View Full Version : Interview with NVIDIA's Tony Tamasi

03-03-03, 05:16 PM
pretty technical... but nvidias tony tamasi tried to answer some of the questions brought up during the beyond3d preview...


03-03-03, 08:19 PM
Correct Me if I missed something, but it looked as though He wasn't addressing what I think is the real problem, Memory Bandwidth. I had no doubt's that the GPU was a good thing.
I also didn't expect much (20-30%) more speed from it over
the 9700PRO. But at this juncture, Why do they bother with the
NV30 and not just get the NV35 out.

03-03-03, 08:31 PM
Oh no, you're quite correct. He was speaking just as technically as he possibly could just so we couldn't understand what he WASN'T talking about. :(

I thought B3D did a good job on the review considering nVidia apparently dogged 'em pretty hard for the conditions and really showed the strengths & weakness of the card...and I thought the nVidia dude just danced around the question. :(

03-03-03, 10:15 PM
Testing with the DirectX _PP precision hint enabled and disabled under the 42.68 drivers show no difference in performance. It was quoted that FP16 is able to run at twice the rate of FP32 - considering there is no difference in speed this would either suggest that the current drivers are forcing FP16, thus meaning that DirectX developers and applications will not be able to benefit from the increased precision afforded by FP32, or that the drivers are currently unable to effectively schedule FP16 processing. Which is it?

The applications choose the precision level of precision and may or may not use it as efficiently as possible.

And he's comparing the NV30 to the R8500...there's a worthy comparison....NV30's implementation is different from Radeon 8500 because it was designed by NVIDIA, with a different design philosophy. No two 3D architectures are exactly the same between different vendors.

03-04-03, 01:44 PM
Here's my favorite part:

B3D's Question:
Would your drivers fail WHQL certification...?

[QUOTE]Tamasi's Answer:
Of course, drivers would fail WHQL if they fail to meet minimum WHQL requirements...

So would the drivers pass WHQL or not? :confused:

John Reynolds
03-04-03, 01:55 PM
I particularly liked the attack on ATi's 24fp when they've been pushing MS to lower DX9 compliancy from 24 to 16fp.

03-04-03, 09:11 PM
Originally posted by John Reynolds
I particularly liked the attack on ATi's 24fp when they've been pushing MS to lower DX9 compliancy from 24 to 16fp.