Go Back   nV News Forums > Hardware Forums > Benchmarking And Overclocking

Newegg Daily Deals

Reply
 
Thread Tools
Old 05-19-03, 01:46 AM   #145
digitalwanderer
 
digitalwanderer's Avatar
 
Join Date: Jul 2002
Location: Highland, IN USA
Posts: 4,944
Default

Quote:
Originally posted by nVidi0t
However, this is a wild guess but it might be a side effect of other cheats...
Hey, I had that thought too over in a thread at R3D earlier too...the whole thought not just the conspiracy minded bit.

It seems like a bug, but it might be a bug that's there because of a cheat....I'm thinking rendering order for some weird reason will end up being the culprit, but that's just a wild-arsed guess.

Quote:
Originally posted by StealthHawk
Wrong. There are visible side effects present from this "optimizing."

Discussion

They also lowered the IQ of Quality AF.

Discussion



As noted above, there are visible side effects.

This totally discounts the ethical reasons of why not rendering everything in a benchmark is cheating.

.......


So what are you saying, that they wouldn't have included these optimizations if they knew that you could go off the rail? If that doesn't mean it isn't a deceptive attempt at a cheat, I don't know what is. Does that fact that some people can go off the rail make it any more or less bad?



That's great, too bad Kyle didn't say that in this article. The ExtremeTech article mentions that they talked to nvidia from the get go. And B3D did too. nvidia hasn't replied to B3D yet.

Hmm, if it's not a cheat why is it taking them so long to come up with an answer, or an official reply?



The next question is of course, how do you know they aren't cheating in some other timedemos or benchmarks? And how do you know that the speed increaes found in benchmarks translate to better gameplay performance?

Serious Sam? 3dmark2001? Hmm.
I LOVE THIS REPLY!!!!!

Thanks for the links to the visual differences, I wasn't aware of that. As for the rest....need I say I agree?



Quote:
Did anyone ever say they were cheating in Doom3?
Uhm...NO! But I think he might be refering to how I implied, (well, stomped about and raved a bit about... ), that nVidia and Johnny-boy Carmack pulled a sort of unfair and cheesey little nVidia love-in benchmark fest without happening to mention it to ATi. (They found out about it when we did?!?! For shame Mr. Carmack, for shame. )

But I never said they cheated at it, just that it wasn't a fair comparison of the two cards actual performance in the real game...at least not to the 9800 Pro and nVidia sort of not only strutted that bit of news as their new crown but orchestrated the whole thing.

Just thought I'd clear that up for ya, let's finish up this most excellent reply of yours.

Quote:
What they are doing does not help games. It only works in fixed benchmarks where the camera angle doesn't change. Do you not understand this? This "optimization" will not help in real games. This is another reason why it shouldn't be called anything but cheating.
Thank you, that sums it all up very nicely.
__________________
[SIZE=1][I]"It was very important to us that NVIDIA did not know exactly where to aim. As a result they seem to have over-engineered in some aspects creating a power-hungry monster which is going to be very expensive for them to manufacture. We have a beautifully balanced piece of hardware that beats them on pure performance, cost, scalability, future mobile relevance, etc. That's all because they didn't know what to aim at."
-R.Huddy[/I] [/SIZE]
digitalwanderer is offline   Reply With Quote
Old 05-19-03, 02:28 AM   #146
Sazar
Sayonara !!!
 
Sazar's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 9,297
Default

Quote:
Originally posted by Nv40
StealthHawk ,are you reading this tread ?

because you are saying things that have been answered ,so Many pages
ago ,since the begining.. so please go back again ,and reread everything..

Nvidia can claim whatever they want ,and they will be right..
SINCE THEY DONT AGREE WITH THE RULES OF THE COMPETITION ,
THEY DONT HAVE TO PLAY BY ITS RULES. and SINCE THEY ALREADY WARNS EVERYONE THAT THE PERFORMANCE IS NOT REPRESENTATIVE OF NVIDIA CARDS .

So THey can Optimize the way they want.

