NV40 64 bit?

[Uttar]

BTW, Maximum PC is full of BS GPU-wise:
1) In their NV30 issue, they didn't even KNOW if it was a 128-bit and 256-bit memory bus, and said it was amazing as it COULD be 32GB/s :rolleyes:
2) One of their GPU explanation in a small "extreme performance(?)" part of the magazine had more errors and mistakes than words!


Uttar

[nutball]

Originally posted by Nv40
SPeed is not everything ,there are things like ImageQuality that are very popular in reviews today ,that gamers really care.. the days of only fast fps has ended.. ;)

Speed and IQ are two sides of the same coin. It was possible to have gorgeous IQ in games five years ago, just not at interactive frame-rates.



so ,you if you dont give developers what they need ->now (not in the future,but right now!!!) for what they want to do in their -today- development ,and wait 2-3 years ,because "its not neccessary today", you will be delaying the creativity of those developers and waiting Twice the time ,that is 2-3 years to finally give the requested features to developers PLUS! 2-3 MORE YEARS to see it in their retail games. game you play today are as advanced in graphics and features as the "not neccesary/wasted features" made 2-3 years ago. not really sure if you are aware of this but - hardware need to comes first- and later comes the games using the new tech. is not the other way.

So what you're telling me is that in order for a few hundred game developers to do their work today, NVIDIA and ATI need to productise a full 64-bit-per-component FP hardware and sell millions of them to the general public, today? I don't think so somehow.

Ahhh, nuts, I just typed several paragraphs in response, but I can't be bothered.

It's perfectly possible to code to hardware specifications you know. That's the whole point of things like DX10, they lay down common ground rules for the hardware and software guys to follow and meet in the middle.

[Razor04]

Originally posted by sxotty
I never said they would have met the requirements, you simply said that fp32 was just for directx9 end of story. Similarly here you are going off on a company calling it/them spoiled brats. It is really rather amusing. I was saying that you had an overly simplistic view, and you are merely confirming it.

You are correct they would not have met the specs I think we can all understand this widely known fact. One fact does not, however make all the other information and reasons surrounding something irrelevant.

What you said last about fp16->64bit and fp32->128 I realize and if you were implying that perhaps maximum pc said is that they will have 64bit i.e. fp16 that may well be the case. It is likely that they asked Nvida whether fp16 would still be there, and nvidia said something like yeah we will have 64bit precision in the next chip... could easily have confused someone.
Can't we all just be friends? :)

No point in arguing over it...I just want to see them get back on track and the competition between the two big brands to get even closer so prices go lower. I think everyone here agrees on this aspect. Although it wouldn't hurt for NV to work on their "optimizations" some... i.e removing them

[sxotty]

Yes friends ;)

I dont mind optimizations, but they should have a little check boxes

"to cheat" "or not to cheat"

Well perhaps different titles...

[nobie]

I think the optimizations they do in the new ForceWare drivers are pretty good (ie they do not trash image quality). IMHO I'll take an extra 10 FPS if it makes no noticeable difference in IQ. AA quality is still not so good though, hopefully that will improve with NV40. AF could be improved too, specifically with regards to preventing moire patterns from appearing.

[Razor04]

Originally posted by nobie
I think the optimizations they do in the new ForceWare drivers are pretty good (ie they do not trash image quality). IMHO I'll take an extra 10 FPS if it makes no noticeable difference in IQ. AA quality is still not so good though, hopefully that will improve with NV40. AF could be improved too, specifically with regards to preventing moire patterns from appearing.
Ah but the big problem with that is they are "optimizing" mainly or only in the benchmark portions of applications inflating their scores and making their products appear to do better than they actually do. That is just low imho...but I am really sick and tired of going over this so many times and there are countless other threads that deal with these issues so I am going to take my leave of this thread.

[nForceMan]

Originally posted by Razor04
Ah but the big problem with that is they are "optimizing" mainly or only in the benchmark portions of applications inflating their scores and making their products appear to do better than they actually do. That is just low imho...but I am really sick and tired of going over this so many times and there are countless other threads that deal with these issues so I am going to take my leave of this thread.

"mainly or only"? You should prove it. ( It'll be hard. :D )
And don't forget this one:

With ATI, we can see that the card doesn't do either anisotropic or trilinear filtering when looking at any texture stage other than 0.
Source (http://www.anandtech.com/video/showdoc.html?i=1931&p=6)

Back to topic.
64bit color channels will be important for cinematic computing. NVIDIA's vision has been aiming at this for a long time.

