maverik-sg1
New member
The Inq reports:
ATI BELIEVES that you need many more pixel operations per clock than the pixel itself. Nvidia disagrees. It believes in a totally different ratio. With the birth of the R580, ATI believes that you should have three to one ratio as R580 has 48 Shaders and 16 pixel pipelines. You don’t need much mathematics to figure out that ATI preaches a three to one ratio.
As the G71 will be able to process 32 pixels per clock, it will physically have 32 pipelines but will have just 16 ROPs. This is no surprise to us as Nvidia's G70, Geforce 7800 GTX actually has 24 pipelines and 16 ROPs.
This means that Nvidia and ATI will take completely new approaches, where Nvidia will try to convince everyone that you need more raw pixels than pixel operations.
The G70, Geforce 7800 GTX has 24 texture memory units and 16 ROPs and can theoretically draw eight pixels with each, having three textures per clock. G71 will be able to do even more pixels per clock - it's as simple as that.
ATI's R580, on the other hand, will still be able to push sixteen raw pixels per clock but will be able to render 48 pixel operations per clock. So ATI will insist on more per pixels operations before you actually draw the pixel. This kind of makes sense when you need more maps and calculations per pixels.
I would fear the very successful and expensive "The way it's meant to be played" program which Nvidia spent some $160+ million dollars on last year. With that kind of budget it's much easier to convince developers and publishers that Nvidia's marchitecture is the best. We can surely say that 2006 will be just as full of marchitecture wars as 2005 was.
Mav's 2p - The R580 is the core used on the Xbox 360 now and I guess it's working rather nicely so I guess the Nvidia unit is the one we will see in the PS3 - I would suspect the reason the R520 was released at all was purely because ATI had a contractual obligation to the XBOX that stipulated that at time of release the GPU in the XBOX had to be better than the desktop unit.
As I am useless at how these things are calculated - can someone do the math here and tell me which will theoretically be faster.
One thing was certain in the last round - the core clock had to be higher on the ATI unit to compete with Nividia unit - now that wafer sizes will be closer (90nm Nvidia Vs 80nm ATI) and Nvidia clocks will be closer to the R580 than they were to the R520 (G71 clocks will probably be 650/1800 Versus 695/1500 R580) compare that to (G70 430/1200 Vs R520 625/1500) It possible that the ATI card wins on the technolgy front but Nvidia's raw power see it grunt it's way to the top. The power consumption of the ATI card seem to be quite high too, it's being rumoured that Nvidia has got em licked in that dept.
Either way - The battle will no doubt be CPU limited before the full potential of either units are realised.............. Enter the physix chip??
Mav
ATI BELIEVES that you need many more pixel operations per clock than the pixel itself. Nvidia disagrees. It believes in a totally different ratio. With the birth of the R580, ATI believes that you should have three to one ratio as R580 has 48 Shaders and 16 pixel pipelines. You don’t need much mathematics to figure out that ATI preaches a three to one ratio.
As the G71 will be able to process 32 pixels per clock, it will physically have 32 pipelines but will have just 16 ROPs. This is no surprise to us as Nvidia's G70, Geforce 7800 GTX actually has 24 pipelines and 16 ROPs.
This means that Nvidia and ATI will take completely new approaches, where Nvidia will try to convince everyone that you need more raw pixels than pixel operations.
The G70, Geforce 7800 GTX has 24 texture memory units and 16 ROPs and can theoretically draw eight pixels with each, having three textures per clock. G71 will be able to do even more pixels per clock - it's as simple as that.
ATI's R580, on the other hand, will still be able to push sixteen raw pixels per clock but will be able to render 48 pixel operations per clock. So ATI will insist on more per pixels operations before you actually draw the pixel. This kind of makes sense when you need more maps and calculations per pixels.
I would fear the very successful and expensive "The way it's meant to be played" program which Nvidia spent some $160+ million dollars on last year. With that kind of budget it's much easier to convince developers and publishers that Nvidia's marchitecture is the best. We can surely say that 2006 will be just as full of marchitecture wars as 2005 was.
Mav's 2p - The R580 is the core used on the Xbox 360 now and I guess it's working rather nicely so I guess the Nvidia unit is the one we will see in the PS3 - I would suspect the reason the R520 was released at all was purely because ATI had a contractual obligation to the XBOX that stipulated that at time of release the GPU in the XBOX had to be better than the desktop unit.
As I am useless at how these things are calculated - can someone do the math here and tell me which will theoretically be faster.
One thing was certain in the last round - the core clock had to be higher on the ATI unit to compete with Nividia unit - now that wafer sizes will be closer (90nm Nvidia Vs 80nm ATI) and Nvidia clocks will be closer to the R580 than they were to the R520 (G71 clocks will probably be 650/1800 Versus 695/1500 R580) compare that to (G70 430/1200 Vs R520 625/1500) It possible that the ATI card wins on the technolgy front but Nvidia's raw power see it grunt it's way to the top. The power consumption of the ATI card seem to be quite high too, it's being rumoured that Nvidia has got em licked in that dept.
Either way - The battle will no doubt be CPU limited before the full potential of either units are realised.............. Enter the physix chip??
Mav
.
