ATI Releases More "R600" Details

HEHHEE ill post the whole thing

320-stream processors, named ATI Radeon HD 2900

AMD has named the rest of its upcoming ATI Radeon DirectX 10 product lineup. The new DirectX 10 product family received the ATI Radeon HD 2000-series moniker. For the new product generation, AMD has tagged HD to the product name to designate the entire lineup’s Avivo HD technology. AMD has also removed the X-prefix on its product models.

At the top of the DirectX 10 chain, is the ATI Radeon HD 2900 XT. The AMD ATI Radeon HD 2900-series features 320 stream processors, over twice as many as NVIDIA’s GeForce 8800 GTX. AMD couples the 320 stream processors with a 512-bit memory interface with eight channels. CrossFire support is now natively supported by the AMD ATI Radeon HD 2900-series; the external CrossFire dongle is a thing of the past.

The R600-based ATI Radeon HD 2900-series products also support 128-bit HDR rendering. AMD has also upped the ante on anti-aliasing support. The ATI Radeon HD 2900-series supports up to 24x anti-aliasing. NVIDIA’s GeForce 8800-series only supports up to 16x anti-aliasing. AMD’s ATI Radeon HD 2900-series also possesses physics processing.

New to the ATI Radeon HD 2900-series are integrated HDMI output capabilities with 5.1 surround sound. However, early images of AMD’s OEM R600 reveal dual dual-link DVI outputs, rendering the audio functions useless.

AMD’s RV630-based products will carry the ATI Radeon HD 2600 moniker with Pro and XT models. The value-targeted RV610-based products will carry the ATI Radeon HD 2400 name with Pro and XT models as well.

The entire AMD ATI Radeon HD 2000-family features the latest Avivo HD technology. AMD’s upgraded Avivo with a new Universal Video Decoder, also known as UVD, and the new Advanced Video Processor, or AVP. UVD previously made its debut in the OEM-exclusive RV550 GPU core. UVD provides hardware acceleration of H.264 and VC-1 high definition video formats used by Blu-ray and HD DVD. The AVP allows the GPU to apply hardware acceleration and video processing functions while keeping power consumption low.

Expect AMD to launch the ATI Radeon HD 2000-family in the upcoming weeks, if AMD doesn’t push back the launch dates further.

AH $CREW THAT !!

I want Laptop Info !!

Oh BTW, nice find.




HEhe, I like emoticons......

Ati previously confirmed this is a 720 million transistor part. The Geforce 8800 has over 680 million transistors. Less than 40 million more transistors to work with.

Either the current G80 cards have a lot of their functionality disabled, or the stream processors on the R600 are not exactly comparable to a stream processor on the Geforce 8800.

Either way, I wish they'd just release this thing so we know how fast it goes. I could be interested in buying a more high end video card soon.

same here

I knew it was gonna be a monster but this exceeds what I thought it was gonna have by a mile...

OMG if that's true it's going to wallop the 8800's

I say... Release the damn thing for craps sake!

so close, yet so far...

I hope it does justice to its extra late release.

You can't compare the stream processors the ATi cards to that of the 8800GTX's. Case in point...R600XT Scores 12K in 3DMark06. So how is that it only scored slightly better, but had over 200 more stream processors? Since everyone is new this design we don't know whether the stream processors are clocked lower or they have a smaller effect on performance than originally believed.

That's because the ATI has 320 stream shaders, but the G80 has 128 unified physical shaders and a higher core clock.

What the R600 really has are 64 unified shaders that run up to 5 operations per clock faster than Nvidia's unified shaders.

The problem is, performance like that is to software/application dependent. What you will realistically see is some games running 100% faster on Nvidia and some games running 25% faster on ATI. Why? Because when an application doesn't utilize the faster shader architecture, Nvidia has the brute force unified shaders combined with core clock to destroy ATI's 64 shaders. But even when applications DO utilize the more operations per clock cycle, they can only yield about 17% more performance than the 8800GTX because of the 8800GTX's brute power.

This is probably why ATI has so many non disclosure acts all over their R600's. Notice some people say, "We sometimes see a 17% performance gain on the 8800GTX" and they also sometimes say "But in other games this gain isn't seen".

Flip a coin, heads = your game will be Nvidia optimized, tails = your game will be ATI optimized.

8O WOW, all I can say is .... Curve ball!

and this is coming from someone running a pentiumD

Its quite ironic that you even mention this because this is exactly what DX10 was supposed to get rid. It was supposed to "unify" the shaders and what not so that games could be more easily optimized to run on these graphic cards. Allowing for programmers to take more advantage of the power they have. Unfortunately if your comment holds true this is just a continuation of the the game to game specific performance of certain graphic cards.

It will... 3D mark is DX9.. under DX10 expect nearly 100% performance advantage for X2900XTX over the 8800GTX.

ATi is trying to pull another R300, this time with DX10. X2900XTX will be the fastest DX9 card but not by much.. just like Orton claimed. But under DX10.. it's built for DX10 and VISTA. nVIDIA will need a new architecture to compete.

its being said that the onboard audio is strictly for HDMI use, meaning sound cards will still be needed if you aren't using HDMI.

