thought, I wouldn't be surprised to see the Xbox 360 GPU end up the more powerful of the two graphics chips. The reason is based on inference: R4xx is very similar to R3xx, meaning ATI didn't spend as much resources creating R4xx as NVIDIA spent on NV4x. If their R&D teams are about equal in size, where did ATI's extra efforts end up being spent? That's right: the Xbox 360 GPU. This is simply a deductive guess, and it could be wrong, but it's something to consider. NVIDIA spent a lot of effort recovering from the NV3x (FX) fiasco.
What makes this all really entertaining to me is the following:
http://www.anandtech.com/news/shownews.aspx?i=2427...
If anything, it seems like the PS3 GPU is more of a PC design with less "future technology". In other words, everything said by MS and Sony is complete hype and should be taken with a massive helping of salt. :)
Iftekharalam - Thursday, May 19, 2005 - link
Which graphics processor will be more powerful?The XBOX 360 or the PS3? The Nintendos future gaming console also uses ATI's GPU codenamed "Hollywood".
Low Roller - Thursday, May 19, 2005 - link
AnandTech's article says they were not able to get a transistor count out of ATI for the Xbox 360.According to IGN, the Xbox 360's GPU only has 150 million transistors, compared to the G70's 300 million.
http://xbox360.ign.com/articles/612/612995p1.html?...
araczynski - Thursday, May 19, 2005 - link
nice info.too bad i could care less which gpu is used in which console, i'm more interested in which console will have some original quality games...
R3MF - Thursday, May 19, 2005 - link
sounds good, shame about the non unified shader model tho.maybe nvidia are right, but i like advanced tech. :p
Shinei - Thursday, May 19, 2005 - link
nVidia's apparently pulling out the 16" battle cannons with RSX/G70--136 shader ops per clock is damn impressive, regardless of whether the GPU is console or desktop...And if nVidia says the desktop G70's going to be even more powerful than RSX, I'm willing to bet that there will be at least 10 shader units pumping to 24+ pipes in full FP32 quality. Nice. :)
AnandThenMan - Thursday, May 19, 2005 - link
Very very interesting article. ATi and NVIDIA seem to have diverging paths. All things considered, this tells me that ATi has the more advanced GPU. Does that mean a faster GPU though.EODetroit - Thursday, May 19, 2005 - link
How about anything about the new physics processor?Garyclaus16 - Thursday, May 19, 2005 - link
...sooo much for my bfg6800U...seems like I'm already way behind again.