[Razor04]

Originally posted by nForceMan
"mainly or only"? You should prove it. ( It'll be hard. :D )
And don't forget this one:

Source (http://www.anandtech.com/video/showdoc.html?i=1931&p=6)

Back to topic.
64bit color channels will be important for cinematic computing. NVIDIA's vision has been aiming at this for a long time.
*sigh* I wasn't going to bother with this anymore but you (nForceMan) always seem to pull BS out of any post. First of all I said mainly or only...that connotates that it would depend on the application. Take 3DMark03...that would be an only case...then look at UT2003 that would be a mainly case as in addition they are optimizing for the game as a whole. And yes it has been proven that their benchmark scores are higher in UT2003 typically than what is seen in games...why do you think so many places now use custom paths and FRAPS? I can't believe you posted that AnandTech article...I guess you didn't read any of the threads on here or B3D or numerous other sites basically saying that the author of the article was pulling stuff out of you know where. One specific example is with the texture stage filtering that you mentioned...where did he get that conclusion from? NV...did he bother to ask ATI for clarification on anything in that article? No... I guess the ATI driver and hardware developers over at B3D know nothing about their own hardware when they say that article has a lot of errors including the thing you brought up. nForceMan...it is always a good idea to do research on your own...don't trust the big sites just cause they are big...cause more often they are wrong or biased as they are trying to please too many people at once.

btw...how many more times are we going to have to go over 64 bit = FP16 in this thread? And also NV's "vision" of cinematic computing just came about with the NV3X and is more marketing than anything else.

[nForceMan]

Originally posted by Razor04
Take 3DMark03...that would be an only case...then look at UT2003 that would be a mainly case as in addition they are optimizing for the game as a whole. And yes it has been proven that their benchmark scores are higher in UT2003 typically than what is seen in games...

(omg) That would be your proof? Pffft. You wouldn't be a successful mathematician.

I can't believe you posted that AnandTech article...I guess you didn't read any of the threads on here...
Nice. You have just shot yourself in the foot. I did start a thread here about that.

where did he get that conclusion from? NV...did he bother to ask ATI for clarification on anything in that article? No...
Are you sure? Can you prove it? (If you can, please don't do it so idiotic than did it above) Or just trying to present an affair in such a way as a fact?

don't trust the big sites just cause they are big...
I said "don't forget". This doesn't mean I trust/don't trust.

btw...how many more times are we going to have to go over 64 bit = FP16 in this thread?
Really? You'll see (sooner or later)! ;)

And also NV's "vision" of cinematic computing just came about with the NV3X and is more marketing than anything else.
This is your opinion. Btw it seems the life is so simple for you. Your basic rules:
pro NV statement = :bs:
pro ATI statement = :super: :dance: :headbang: :thumbsup: (salute)
I think this can be one of the definition of 'fanboi'.
I'm awfully sorry, but the things are a lot more complicated than that.

[StealthHawk]

Everybody should know by now that when AF is selected in the control panel ATI does full degree trilinear AF on texture stage 0 and full degree bilinear AF on the other texture stages.

When AF is set to application it can do full degree trilinear AF on all texture stages.

In fact, NVIDIA is doing something similar. For NVIDIA quality see here (http://www.nvnews.net/vbulletin/showthread.php?s=&threadid=19604).

[nobie]

Originally posted by Razor04
Ah but the big problem with that is they are "optimizing" mainly or only in the benchmark portions of applications inflating their scores and making their products appear to do better than they actually do. That is just low imho...

I can't argue with that, optimizations that artificially inflate benchmark scores don't do anyone any good (well, except maybe Nvidia :rolleyes: ).

[nvnews-reader]

Why are they keeping FX16 around?

[Paul]

Originally posted by nvnews-reader
Why are they keeping FX16 around?

For PS1.4 speed.

[oqvist]

Why would nvidia opt for 64 bit???

I mean seeing how the 5900:s suffers in 32-bit? Why not at least offering a 24-bit like the Radeons for best compromize? I would bet the NV40 is slower in 64-bit than the NV35 ever was in 32-bit. But maybe it´s just a marketing ploy as the 8x agp on ti-4200:s video cards.