I don't see why an HDMI-to-DVI connector with digital audio plug wouldn't work to use it on a external set of speakers, via digital input from the digital out adapter on the adpater piece.

That's because the ATI has 320 stream shaders, but the G80 has 128 unified physical shaders and a higher core clock.

What the R600 really has are 64 unified shaders that run up to 5 operations per clock faster than Nvidia's unified shaders.

The problem is, performance like that is to software/application dependent. What you will realistically see is some games running 100% faster on Nvidia and some games running 25% faster on ATI. Why? Because when an application doesn't utilize the faster shader architecture, Nvidia has the brute force unified shaders combined with core clock to destroy ATI's 64 shaders. But even when applications DO utilize the more operations per clock cycle, they can only yield about 17% more performance than the 8800GTX because of the 8800GTX's brute power.

This is probably why ATI has so many non disclosure acts all over their R600's. Notice some people say, "We sometimes see a 17% performance gain on the 8800GTX" and they also sometimes say "But in other games this gain isn't seen".

Flip a coin, heads = your game will be Nvidia optimized, tails = your game will be ATI optimized.

ATI's 'stream' shaders are unified shaders... they are second generation shaders in comparison to Nvidia's unifed shaders, which are technically their first generation of unified shaders. ATI's complex shader approach can yield just as much performance boost, if not more, than the brute force shaders which are clocked to high frequencies. So immediately, you can't pass off ATI's r600 as weak in comparison to Nvidia's because if you try to compare them directly and not through actual benchmarks and game performance, then its impossible to see how ATI's method could beat Nvidia's, but considering ATI's 4-5 issue shaders can do a lot more than Nvidia's shaders, and knowing ATI's experience with DX10 capable chipsets, then its best not to underestimate them. I'm sure the R600 won't knock the socks off G80 in DX9, but once DX10 comes into full swing, I think we'll the true power of both chips come to light, and then we'll be able to truly judge which one's the more powerful one. For one, I would like to see how geometry shaders would implemented in games and how it would improve performance, considering what Crysis will probably demand. So considering what you've said, I just can't agree with you completely. Sure, the whole application dependency due to driver optimisation will still be an issue, though ATI have already shown how threaded GPU architectures do bring about performance benefits over the old pipeline architectures, so I feel confident their complex shader approach will yield similar results. Though, the only really way to find out is to wait for the card to be released and then to be tested and benchmarked.

entirely software (maybe driver) dependent

They never said how or what it is about the card that made it have 320 stream processors, which is yay we can F@H much faster now, streams don't mean much for gaming unless they're doing physics. Then the question arises at what cost does utilizing stream processors along side unified shaders, bring about? Do you lose some shaders to gain a few streams or is nothing lost?

for some reason i do not like this stream processor tag being applied to ATI. i thought that was a term that nvidia used for the 8800. i don't think that is what ATI told dailytech, unless they are trying to play the numbers game like AMD did with intel. i.e even though AMD did more with what speed they had people still liked the high ghz numbers intel used.

still, until i see some proper reviews, i am not going to assume anything and definately won't pass final judgement on dx10 perfromance until a game comes out which doesn't have dx10 tacked on.

I have 2 Gefen HDMI splitters that i use to split a HD single from 2 recievers to four high defintion TV's. Unfortunately Gefen's tech support is not the greatest and their product didn't work very well either. Regardless, through numerous trials i was able to get it work. Also it seems as though Gefen is really the only quality maker of units like this dealing with DVI and HDMI splitters and what not.

If ATi were smart they would make their own unit that could split the HDMI single into audio and video, since obviously no computer monitors have HDMI connectors or would never have quality speakers. Even more smart would be if they package it with the the 2900XT

Wanna know what game I play? Counter Strike: Source.

Flight Simulators are for nerds, Crysis is going to be a flop.


Unless you were mucking about, that's just a tad daft.

What I really want to know is what are they actually doing with the 8900gtx? Because if you remember with the transistion from 7800gtx to 7900gtx, all nvidia did was clock the 7900gtx higher, same amount of pipelines, the whole shebang. And if nvidia expects people to want to pay more for an extremely high clocked 8800gtx with a massive noisy fan like the 7900gtx's and no extra shaders to compete with teh r600's 320, I'm gonna say it now, they're screwed. Ati killed nvidia when they unvieled the 7900gtx because ati added more shader processors and gained a bit more performance aside from the higher clocks

I agree, I'd like to see if nvidia is actually doing anything to the g80 core for the 8900gtx besides clock it up higher, since that's all they did pretty much with the 7900gtx

The GTX 512 was a true powerhouse at its time of release, it swept the board with both the vanilla 7800GTX and X1800XT. Unfortunately for Nvidia as the X1800 was delayed ATI was already done with the X1900XTX. Nvidia also had the 7900GTX done by then, but for all of its muscle it still fell short of the X1900 IMHO.

I also think that last generation is being mirrored in this current generation, as Nvidia got the first cards in a generation out first, and ATIs were delayed again and again. If the pattern is followed again, the HD2900 will compete with the 8800GTX, Nvidia will release a new card (Ultra or 8900?) then ATI releases its real powerhouse of the generation (XTX or 2950?) which will hold its own until Nvidias 9800 release.

lol at the newbie.