FutureMark & Nvidia joint statement on 3DMark03; FutureMark tucks its tail between its legs.

[Gstanfor]

Quote

--------------------------------------------------------------------------------

nVidia originally withdrew their membership and support for BRIBEmark, whoops, thats not the name..., err, QUAKmark, no thats not it either - close though..., 3DMARK 2003, when futuremark refused to consider benchmark optimizations nVidia put forward in the development stages.

--------------------------------------------------------------------------------


FALSE. Nvidia Withdrew in December, 13 months after development was started and AFTER everything but Bug hunting was complete. a mere 3 months before 3dmark03 was released. They were Fully 100% behind 3dmark03 until it became clear that the Nv3x was going to have troubble with it. Which is Due almost entirely to design flaws, or poor DEsign choices. Like limited Memory bandwidth, poor single texture performance, poor DX9 shader support among other things.

Further, this has nothing to do with bribes. Nvidia has 3-4 Times the Cash and resources ATi has. This has to do with Standing for what is Right and not caving to manipulation and Favrotism.

Quote

--------------------------------------------------------------------------------

I wonder if ATi's "membership subscriptions" have made up in any way for the damage Futuremark have inflicted upon themselves?

--------------------------------------------------------------------------------


It is only 5,000$ a year for membership.

Actually $5000 is the minimum cost of membership for a year, not the "membership subscriptions" that ATi paid...

So what if nVidia has 3 to 4 times the cash of ATi? That doesn't stop ATi from spending exhorbidant amounts on "membership subscriptions" if that is what they wish to do.

 

[BFG10K]

First of some games do use clipping planes. Ultima 9: Ascension is a particular case in point.

Correction: all games use dynamic clipping planes and even if they don't explicitly ask for them the drivers do it for them.

It actually uses 3 clipping planes, statically set in the options.ini file.

You might be setting the distance to clip but it's still dynamic. You can still move anywhere you like and it won't break the game, unlike nVidia's clip planes which are completely static and can only exist thanks to prior knowledge of the animation loop.

Second, I re-iterate, the user cannot control the viewpoint in 3dmark03.

They can't control it while running benchmarks in real games either and that's where the suspicion comes in. The proven clip planes in 3Dmark can be done just as easily in 3D gaming benchmarks and thus the results obtained in the tests mean absolutely nothing when playing the game properly as the static clips planes cannot exist in such a case. It's outright deception as the benchmarks show one thing but reality is completely different.

Static clip planes can only serve to inflate benchmark results but can never translate into the real world. It's simply artificially inflated performance in an attempt to deceive customers.

Pre-rendered or rendering on the fly is irrelevant here

the viewpoint will remain the same in both cases.

So what happens when Future mark makes a new test that follows a different path or direction? What then? Do you advocate nVidia builds new clip planes just for the new benchmark?

Think about the idiocy of what you're saying. These are not optimisations by any stretch of the imagination.

 

[Hellbinder]

The same can be said about ATI..they cheated in Quack3.exe,AND they admitted to cheating in 3dMark2k3 (although with lesser performance gain) I see it this way, ATI cheated on a GAME, NV didnt, which is the lesser evil? whether it be 3 years ago, or yesterday, they still did it.

No, Ati did not admit to Cheating in 3dmark. What ATi did and What Nvidia did are completey Different in their intent and method. This has been addressed at Beyond3d in great detail and Several developers also posted their View on the matter. However, what both did is considered cheating becuase no optimizations are allowed in 3dmark period.

Which is worse cheating in a game or benchmark. Did you even think about this before you posted??? There is no difference. Because Quake was being used AS A BENCHMARK!. Which is the entire issue here. If you can do an optimization that affects a whole game or games then it is GOOD. As long as it does not negatively affect the IQ or the intent of the developer. In a BENCMARK situation the rules are completely different. Because in Rail bencjmarks, or Timedemos you can examine them Frame by Frame and make tweaks that have no effect at all in real gaming. Which artificially infaltes the score or expectation of the cards performance. Thus it is CHEATING. The difference between what ATi and Nvidia did in 3dmark is. ATi's is an HONEST *Optimization* the slight reorder of exsisting shader code for water and Sky in GF4. NVidias are *CHEATS* that cannot be applied to real games without affecting the IQ or changing the Intent and purpose of the Developer. Which include lowering IQ and inserting clip planes.

Further what ATi did with Quake also increased your frame rate while playing the game not just the timedemo. It did so by lowing the IQ of certain textures in Q3. Something that is Commonly practiced today by several IHVs. And can easily be shown that Nvidia has a LON HISTORY of lowering IQ for speed accross ALL GAMES going back all the way to the origional TNT.

And noone has STILL answered my question...if this is a driver issue and not hardware, why is NO ONE getting these optimizations with other NV cards? I see NO degradation in my 44.03 drivers, with my GF3 Ti200. Performance is up a little bit in my games (this is normal for NV drivers to gain a LITTLE bit), and I wont run 3dMark AT ALL, coz I dont find it viable and real world. You cant play it, so why bother?

This is a Very obvious reason. Becuase the FX is a new architectre and Driver optimizations and Cheats will simply not affect older cards. Can your GF3 run GT4 in 3dmark03?? can it Run pixel shader 2.0??? does it have any DX9 functionality???

NO.

