NVIDIA Denies Unified Compiler Turn Off by 3DMark03

Tafinho

Tafinho

Power Member
AHAHAHAHAHAHAHHAHA
Finalmente abriram a pestana...
Ainda esta a tentar perceber como é que atravez da API se desligava alguma coisa numa driver...

in xbitlabs

Luciano Alibrandi, European Product PR Manager for NVIDIA Corporation, has made a correction in regards previous information about NVIDIA’s Unified Compiler and 3DMark03 benchmark after getting into details with the company’s engineers. Apparently, the statement claiming that NVIDIA’s Unified Complier deployed to optimize pixel shader performance is disabled by the new version of 3DMark03 is not fully correct.

“I would like to inform you that a part of my response was not accurate. I stated that the compiler gets disabled, by 3DMark and that is in fact not true,” he said.

So, after all NVIDIA denied the problems between the Unified Compiler technology and the latest version of popular 3DMark03 benchmark. As a result, we may now conclude that the accusations in Futuremark direction from Hans-Wolfram Tismer, a Managing Director for Gainward Europe GmbH were not correct at all.


Ou seja, os 25% de perda de performance deviam-se a "application-specific optimizations" cheaters...
 
E o software impossivel de cracar também é verdade.
Mostrem-me o primeiro programa que não foi ainda crakado.
 
hmmm um kromo kkr bem ke pode krackar o 3dmark, paquilo lhe dar 40000marks com uma tnt2 :D pena é ninguem acreditar no score dele e tirarem da orb :(
N percebo pq nao acreditam :rolleyes:
 
Tafinho,

""application-specific optimizations" cheaters..."

????

Isto é muito vago.

Eu gostava era de ver uma Radeon
a bombar em Linux. Infelizmente
ainda não se vislumbra nada.
 
Deves andar cegueta porque já vi umas tantas a trabalharem... em OpenGL e tudo...

Creio que até foi criado um tutorial aqui no forum para explicar como instalar os drivers Cat 3.5 e tudo... ou aqui ou vi no Rage3d...
 
Para alem do que o Raptor disse, existem tb drivers de Fire GL/XL (ou la como se chamam) que podem ser instalados nas 9500/9700 e bombam fixe em Linux. :)
 
Originally posted by Metro
E o software impossivel de cracar também é verdade.
Mostrem-me o primeiro programa que não foi ainda crakado.
Clicar para expandir...

é diferente tu cracares um programa mudando-lhe o executável, de tu mudares o executavel sem o alterares...
 
Curioso que o Derek Perez disse que o compiler não sei das quantas foi desactivado neste patch do 3DMark, e passado algum tempo vêm outro PR-Manager da NV dizer que não era disso.

Sinceramente... alguns PR da Nvidia deviam ir para o caixote do lixo... parece que ninguém se entende.

Ou os PR nem percebem o que a empressa anda a fazer... no minimo ridículo.
 
"An official from NVIDIA Corporation confirmed Mr. Tismer’s accusation that “patch 340 disables the GPU compiler. The compiler has to run on the CPU instead resulting in code harder to digest and taking away 20% of the performance.” “Yes, that is actually the case with the new patch 340 that Futuremark posted,” said an NVIDIA spokesperson on Wednesday.

“Few weeks ago we released our 52.16 driver that includes our brand new unified compiler technology. With the new patch the benchmark, our unified compiler gets not used by the app so it goes to CPU and we are definitely slower,” Luciano Alibrandi, NVIDIA’s European Product PR Manager, added."

http://www.xbitlabs.com/news/video/display/20031112181114.html

Lembro-me do Derek ter dito a mesma coisa... vou ver se encontro.
 
Originally posted by iJFerreira
Raptor,

para além dos drivers não serem
muito estáveis, a performance deixa
algo a desejar.

Pelo menos é o que este diz e ele
deve perceber do assunto, não?

http://homepage.hispeed.ch/rscheidegger/atilinux_oct03/ati_linux_comp_oct03.html
Clicar para expandir...

bombar é funcionar... logo, bombam! lol ah pois.... português é f*dido

Eu gostava era de ver uma Radeon
a bombar em Linux. Infelizmente
ainda não se vislumbra nada.
Clicar para expandir...

