Fuad of The Inq, trying to explain NV40 ... AGAIN

[KimB]

Since the world was taken over by the machines. Oh, wait, or was it immortals? Bah.

 

[StealthHawk]

Since the world was taken over by the machines. Oh, wait, or was it immortals? Bah.

Immortals.

 

[KimB]

But there was only One when the world was taken over by machines, too...

 

[Ailuros]

O Leaf Munching Oracle, when will Series 5 come out in AGP or PEG form?

You mean "Leaf Smoking Oracle" don't ya 8)

Nope Oracles used to munch on leaves; as a greek I should know better.

 

[pakotlar]

We understand that the NV40 will actually only have eight physical pipelines, but these will appear act like 16 in certain games and, indeed, in 3DMark 2001 Nvidia was telling people how 3DMark01 is a very nice benchmark since they can render 16 textures per pass in it, using only eight pipelines. What Nvidia is using is the ability of the Vertex Shader model 3.0 (known as PS 3.0 or VS 3.0) where the Shader can actually render textures as a virtual pipeline but can only render them without filtering information.

Link: Nvidia's NV40 mimics a 16x1-pipeline architecture with just eight pipelines

Essentially, nVidia is now abusing a 3-years-old fillrate benchmark? :? :?

incurable, incredibly confused.

thats Funny.. Now where have i heard that one before... oh yeah..

Its exactlly what i have been saying the Nv40 is for oh.. a couple months now.. an 8 pixel engine that can occasionally do 16 pixels under the right conditions.

Will be interesting to see what Nvidias "hardware support" for PS 3.0 is as well...

Likely not what some are thinking.

why do you feel the need to brag? and when did you say this?

 

[pakotlar]

to quote myself from nVnews

[speculation]nV releases the nV40 first.
if nV40> 3 quad R420 then ATi releases the 4 quad version as the pro and the 3 quad as the np.
if nV40</= 3quad R420 then ATi release it as the pro and a 2quad version as the np[/speculation]

In either case clock speeds can still be tweaked to set performance goals in relation to the competitors card.


plus in the second case they would then have a revision for when nV releases the nV45 PCIE card

unless they were planning to delay the r420 by several months, case 1 is unlikely.

 

[Geeforcer]

why do you feel the need to brag? and when did you say this?

My guess is some time after the "R360 = 8x2" prediction.

 

[pakotlar]

why do you feel the need to brag? and when did you say this?

My guess is some time after the "R360 = 8x2" prediction.

hehe

 

[Tim Murray]

But there was only One when the world was taken over by machines, too...

Actually, there were six.

 

[digitalwanderer]

thats Funny.. Now where have i heard that one before... oh yeah..

Its exactlly what i have been saying the Nv40 is for oh.. a couple months now.. an 8 pixel engine that can occasionally do 16 pixels under the right conditions.

Will be interesting to see what Nvidias "hardware support" for PS 3.0 is as well...

Likely not what some are thinking.

why do you feel the need to brag? and when did you say this?

I don't know where he said it, but if he is correct on his "8 pixel engine that can occasionally do 16 pixels under the right conditions" I know a whole lot of geek rumor-mongers who will never live it down!

If that is the case, nVidia has been pushing the 16x1 like a mofo on all their regular insiders....and those insiders will be PISSED at being used for disinformation, again.

 

[PaulS]

I don't know where he said it, but if he is correct on his "8 pixel engine that can occasionally do 16 pixels under the right conditions" I know a whole lot of geek rumor-mongers who will never live it down!

If that is the case, nVidia has been pushing the 16x1 like a mofo on all their regular insiders....and those insiders will be PISSED at being used for disinformation, again.

The key point here is whether it's more accurate to describe it as a 16 pipe architecture which has negative conditions attached, or an 8 pipe architecture which has positive conditions attached. I'd suggest the former, but it really doesn't matter too much either way because it's just semantics.

The performance figures will speak for themselves, and the internal figures/arithmetic (not the publically released benchmarks) have received a lot of praise from some very knowledgable people - however you want to describe it, it will surely perform well (as will the ATi offering, of course).