Nvidia says DONT DO IT , IT IS NOT REAL , but people do it anyway ,
them later the same people blame them ....

what can they do ? an ADvertisment in MTV ? Maybe CNN ?
"the performance is not representative of our cards..
use it at your risk ,for entertaiment purposes."

it is very clear to see that People are not interested anymore if the
contest is valid or not ,they only want to see the Scores...
no matter what

The Performance is not accurate ,because both cards do things in diferent
ways , REnders things in diferent ways ,with cheats or wihtout cheats ,with optimizations or without it. with fixed drivers or wihout it.

it is pointless so many repetition... seem you can only see things
in two ways.. BLACK or WHITE.. if that is not Black then it should be
WHite ... .. there are more colors , you know .
I don't know if its the english or lack of something else but... um.. WHAT >

didn't understand a word there cept for CNN... I Think...
Sazar is offline   Reply With Quote
Old 05-19-03, 03:52 AM   #147
zakelwe
Registered User
 
Join Date: Dec 2002
Posts: 768
Default

I ran some benchmarks on various drivers from 40.72 upwards. i have a GF4 so cannot run GT4 but the problem is apparent in GT2 which my card can run.

I found that the speed for GT3 was constant for the later drivers but the speed for GT2 rose considerably. However, it did not peak at 43.51 or 44.03 but peaked at 42.68 and then stayed constant until 44.03 where it dropped, maybe because these drivers are not optimised for GF4 cards ?

I ran nature in 3dmark2001SE as a DX8 baseline and that showed that the scores stayed more or less the same apart from two earlier drivers and the 44.03 again.

So Gt2 does go up, but there is no performance gain from 42.68 onwards and 42.68 have been judged OK by Futuremark then ther is no problem for GF 4 owners, ie nvidia have not gained anything for them .

We need someone to test GT4 with a FX5800 or 5900 with 42.68 and 44.03 drivers to see if there is any difference.

If there isn't then there is no extra performance gained from this bug/cheat, unless the 42,68's also show this problem as well.

Regards

Andy
zakelwe is offline   Reply With Quote
Old 05-19-03, 03:54 AM   #148
Filburt
Jeimei Frugiunglagią
 
Join Date: Jan 2003
Posts: 105
Send a message via ICQ to Filburt Send a message via AIM to Filburt
Default

Quote:
Originally posted by jimmyjames123
There is no indication that the majority of the data is cut out. There is simply no way to say how much of a boost is given.



Since when is there a handbook on the intent of a benchmark? We already know that both NVIDIA and ATI both optimize for this benchmark, and we know that ATI has access to the developer's version of this benchmark. And obviously Futuremark's intent is not the same as NVIDIA or ATI. NVIDIA (and ATI) wants the benchmark to be as fast and as smooth as possible on their cards without corrupting image quality that we can see.



The scores one receives in a benchmark are never perfectly congruous to actual gaming performance. That was the major beef with 3dmark03 in the first place. The GeForce FX 5900 Ultra has had it's performance well documented by several reviewers, and for the most part extremely well in virtually all benchmarks and games even compared to everything else that is out there.
Are you being facetious?

The point of a benchmark is to give some objective means of comparison between different pieces of hardware. To run the benchmark correctly, all products must pass the entirety of data through themselves, and render all scenes to the same specifications. To cut out a major chunk of the data passed is to in effect not run a benchmark in the same manner as it is run on other cards and to therefore provide a score not comparable to scores obtained on other cards due to their respective scores meaning very different things. The *reason* benchmarks have a "rail" of sorts is to provide a control factor in the exercise as to ensure the different pieces of hardware are rendering the data and thus the scores are reasonably comparable.

FX5900 score = how quickly it renders when only the data of the "visible" scene is all that is passed to the actual hardware.

9800 score = how quickly it renders the scene when the whole of the data is passed to the actual hardware.

Again, to apply static clip planes is to run what is effectively a radically different benchmark than what the other cards are running. Doing so, and then providing a score that is presented as being produced in the same manner as the scores of other products, I would certainly term as being dishonest.

Do you understand, now? Given that the 9800 is rendering the scene in accordance with the intent of a benchmarking program, which is to pass data in its entirety to the hardware. Optimizations applied should *ONLY* be those that are transferrable to non-deterministic flight paths since the purpose of the benchmark is to gauge performance that can be reflected to a reasonable degree in other tasks. Applying static clip planes is to optimize in such a fashion that reflects nothing of how the hardware will handle a non-deterministic scenario and is therefore cheating the benchmark because it has destroyed that which the score is intended to reflect.

Making an argument that the image quality is not affected is not only a bogus rebuttal, it addresses nothing of what a benchmark is designed for. Benchmarks are means of objective comparison between multiple products, keeping image quality is only important should it fall out of spec. Saying it *looks* the same as the 9800 and therefore it isn't doing some sort of cheating only means it isn't cheating in degrading the perceivable image quality from spec, but it in no way is a means of defense against the criticism I'd made in the preceding paragraph.

