R600 not going to make it on time...again?

[SnipingWaste]

I agree nVidia's recent cards have been poor in terms of bandwith, Im not sure if I said that before or not but I'll say it now.

What I meant by my previous comment is that all nVidia would potentially have to do is add more clusters to get to 512bit. Granted that would be a hell of alot of chips but then again they could use both sides of the card which should be enough.

It'll take a new core and PCB but it will not take much at all for the "G80" type core to match/beat the R600. The R600 on the other hand would be much harder to redesign to tackle an upgraded G80 since their design isnt nearly as modular as nVidia's

Nvidia will have to come up with a new memory controler and want be able to just add on to it. If you look at the G80 its realy 2 memory controlers 1 256bit and 1 128bit controler. The R600 looks to be 1 512bit memory controler and with ringbus so reads to the memory will be like a 1024bit bus. There is no way G80s memory controler can do that even with upgrads. They will have to come up with a new memory controler and I beleave they will not do that (They like to re use memory controlers for many GPU rev's). They spent a lot of time and money and can't beleave they well throw it away so soon. Look at the NV30 and NV35. It took time to redo the memory controler to 256bit form 128bit and it still was not much of an improvment. If they do redo the memory controler then look for at best 2% to 5% improvment only.


Another thing is when DX10 comes out. DX10 will get rid of many of the bottle necks in the DX9.X API and that will leave the bandwidth being the next bottle neck in the DX10 API and that is wear the R600 "should" shine with its huge bandwidth.