 

[Hellbinder]

Actually $5000 is the minimum cost of membership for a year, not the "membership subscriptions" that ATi paid...

So what if nVidia has 3 to 4 times the cash of ATi? That doesn't stop ATi from spending exhorbidant amounts on "membership subscriptions" if that is what they wish to do.

You have absolutly no idea what you are talking about. You clearly have no concept of how the beta program for Futuremark works.

It is blatantly infantile for you to be suggesting that ATi bought 1,000,000 copies of 3dmark just to get special treatment. No, it is flat out SICKENING that you are posting trash like this instead of standing for what is right. There are several Beta members who have drrect input into the creation of 3dmark. Dont you think they would all notice and take exception if something like this was going on. Like SIS or DELL for instance???

 

[Gstanfor]

Originally posted by: Hellbinder

Actually $5000 is the minimum cost of membership for a year, not the "membership subscriptions" that ATi paid...

So what if nVidia has 3 to 4 times the cash of ATi? That doesn't stop ATi from spending exhorbidant amounts on "membership subscriptions" if that is what they wish to do.

You have absolutly no idea what you are talking about. You clearly have no concept of how the beta program for Futuremark works.

It is blatantly infantile for you to be suggesting that ATi bought 1,000,000 copies of 3dmark just to get special treatment. No, it is flat out SICKENING that you are posting trash like this instead of standing for what is right. There are several Beta members who have drrect input into the creation of 3dmark. Dont you think they would all notice and take exception if something like this was going on. Like SIS or DELL for instance???

Of course Intel never paid BAPCO off to make them look good in benchmarks either... Wait - aren't ATi and Intel strategic partners?...

EDIT: The other members of BAPCO really objected to Intel's behaviour and forced them to stop it too - not! Isn't Dell a BAPCO member?

EDIT2: bapco membership Guess who are BAPCO members??? ATi & Futuremark!!! ROFLMAO!

 

[ViRGE]

Originally posted by: HellbinderUsing custom compiled shaders with Nvidias PROPRIETARY Back end Compiler for Cg..

For the last damn time, Cg isn't proprietary; you are free to write your own damn compiler for it

 

[Insomniac]

Originally posted by: Adul
http://www.hardocp.com/article.html?art=NDgy

Kyle has yet another thing or three to say about this.

You can flip that thing the other way. All of Dell's and Gateway's money (if you want to assume that the guys at FM play favorites to the guys with $$$) is paid to FM to ensure that they don't optimize or play favorites to any company. Dell uses that to help them choose which cards to put in their system. They don't sell every card and they certainly don't validate them in their systems. These guys have millions of customers and their own reputation to protect. Why would Dell want to buy products that appear to be better when they really aren't?

Gstanfor: You think ATi and FM conspired to make it look like nVidia cheated? They decided to set them up, then tell someone about it to expose them? It has been 4 days now. If nVidia did nothing, they would know tht immediately. Too bad they're not defending themselves as vigorously as you are. I wonder why?

 

[NOX]

All this arguing over a synthetic benchmark! No wonder I hardly post here anymore, and probably why I wont be renewing my subscription.

What a waste of forum space! :|

 

[DaveSimmons]

Originally posted by: NOX
All this arguing over a synthetic benchmark! No wonder I hardly post here anymore, and probably why I wont be renewing my subscription.

What a waste of forum space! :|

not arguing over a benchmark, arguing over nvidia's attempt to deceive cnsumers by cheating on a synthetic benchmark.

 

[MercenaryForHire]

Originally posted by: NOX
All this arguing over a synthetic benchmark! No wonder I hardly post here anymore, and probably why I wont be renewing my subscription.

What a waste of forum space! :|

Waste of space?

God forbid you should ever stumble into Off-Topic by accident.

- M4H

 

[NOX]

Originally posted by: DaveSimmons

Originally posted by: NOX
All this arguing over a synthetic benchmark! No wonder I hardly post here anymore, and probably why I wont be renewing my subscription.

What a waste of forum space! :|

not arguing over a benchmark, arguing over nvidia's attempt to deceive cnsumers by cheating on a synthetic benchmark.

You?re kidding right? You expect me to believe that consumers will purchase a particular product because of a synthetic benchmark, which they don?t even know exist?! Less then 1/2 a percent of the entire planet knows 3dmark! Please, this is silly.

The only two things I?ve learned from this thread is 1) that ATI cheated, then Nvidia recently cheated! 2) ATI and Nvidia Fanboys are even more unpleasant then GUTB!

 

[Megatomic]

Originally posted by: NOX
You?re kidding right? You expect me to believe that consumers will purchase a particular product because of a synthetic benchmark, which they don?t even know exist?! Less then 1/2 a percent of the entire planet knows 3dmark! Please, this is silly.

The only two things I?ve learned from this thread is 1) that ATI cheated, then Nvidia recently cheated! 2) ATI and Nvidia Fanboys are even more unpleasant then GUTB!

My sentiments exactly.

 

[Dean]

More new "Optimizations"

3dmark01 found with them also.

"Detonators have 73-element table of "recognized" applications
- ~50 "recognized" shaders"


Thats application detection for 73 different programs folks!!

Read about it here Beyond 3D

The guy who detected them also makes Riva Tuner ;-)

 

[Gstanfor]

I think you will find that for the recognised shaders at least, they are being passed through Cg and not through HLSL.