Thus, for you to make such a defense, you are either actively choosing to ignore the above, or you do not understand that which you are talking about. Either way, the defense does not hold water.
Filburt is offline   Reply With Quote
Old 05-19-03, 03:56 AM   #149
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

my thought on optimization is, if it only works in a predefined way, its a cheat.

if nvidia told you 3DMark represented its card performance, you were cheated.
Behemoth is offline   Reply With Quote
Old 05-19-03, 04:00 AM   #150
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

Quote:
Originally posted by zakelwe
I ran some benchmarks on various drivers from 40.72 upwards. i have a GF4 so cannot run GT4 but the problem is apparent in GT2 which my card can run.

I found that the speed for GT3 was constant for the later drivers but the speed for GT2 rose considerably. However, it did not peak at 43.51 or 44.03 but peaked at 42.68 and then stayed constant until 44.03 where it dropped, maybe because these drivers are not optimised for GF4 cards ?

I ran nature in 3dmark2001SE as a DX8 baseline and that showed that the scores stayed more or less the same apart from two earlier drivers and the 44.03 again.

So Gt2 does go up, but there is no performance gain from 42.68 onwards and 42.68 have been judged OK by Futuremark then ther is no problem for GF 4 owners, ie nvidia have not gained anything for them .

We need someone to test GT4 with a FX5800 or 5900 with 42.68 and 44.03 drivers to see if there is any difference.

If there isn't then there is no extra performance gained from this bug/cheat, unless the 42,68's also show this problem as well.

Regards

Andy
i will test the GT4 with FX5800
Behemoth is offline   Reply With Quote
Old 05-19-03, 04:27 AM   #151
zakelwe
Registered User
 
Join Date: Dec 2002
Posts: 768
Default

Quote:
Originally posted by Behemoth
i will test the GT4 with FX5800
Thanks !

Can you test it against 42.68 to see what improvement there is ? Maybe do an average of three runs ?

If there is none then someone needs to test 42.68 to see if this off rail clipping occurs in that also so that means the beta testers.

I thought 42.68 just didn't show some of the bullets and explosions in GT1 !

Regards

Andy
zakelwe is offline   Reply With Quote
Old 05-19-03, 05:24 AM   #152
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

Quote:
Originally posted by zakelwe
Thanks !

Can you test it against 42.68 to see what improvement there is ? Maybe do an average of three runs ?

If there is none then someone needs to test 42.68 to see if this off rail clipping occurs in that also so that means the beta testers.

I thought 42.68 just didn't show some of the bullets and explosions in GT1 !

Regards

Andy
since my scores were very consistent, i only did one run on 44.03
not sure about the bullets, didnt focus at them.
here are results

42.68
--------
3DMarks 3683 3697 3692
GT1 92.4 94.2 94.1
GT2 25.7 25.9 25.7
GT3 22.0 22.0 21.9
GT4 26.4 26.3 26.4

44.03
--------
3DMarks 4051
GT1 96.8
GT2 28.9
GT3 24.2
GT4 29.4

enjoy!
Behemoth is offline   Reply With Quote

Old 05-19-03, 05:29 AM   #153
StealthHawk
Guest
 
Posts: n/a
Default

Quote:
Originally posted by Nv40
StealthHawk ,are you reading this tread ?

because you are saying things that have been answered ,so Many pages
ago ,since the begining.. so please go back again ,and reread everything..
Yes, I read the whole ****ing thread...much to my chagrin. Basically I had no choice since this is a popular topic, and I want to make sure things don't get out of hand.

As to your accusations, obviously they have not been sufficiently answered, because arguments are still being made, based on the untrue premises which were originally stated. Also, jimmy either ignores or does not see posts where parts of his argument are debunked, and continues saying things like "IQ is not affected, so its ok."

Since much of the debate stems from what was said in the beginning, I think returning to the roots of the problem was necessary.
  Reply With Quote
Old 05-19-03, 05:56 AM   #154
zakelwe
Registered User
 
Join Date: Dec 2002
Posts: 768
Default

Quote:
Originally posted by Behemoth
since my scores were very consistent, i only did one run on 44.03
not sure about the bullets, didnt focus at them.
here are results

42.68
--------
3DMarks 3683 3697 3692
GT1 92.4 94.2 94.1
GT2 25.7 25.9 25.7
GT3 22.0 22.0 21.9
GT4 26.4 26.3 26.4

44.03
--------
3DMarks 4051
GT1 96.8
GT2 28.9
GT3 24.2
GT4 29.4

enjoy!
Thanks for doing those.

Compared to the GF4 GT2 and GT3 increased from 42.68, for the Gf4 they did not with my results.

The increase was :-

GT1 => 3%
GT2 => 12%
GT3 => 10%
GT4 => 10%


So it looks as if they managed to optimise all games from the earlier drivers by about the same amount, apart from GT1, which is as I said earlier was slammed for not displaying correctly so bumping up the scores in 42.68, so they only got 3% for that.

If there is a problem with 44.03 in GT2 and GT4 it is not significantly upping the scores from GT3's increase, if these figures
are correct.


Regards

Andy
zakelwe is offline   Reply With Quote
Old 05-19-03, 06:04 AM   #155
DaveBaumann
Registered User
 
Join Date: Jan 2003
Posts: 98
Default Re: My thoughts on "Optimization"

James, let me point a fe things out here.

Quote:
Originally posted by jimmyjames123
[b]For one, I think the journalism at Extremetech and Beyond3d was a bit odd. Never do they mention the following relevant points:

1) Image quality was NOT compromised with the new FX 44.03 driver. In fact, most people have noted improvements in Image Quality, even in 3dmark03.
First of, we are reporting that we have found similar rendering issues to ExtremeTech, but this is to be expected since we are one of the few media outlets that have these tools available to use (the other being C|Net, and they are not quite as active in the same way). We have not drane any conclusions as to what they are and we have gone back to NVIDIA to ask what the issue was.

As for quality of the 44.03, our post on 3DMark03 issues does not relate to 44.03 as we hadn't tested them at that time - we were using both the 43.45's and 44.51's and saw these issues. The IQ improvements are spearate as far as we are concerned since we were not using those drivers and are outside the scope of this issue.

Quote:
3) [H]OCP talked with ATI privately about the quake/quack driver cheat issue for more than a month before writing their article. ATI repeatedly denied cheating, and [H]OCP released their article after they proved them wrong (showing that image quality was compromised on Quake 3 while enhancing performance).
We informed NVIDIA on the 8'th of May, breaking our Futuremark NDA in the process by providing screenshots from the developer build to give them an accurate deptiction of what is occuring.
DaveBaumann is offline   Reply With Quote
Old 05-19-03, 06:40 AM   #156
ChrisRay
Registered User
 
ChrisRay's Avatar
 
Join Date: Mar 2003
Location: Tulsa
Posts: 5,101
Default

Quote:
We informed NVIDIA on the 8'th of May, breaking our Futuremark NDA in the process by providing screenshots from the developer build to give them an accurate deptiction of what is occuring
And you're proud of this? I'm normally a pretty big fan of your articles. But breaking you're NDA is a pretty stupid thing to do
__________________
|CPU: Intel I7 Lynnfield @ 3.0 Ghz|Mobo:Asus P7P55 WS Supercomputer |Memory:8 Gigs DDR3 1333|Video:Geforce GTX 295 Quad SLI|Monitor:Samsung Syncmaster 1680x1080 3D Vision\/Olevia 27 Inch Widescreen HDTV 1920x1080

|CPU: AMD Phenom 9600 Black Edition @ 2.5 Ghz|Mobo:Asus M3n HT Deluxe Nforce 780A|Memory: 4 gigs DDR2 800| Video: Geforce GTX 280x2 SLI

Nzone
SLI Forum Administrator

NVIDIA User Group Members receive free software and/or hardware from NVIDIA from time to time to facilitate the evaluation of NVIDIA products. However, the opinions expressed are solely those of the members

Last edited by ChrisRay; 05-19-03 at 06:45 AM.
ChrisRay is offline   Reply With Quote
Reply


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
Bored, impressed, and giddy: Our final thoughts on E3 2012 (with photos) News Archived News Items 0 06-13-12 07:00 AM
Thoughts from console owners on NVIDIA's GEFORCE GRID MikeC Console World 11 05-27-12 09:43 AM
Looking for a good 21"/22" Monitor...any thoughts? Guuts General Hardware 13 09-22-02 12:04 PM
Thoughts on the command line as an interface. lunix Microsoft Windows XP And Vista 10 09-12-02 09:44 PM
GTA Thoughts? Typedef Enum Gaming Central 5 09-03-02 05:51 PM

All times are GMT -5. The time now is 10:00 PM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©1998 - 2014, nV News.