Anda a minha casa, tenho aqui um portatil com Ati a gostar muito dos drivers em linux... (ou tive...)
Pois é, basta IR ao google antes de falar, e nao só depois de ser contestado...

vai aos forums do rage3d, vais ficar admirado na parte linux... cumps!
 
"With the introduction of the GeForce FX - we built a sophisticated real-time compiler called the Unified Compiler technology. This compiler does real-time optimizations of code in applications to take full advantage of the GeForce FX architecture.

Game developers LOVE this - they work with us to make sure their code is written in a way to fully exploit the compiler.

The end result - a better user experience.

One of the questions we always get is what does this compiler do? The unified compiler does things like instruction reordering and register allocation. The unified compiler is carefully architected so as to maintain perfect image quality while significantly increasing performance. The unified compiler a collection of techniques that are not specific to any particular application but expose the full power of GeForce FX. These techniques are applied with a fingerprinting mechanism which evaluates shaders and, in some cases substitutes hand tuned shaders, but increasingly generates optimal code in real-time.

Futuremark does not consider their application a "game". They consider it a "synthetic benchmark". The problem is that the primary use of 3DMark03 is as a proxy for game play. A website or magazine will run it as a general predictor of graphics application performance. So it is vital that the benchmark reflect the true relative performance of our GPUs versus competitors.

And, while they admit that our unified compiler is behaving exactly the way it behaves in games and that it produces accurate image quality, they do not endorse the optimizations for synthetic use. Hence, Futuremark released a patch that intentionally handicapped our unified compiler.

So, we advocate that when reviewers are using 3DMark as a game proxy, they must run with the unified compiler fully enabled. All games run this way. That means running with the previous version of 3DMark, or running with a version of our drivers that behave properly.

Derek Perez
Director of Nvidia PR
"
Clicar para expandir...

http://www.theinquirer.net/?article=12631

Derek strikes again... :D
"At one we asked Derek how this sat with the optimisations guidelines that were given to press by NVIDIA, specifically the guideline that suggests "An optimization must accelerate more than just a benchmark" To which Derek's reply was "But 3DMark03 is only a benchmark" -- it was suggested that this particular guideline should read "An optimization must accelerate more than just a benchmark unless the application is just a benchmark"! "
Clicar para expandir...

http://www.beyond3d.com/articles/3dmark03/340/

Eu gosto disto... :D

For immediate release

Recently, FutureMark released a new patch, version 340 for its once industry-leading 3DMark2003 benchmark. We at Nvidia are part of FutureMark's beta program (a program that can cost up to hundreds of lawyers per year to bully your way into), and we would like to say that we completely agree with FutureMark's optimization guidelines (we internally refer to those as "wishful thinking"). Since FutureMark are our good pals now, we will not say they did something to make us look bad with the 340 patch, but we will instead remind them that their finnish headquarters are mostly made of wood, and we would hate if something bad was to happen.

Here at Nvidia we are dedicated to provide our 120 450 465 406 546 504 654 065 461 0425 406 546 876 406 903 541 658 706 013 210 601 300 304 650 465 044 056 106 473 customers with the best benchmark results available. We would love to see FutureMark release "3DMark 2003 : the game" so we could optimize for both it and the benchmark and be consistent with our own guidelines (we internally refer to those as "PR crap for gullible reviewers"), but they don't want to so we have to settle for optimizing for the benchmark itself. Regarding the recently released numbers, here is what our superhuman chief scientist David Kirk has to say : "We don't object to those performance numbers as released by FutureMark, but don't forget our direct competitor is cheating in the benchmark : since they are using only 24 of those complicated "bits" thingies, it is likely they use the remaining 8 bits (remember, bits can only come into powers of 2) to hide their 3DMark2003 cheats. We tried to raise the problem with FutureMark, but so far they haven't listened to us.
I still personally believe that patching benchmarks is the wrong answer, since it can disrupt the result of perfectly legit and totally generic optimizations. This is just a slight bug in our software that our perfectly legit and totally generic optimizations only kick in when 3DMark 2003 is started, but our next driver revision should bring back the exact same level of performance, together with bilinear filtering (which looks just as good as the real thing in a variety of situations)"

Nvidia is a global cheater in the communication age, and our goal is to "deface every pixel on the planet".
Clicar para expandir...
 
