PDA

View Full Version : More woes (new ones!) for nVidia over the FX?


Pages : [1] 2

digitalwanderer
05-28-03, 12:51 PM
There is an interesting thread (http://www.beyond3d.com/forum/viewtopic.php?t=6129) over at Beyond3d concerning some intriguing new driver findings....

I don't understand all of it and it's still developing, but I figured you folks might like to know about it. Credit goes to Hellbinder for posting it at Rage3D (http://www.rage3d.com/board/showthread.php?s=&threadid=33688771) for the find.

Richthofen
05-28-03, 12:56 PM
to be honest i don't care.
My games run great and that is all i have to know.

Skynet
05-28-03, 01:02 PM
to be honest i don't care.
My games run great and that is all i have to know.
head in the sand?

Personally I think this will be HUGE. Nvidia you better come clean now, not later or much later. Does Nvidia really feel the FX series is THAT inferior to have to hack away to make it perform the way they want? Sure makes a person wonder...

silence
05-28-03, 01:13 PM
Originally posted by Richthofen
to be honest i don't care.
My games run great and that is all i have to know.


basically i agree.....if card needs those shaders to run better,ok..why not?

but if that's something that hurts my gaming...i dun wanna it.....

so i think i'll start looking at B3D for a bit.....i would really like to know what's going on....

Sazar
05-28-03, 01:18 PM
is this what uttar has been hinting @ for a while ?

Grrrpoop
05-28-03, 01:18 PM
Originally posted by Skynet
head in the sand?

Personally I think this will be HUGE. Nvidia you better come clean now, not later or much later. Does Nvidia really feel the FX series is THAT inferior to have to hack away to make it perform the way they want? Sure makes a person wonder...
While I agree this warrants attention and could be a cause for concern, I don't think it's safe to assume that there's anything nefarious about these new discoveries.

Admittedly, with their recent track record it doesn't look good hehe :D

Oops, just noticed what Silence said, which is basically what I agree with :)

digitalwanderer
05-28-03, 01:23 PM
Originally posted by Sazar
is this what uttar has been hinting @ for a while ?

I don't know for sure...but he did post up the entire text of the shader files and I can't make heads or tails of 'em. :(

To those who enjoy their games, I'm very glad for ya and hope you continue to...but that is NOT the issue here. :rolleyes:

Uttar
05-28-03, 02:01 PM
Nah, nothing to do with that.
I was atually first hinting at the GeForce FX 5000 beign a renamed GeForce 4 MX AGP 8X. Still no news on that though. It originally came from a generally reliable source ( and in a good position to know that type of things, too ) - but he might be wrong.

Anyway, there have been other things I've been hinting to. Nothing major though. Really, you shouldn't think I'm trying to hide some type of top secret info :)


Uttar

dohcmark8
05-28-03, 02:23 PM
Big deal.....

A bunch of idiots who know nothing about writing drivers are reading the source code for Detonators, i see nothing coming out of this except... "Looky Uttar it says, MOVL1 0111 Path Fast here" and then another bunch of idiots who have not idea why its there and what its for going OMG theyre cheating, in other words i see no good coming out of this.

digitalwanderer: Your ATi bias is extremely obvious its not even funny, why dont you go to Rage3d, unless you know something we dont? :confused:

Uttar
05-28-03, 02:26 PM
Originally posted by dohcmark8
A bunch of idiots who know nothing about writing drivers are reading the source code for Detonators, i see nothing coming out of this except... "Looky Uttar it says, MOVL1 0111 Path Fast here" and then another bunch of idiots who have not idea why its there and what its for going OMG theyre cheating, in other words i see no good coming out of this.

Frankly, I don't believe they're cheating with those things. As shown by a few members, all of those are pretty much obviously innocent ( standard work, nothing which would help identify a shader ) - the first one seems rather fishy though, but it could also be innocent.

I do find it strange that nVidia is hardcoding this in their drivers however, and figuring something out of it would be nice.

Also, I believe you're confusing the B3D community with another one. It's a lot better than you think.


Uttar

dohcmark8
05-28-03, 02:30 PM
Originally posted by Uttar
Frankly, I don't believe they're cheating with those things. As shown by a few members, all of those are pretty much obviously innocent ( standard work, nothing which would help identify a shader ) - the first one seems rather fishy though, but it could also be innocent.

