Nvidia losing influence due to Ps3 involvement?

[radar1200gs]

Democoder, what precision FP temps are you talking here? If thats 4 FP32 temps, then that translates to 8 FP16 temps.

 

[Fred da Roza]

But you can't get equivalent clocks if you have to have significantly more transisitors to support FP32 than FP24. It's probably the reason why ATI chose to design around FP24.

This is assuming that the transistors in the FP processing units are a large portion of the total number of transistors. I claim that they are not. What's more, how do you explain that the GeForce 6600 GT is, to date, the highest-clocked retail GPU available, with full support for FP32?

And around we go again

http://www.beyond3d.com/forum/viewtopic.php?p=220880&highlight=#220880

Then followed up by Dio

That assumes that ATI had all the information about the implications of supporting FP24 vs. FP32 beforehand. To put it bluntly, decisions aren't always (or even often) made because they are the best decisions to make.

I find it quite astonishing that after everything that has been written about FP24 vs FP32 on this board you would suggest that ATI might be going "Oh god, how could we have been so stupid as to do FP24 instead of FP32!"

I suggest you see sireric's post : http://www.beyond3d.com/forum/viewtopic.php?t=6223&postdays=0&postorder=asc&start=17

I think that shows that we think we made the right decision in hindsight. I certainly think we did because I cannot see the benefit to FP32 right now. Which isn't to say it won't be right in the future, but it just works right now.

 

[radar1200gs]

The only bit I never got with the NV30 launch was why nVidia insisted on bringing nV30 to market after R300 and the failed tapeout around that time. They should have gone straight to nV35 right then and there and saved a lot of embarrasment.

they had no choice, they had contracts to fill. And the n35 wasnt due till may but didnt hit any real amount untill july. Needed to be in the mix. Have websites like TH say the new king of graphix. and all the rest. Then they hide the flaws and lack of supply by cheating 3dmark. All to keep there name in the mix untill supply hits in august.

Sorry, had to address this.

You don't honestly believe that nVidia designed and built nV35 in between R300 launch and nV30 availability do you??? If you do, ask yourself, why didn't nVidia directly fix their biggest problem, register congestion?

nV35 was in development for quite a long while and was going to succeed nV30 no matter what, just not so quickly initially.

What I don't get is, with the lateness of nV30 (late enough that nV35 was almost completely ready at nV30 launch), and the trouble encountered manufacturing nV30 why didn't nVidia simply use the respin and fab time that that did on nV30 to accelerate completion of nV35 and launch it instead.

The contracts didn't matter that much, nV30 was late anyway and a better chip than nV30 would have been supplied.

One of life's little mysteries I guess.

 

[karlotta]

more OT, if nvidia didnt run with the nv30, they would have broken there contracts with Asus..Abit.ect.. and then lost them to ATI even earlier. As for the lame engineering, well i blame that on the loss of old time engineers who took the money and run. Never mind the poor leadership by the JH who was hell bent on being bigger than intell. And the bitter 3dfx employes who had to work for there arch rival. And for the same PR guys who drove 3dfx into the ground.Opps they still work there. It took some major personal changes to ATI to turn them around.

 

[radar1200gs]

Why do you keep going on about contracts? If nVidia had put the effort into nV35 they could probably have launched it about the same time you could buy nV30. The vendors would have nV35, which nVidia could simply label NV30 if they chose to...

 

[hovz]

radar its more complicated than that. they HAD certain criteria that HAD to be met. a card had to be launched, regardless of whether or not it was faster than the competition.