For immediate release

Recently, FutureMark released a new patch, version 340 for its once industry-leading 3DMark2003 benchmark. We at Nvidia are part of FutureMark's beta program (a program that can cost up to hundreds of lawyers per year to bully your way into), and we would like to say that we completely agree with FutureMark's optimization guidelines (we internally refer to those as "wishful thinking"). Since FutureMark are our good pals now, we will not say they did something to make us look bad with the 340 patch, but we will instead remind them that their finnish headquarters are mostly made of wood, and we would hate if something bad was to happen.

Here at Nvidia we are dedicated to provide our 120 450 465 406 546 504 654 065 461 0425 406 546 876 406 903 541 658 706 013 210 601 300 304 650 465 044 056 106 473 customers with the best benchmark results available. We would love to see FutureMark release "3DMark 2003 : the game" so we could optimize for both it and the benchmark and be consistent with our own guidelines (we internally refer to those as "PR crap for gullible reviewers"), but they don't want to so we have to settle for optimizing for the benchmark itself. Regarding the recently released numbers, here is what our superhuman chief scientist David Kirk has to say : "We don't object to those performance numbers as released by FutureMark, but don't forget our direct competitor is cheating in the benchmark : since they are using only 24 of those complicated "bits" thingies, it is likely they use the remaining 8 bits (remember, bits can only come into powers of 2) to hide their 3DMark2003 cheats. We tried to raise the problem with FutureMark, but so far they haven't listened to us.
I still personally believe that patching benchmarks is the wrong answer, since it can disrupt the result of perfectly legit and totally generic optimizations. This is just a slight bug in our software that our perfectly legit and totally generic optimizations only kick in when 3DMark 2003 is started, but our next driver revision should bring back the exact same level of performance, together with bilinear filtering (which looks just as good as the real thing in a variety of situations)"

Nvidia is a global cheater in the communication age, and our goal is to "deface every pixel on the planet".
Clicar para expandir...

Mega LOL

Um clássico
 
Tipo, curtam só! :D

Regarding the recently released numbers, here is what our superhuman chief scientist David Kirk has to say : "We don't object to those performance numbers as released by FutureMark, but don't forget our direct competitor (ATI) is cheating in the benchmark : since they are using only 24 of those complicated "bits" thingies, it is likely they use the remaining 8 bits (remember, bits can only come into powers of 2) to hide their 3DMark2003 cheats. We tried to raise the problem with FutureMark, but so far they haven't listened to us.
Clicar para expandir...

Mega LOL

I still personally believe that patching benchmarks is the wrong answer, since it can disrupt the result of perfectly legit and totally generic optimizations. This is just a slight bug in our software that our perfectly legit and totally generic optimizations only kick in when 3DMark 2003 is started, but our next driver revision should bring back the exact same level of performance, together with bilinear filtering (which looks just as good as the real thing in a variety of situations)"
Clicar para expandir...

Again... Mega LOL


Epá, sinceramente, se um dia apanhasse um PR da Nvidia com um discurso destes, acho que simplesmente cuspia-lhe para a cara...
 
......esse press release é no gozo.........

......não leves a mal, posso ter percebido mal o teu post, mas parece-me que levas-te aquilo a sério..........
 
LOL

Hey... esse último quote não foi do PR da NV. ;)

Anyway... acho que o que fez mais rir foram aqueles photochops do cooler da FX 5800U... até havia por aí um vídeo.

Enfim este ano desde o lançamento do NV30 vai ficar para a história. :)
 
Inicie sessão ou registe-se para poder participar.
Back
Topo