I do find it strange that nVidia is hardcoding this in their drivers however, and figuring something out of it would be nice.

Also, I believe you're confusing the B3D community with another one. It's a lot better than you think.


Uttar

No, i was just refering to all the fanATics on this board who will have a field day with this while not having a clue what they are reading, i visit the B3D forums often and i know most of the members there are extremely knowledgeable on the subject.

Why dont they rip apart ATi's CATALYST drivers too?

digitalwanderer
05-28-03, 02:34 PM
Originally posted by dohcmark8
digitalwanderer: Your ATi bias is extremely obvious its not even funny, why dont you go to Rage3d, unless you know something we dont? :confused:

Oh I go there too, don't worry. The only thing I seem to know that you don't is about nVidia's underhanded business practices, but I'm trying to share me knowledge. :)

Originally posted by dohcmark8
i visit the B3D forums often and i know most of the members there are extremely knowledgeable on the subject.

Why dont they rip apart ATi's CATALYST drivers too?

Because the Catalyst drivers are good, and if they have a question about them ATi is more than happy to respond and clarify....unlike a certain evil/bad company starting with an "nV".... :p

Gar
05-28-03, 02:35 PM
Originally posted by dohcmark8
.

Why dont they rip apart ATi's CATALYST drivers too?

I believe that is also being done. THis could turn out to be nothing, but it could also turn into something just as big when people in this forum started taking screen shots of 3dmark03 and pointing out anomalies with the sky and water in nature test.

SurfMonkey
05-28-03, 04:01 PM
Oh, they'll be next on my list ;)

nin_fragile14
05-28-03, 04:08 PM
Originally posted by dohcmark8

digitalwanderer: Your ATi bias is extremely obvious its not even funny, why dont you go to Rage3d, unless you know something we dont? :confused:

He is biased now, but three weeks ago, he was pretty pro-nvidia. His dislike is definitely understandable considering the events of the past week--not just because of their cheating, but because of personal issues he had with them as well.

Hanners
05-29-03, 06:52 AM
Originally posted by dohcmark8
A bunch of idiots who know nothing about writing drivers...

When I read this far into your post, I actually thought you were talking about nVidia! :lol:

;)

Behemoth
05-29-03, 07:14 AM
yeah it would be very interesting if they study how ati shuffle the instructions as well.

jbirney
05-29-03, 07:51 AM
Originally posted by Behemoth
yeah it would be very interesting if they study how ati shuffle the instructions as well.

Since both Tim Sweeny and John Carmack say that suffeling instructiongs are "legal" optimizations..whats the point of that?

Behemoth
05-29-03, 08:16 AM
Originally posted by jbirney
Since both Tim Sweeny and John Carmack say that suffeling instructiongs are "legal" optimizations..whats the point of that?
the point is to prove if they are really doing what they are saying

Sazar
05-29-03, 08:20 AM
Originally posted by Behemoth
the point is to prove if they are really doing what they are saying

considering visual data.. it would appear to be the case don't you think ?

Behemoth
05-29-03, 08:26 AM
Originally posted by Sazar
considering visual data.. it would appear to be the case don't you think ?
i believe there are good reaons why futuremark is investigating further into this matter.

digitalwanderer
05-29-03, 08:37 AM
Originally posted by Behemoth
i believe there are good reaons why futuremark is investigating further into this matter.

"The truth will come out in the fullness of time."

:)

Behemoth
05-29-03, 08:38 AM
Originally posted by Sazar
considering visual data.. it would appear to be the case don't you think ?
by the way, rewrite shader is more likely to be the case, i think, but they are saying "shuffle instructions", although rewriting shaders for developers is generally acceptable if the visual outputs are the same, as JC put it.

Behemoth
05-29-03, 08:39 AM
Originally posted by digitalwanderer
"The truth will come out in the fullness of time."

:)
yeah at least i will have no doubt in their integrity when i jump ship :)

dohcmark8
05-29-03, 04:41 PM
But i can kinda understand digis enthusiasm, i mean if Trident gave me some free XP4's i would be all over them and would post evry negative thing i see about ATi or nVIDIA all over the net and praise Trident. :nana: :headbang: