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

Newegg Daily Deals

View Poll Results: Are they cheating again?
Yes 132 67.01%
No 53 26.90%
I don't have a clue by four. 12 6.09%
Voters: 197. You may not vote on this poll

Closed Thread
 
Thread Tools
Old 05-15-03, 11:58 PM   #229
DorXtar
3Dlabs Wildcat IV 7210
 
Join Date: Feb 2003
Location: Los Angeles
Posts: 52
Default

Quote:
Originally posted by PsychoSy
Since the other thread was closed due to a number people who thought it was a [...............] develope independantly.
I must agree with whatever this guy says.... He has vic mackey as his avatar...
__________________
I am Dork.
DorXtar is offline  
Old 05-16-03, 12:04 AM   #230
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

Quote:
Originally posted by Hanners
Just read the thread in question, I think you omitted to post the other side of the story being given out over there:
so do you think ati has cheated specviewperf?
i believe ati has cheated specviewperf, i am not gonna to prove it or make you to believe it but bindly trust anything is not clever.
screw up 3DMark is the right step in the right direction, since i dont want the 3DMark optimization in my video card driver, 3DMark is too much of an interference in the video card industry now.
Behemoth is offline  
Old 05-16-03, 12:22 AM   #231
PreservedSwine
I'm always hungry
 
PreservedSwine's Avatar
 
Join Date: Aug 2002
Posts: 548
Default

Quote:
3DMark optimization in my video card driver, 3DMark is too much of an interference in the video card industry now.
If that's true (which I don't beleieve), there is no one to blame but users. To me, it's just a benchmark, one thing upon many to consider when considering a new video card, and what it's limitations will be. In that regard, I feel the more info, the better.

At what point do you draw the line and say which benchmarks are valid (for you), and which are in-valid? Certainly they don't carry the same amount of weight. I've never like the idea of morons who buy cards just so they can boast about high 3Dmartk scores, and tweak their cards just for 3Dmarks. Those guys/gals are missign the point completely, IMO. But should you blame the benchmark for that, or the fools who abuse it?

3Dmarks serves a limited purpose, no more, no less. But to cheat in it is deceptive, to say the least.
PreservedSwine is offline  
Old 05-16-03, 12:27 AM   #232
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

Quote:
Originally posted by Hanners
Elegant?

If I owned a nVidia product, I think I'd be more upset about all the man hours that must have been wasted changing the clip planes in every frame of the benchmark than anything else. Think how much work could have gone into optimising real games and applications, instead of hacking every frame of a benchmark?
i would not think its a waste of man hours, knowledge gain in any kind of optimization can be used to further sophisticate the in-game driver optimization code.
this kind of 3DMark optimization may very well evolve to be used on in-between scenes' real time animations automatically too.
Behemoth is offline  
Old 05-16-03, 12:30 AM   #233
reever2
Registered User
 
reever2's Avatar
 
Join Date: Apr 2003
Posts: 489
Default

http://www.the-inquirer.com/?article=9486

I wonder why the Inq never caught onto the cheating, surely they thought such a monumental increase in 3dmarks would equate to more than 1-2 fps in games?
reever2 is offline  
Old 05-16-03, 12:39 AM   #234
Behemoth
radeon 9800 pro
 
Behemoth's Avatar
 
Join Date: Apr 2003
Location: Darkness Falls
Posts: 841
Default

Quote:
Originally posted by PreservedSwine
If that's true (which I don't beleieve), there is no one to blame but users. To me, it's just a benchmark, one thing upon many to consider when considering a new video card, and what it's limitations will be. In that regard, I feel the more info, the better.

At what point do you draw the line and say which benchmarks are valid (for you), and which are in-valid? Certainly they don't carry the same amount of weight. I've never like the idea of morons who buy cards just so they can boast about high 3Dmartk scores, and tweak their cards just for 3Dmarks. Those guys/gals are missign the point completely, IMO. But should you blame the benchmark for that, or the fools who abuse it?

3Dmarks serves a limited purpose, no more, no less. But to cheat in it is deceptive, to say the least.
i agree 3DMarks serves a limited purpose, but futuremark have not done enough to avoid its abusive misuse, and actually i think quite the opposite that futuremark is using this kind of misuse, confusion, using its benchmark using average users using its 3DMark influence just to make money out of video card manufacturers, make money out of 3DMark lovers.
there is no place for 3DMark optimization code in driver, working with manufacturer to make the cards look good in this benchmark is deceptive enough in the first place, screw it.
Behemoth is offline  
Old 05-16-03, 01:04 AM   #235
MrNasty
MrNasty
 
Join Date: Dec 2002
Location: Deli
Posts: 213
Default

Quote:
Originally posted by CoWBoY
LMAO... ya, 3DMark is a valid benchmark. Stick to the games to dictate performance, not some silly benchmark. Sad thing is if ATi continues to support 3DMark while Nvidia increases their performance in games, ATi will soon realize they are just wasting time and hopefully users will become more enlightend instead of dumb and lazy.
That was dumb of soooo many levels.
__________________
You are not your system specs.
MrNasty is offline  
Old 05-16-03, 01:47 AM   #236
threedaysdwn
Registered User
 
Join Date: Aug 2002
Posts: 90
Send a message via AIM to threedaysdwn
Default

Quote:
Originally posted by Rogozhin
bk

extremetech verified that the scores were affected by this hardcoding.

They had to hard wire each individual FRAME- that takes time lots of it by their dev team-and it is a cheat since it was done on purpose and it was done with the intent of running edmark2003 FASTER.

I just can't believe they thought they could do this sort of thing and not get caught. They did it with the second set of fx 5800 drivers too I'm sure.

wonder how deep this rabit hole goes-ut2003 time demo? SC time demo? doom 3 time demo?



rogo

That's quite an assumption. I suppose you have access to Nvidia's driver code?

I mean, seeing as that's the only way you'd know exactly how they did this.

Maybe the 3DMark movement code is flawed and causes an anomaly with the culling and clipping algorithms in Nvidia's drivers.

I find it unlikely that Nvidia would go through the entire benchmark "hand crafting" a clip plane for the 3DMark demo.

Hell, just looking at how that clip plane only exists in *some* of the tests and seems to have quite different effects in each one brings some question to this "theory."

I for one am not ruling out any possibilities... but as I said above, I doubt they'd go to the lengths you suggest.

Anyhow, it was clearly outlined in Anand's review that nvidia had NEVER seen the timedemo that JC put together for the review tests.

Futher, the Nv35 is winning all other comparisons regardless of what timedemo is used (several reviews just recorded 1 or 2 minutes of game time with FRAPS... I suppose they're "cheating" here too?)


Even more interesting is that this "cheat" seems to buy Nvidia no appreciable performance gain over previously released drivers, as the Detonator FX's advantages are all in games and not 3Dmark.
__________________
New! Compy 8.0 - Intel Core 2 Duo E6600 @ 3.0Ghz - EVGA Nvidia nForce 680 - 4GB DDR800 - 2x WD Raptor 10,000RPM 74GB in RAID 0 - 2x 300GB 7200RPM local storage - Geforce 8800 GTX 768MB - HP 23" Widescreen
threedaysdwn is offline  

Old 05-16-03, 01:54 AM   #237
threedaysdwn
Registered User
 
Join Date: Aug 2002
Posts: 90
Send a message via AIM to threedaysdwn
Default

ack how did the boards screw this up.
__________________
New! Compy 8.0 - Intel Core 2 Duo E6600 @ 3.0Ghz - EVGA Nvidia nForce 680 - 4GB DDR800 - 2x WD Raptor 10,000RPM 74GB in RAID 0 - 2x 300GB 7200RPM local storage - Geforce 8800 GTX 768MB - HP 23" Widescreen
threedaysdwn is offline  
Old 05-16-03, 01:59 AM   #238
rwolf
Rock Star
 
Join Date: Oct 2002
Posts: 122
Default

It all comes down to $$$$.

If your benchmarks look good. Consumer buy your card and oems buy it.

Look at the huge boost the 5600 and the 5200 received. These cards are poor performers. It looks like Nvidia is cheating to put these cards in a better light. People are going to buy them and have slideshows instead of games.


Now that Nvidia has been caught, I know that there is no correlation between the benchmarks and gaming performance.

If you read this review:

http://www.digit-life.com/articles2/gffx/5900u.html

You will see that the 5900 gets creamed by the 9800Pro in the Rightguard benchmark. But does well in the "standard" benchmarks.

Has Nvidia tainted them? Who knows, but I have lost faith in Nvidia's benchmarks.
__________________
Rock on.
rwolf is offline  
Old 05-16-03, 02:10 AM   #239
rwolf
Rock Star
 
Join Date: Oct 2002
Posts: 122
Default

Quote:
Originally posted by CoWBoY
LMAO... ya, 3DMark is a valid benchmark. Stick to the games to dictate performance, not some silly benchmark. Sad thing is if ATi continues to support 3DMark while Nvidia increases their performance in games, ATi will soon realize they are just wasting time and hopefully users will become more enlightend instead of dumb and lazy.
Looks like Nvidia is spending the time coding special effects (FX) into their drivers not ATI.

The nice thing about 3dMark03 is that it allowed us to see that they are not rendering the scenes properly and that makes you wonder if all the other "Game" benchmarks are custom manipulated frame by frame.

If this is truely a cheat and not a bug it is a dark day for Nvidia.
__________________
Rock on.
rwolf is offline  
Old 05-16-03, 02:21 AM   #240
MrNasty
MrNasty
 
Join Date: Dec 2002
Location: Deli
Posts: 213
Default

Hey cowBOY how come your precious nvidia does not care about 3dmark as you claim yet cheats to make it's drivers perform faster in it ? Get it right.
__________________
You are not your system specs.
MrNasty is offline  
Closed Thread


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
Need Help Installing NVIDIA Tesla M2070Q in Linux RHEL5 Ferianto85 NVIDIA Linux 0 05-18-12 09:35 PM
Rumor regarding lack of 680 availability ViN86 Rumor Mill 6 05-09-12 05:48 PM
NVIDIA could rule if they really wanted to Vid_craze NVIDIA GeForce 7, 8, And 9 Series 25 08-16-02 06:24 PM

All times are GMT -5. The time now is 12:30 PM.


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