PDA

View Full Version : Digit-Life Anti-Detection Article Up


Pages : [1] 2 3 4 5 6 7 8 9

StealthHawk
07-03-03, 05:29 AM
http://www.digit-life.com/articles2/antidetect/index.html

It's a translation of what was up in Russian at ixbt. I'm disappointed in the lack of tests. An interesting read if you haven't kept up with the situation, but there really is no new information here. Also, results with a GeForceFX card instead of a GeForce4Ti surely would have been more interesting, to see how much cheating is going on there- more, less, or the same.

DSC
07-03-03, 05:49 AM
Someone should lock up the other thread, so we can keep up all the comments here.

Reading atm, but it's a little big vague, someone needs to write it in a simple language. :o

deejaya
07-03-03, 06:07 AM
I think it's pretty amusing and ironic ATi got clocked for almost 30% (for all the cheat measurers out there :p), and also got caught lowering bit depth.

The best part of that article is this bit, to me:

Unfortunately, the conclusion is rather distressing. First of all, I want to ask Futuremark where they were before and why they started fighting the cheating just a short time ago. Weren't they aware of aggressive optimizations made by vendors for their benchmark? I don't think so. Weren't there any reasons or facts before? They were. Just consider the case when the splash screen in the Detonator go off or the notorious drivers of the SiS Xabre. Weren't they qualified enough to locate and do away with the cheats? I don't believe it. It took us only one day to find and eliminate tricks in the drivers of two largest vendors though we didn't have the source code of the 3DMark2001. We don't want to believe that participation in a beta-program automatically gives the right for such optimizations, but this is the only idea we have. We hope we are wrong.

BigFish7
07-03-03, 06:22 AM
just a great thing to read, everyone should take a look ;)

Kruno
07-03-03, 06:43 AM
Why use 3dmark 2.001K3?

I hate that benchmark, I find 3DMark 2.003K3 much better as it stresses video card more than the CPU.

Why not use those scripts on 3DMark 2.003K3?

How do we know that those scripts are only disabling cheats?

deejaya
07-03-03, 06:51 AM
Originally posted by K.I.L.E.R
Why use 3dmark 2.001K3?
But if you remember, Futuremark always tried to make benchmarks with an eye to the future, that is why its complexity corresponds to what we can see in today's games. Moreover, looking for tricks in this benchmark will help us prove or disprove NVIDIA's statements made right after the release of the Futuremark's patch 330 that eliminates Detonator FX's cheats for the 3DMark2003. If you remember, the company tried to justify the tricks saying that the 3DMark2003 tests didn't comply to what we could see in future games and the 3DMark2003 discredited their products. Let's have a look at the previous Futuremark's solution which didn't produce such a negative reaction and wasn't so biased from NVIDIA's standpoint.

How do we know that those scripts are only disabling cheats?

He doesn't say they only uncover cheats. We don't know, for certain if what he has found are cheats, but you could also say the same for the "cheats" FM uncovered in both ATi's and nVidia's drivers. Reading that article, and looking at the captures, it's fairly obvious both companies have been caught here.

He also starts off saying there will be more things tested when they can figure out a way of doing so, to try and uncover more.

Deathlike2
07-03-03, 07:14 AM
I guess all this reaffirms the cheats in 3DMark03....

Unfortunately Unwinder found more in ATI... ow...

I'm willing to believe NVidia will try to break the patch detection scripts.... at this rate NVidia will be developing "better driver security" than better driver performance... really sad...

On ATI's case... they REALLY REALLY need to rid of their optimizations for benchmarks... PERIOD... since the article says that this whole benchmark is CPU limited... that's something...

If ATI really wants to stay as the honest company that they are claiming to be.. they need to remove optimizations FROM ALL BENCHMARKS... they need not sink to NVidia's level. They can beat NVidia on any level... just don't mess with the benchys

Synthetic Benchmark Optimization = Cheating

The problem with the article.. is that it makes NO mention of what Catalyst drivers that were used... it should've been more thorough on the Dets used (using more Dets versions)

All I can officially say is this...

PR from any company = crap

StealthHawk
07-03-03, 07:23 AM
Originally posted by Deathlike2
On ATI's case... they REALLY REALLY need to rid of their optimizations for benchmarks... PERIOD... since the article says that this whole benchmark is CPU limited... that's something...

If ATI really wants to stay as the honest company that they are claiming to be.. they need to remove optimizations FROM ALL BENCHMARKS... they need not sink to NVidia's level. They can beat NVidia on any level... just don't mess with the benchys

Synthetic Benchmark Optimization = Cheating

Well, Catalyst3.5 was not used or tested, so it might be unfair to jump to any conclusions. 3dmark03 optimizations were removed in Cat3.5, just as ATI stated. I doubt the 3dmark2001 optimizations were removed, though.

If you look at the graph, it says that Catalyst3.4 was tested.

3dmark2001 is not entirely CPU dependent. It is both GPU and CPU dependent.

Deathlike2
07-03-03, 07:33 AM
I guess that's a plus for ATI.. almost.

I went back to look at the article... was a GF4 used instead of an FX?..

I just want to make sure here...

StealthHawk
07-03-03, 07:38 AM
Originally posted by Deathlike2
I went back to look at the article... was a GF4 used instead of an FX?..

I just want to make sure here...

Yes, that is correct.

Deathlike2
07-03-03, 07:43 AM
since the article says that this whole benchmark is CPU limited

I meant 3DMark03.. not 3DMark01... I may be wrong.. but I thought I read that the Nature benchmark on the R300 was CPU limited...

I won't disagree with what you said about 3DMark01 though..

Deathlike2
07-03-03, 07:47 AM
What you just said makes my GF3 looks "cheat-free"... ALMOST.. lol

Have you noticed all these cheats are "tree based"?

Apparently ATI and NVidia don't like the way the trees are done.... LOL.. maybe they aren't tree lovers.. hehe

Kruno
07-03-03, 07:54 AM
3dmark2001 is not entirely CPU dependent. It is both GPU and CPU dependent.

I am interested in video card architecture performance and not total system performance.

DSC
07-03-03, 07:57 AM
For those that are using Catalyst 3.5, you can try the ATIAntiDetector script and see if you lose performance in Nature. That should answer questions about 3.5.

digitalwanderer
07-03-03, 09:07 AM
...that Unwinder did supply his anti-detection script (http://www.ixbt.com/video2/images/antidetect/antidetectorscripts.zip) and you can just download RivaTuner (http://www.nvworld.ru/download/rivatuner.zip) and under the "Power User" option just click on the lil folder icon at the bottom of the page to use it.

Run it on a driver set before you install them, then just go test 3dm2k3 yourself. That's my fave part of this article, it should be easily re-creatable by a number of people so we should have confirmation/negation of it's accuracy pretty soon as well as more "optimization finds" as well as some average percentages.

This is just the begining of it.

The Baron
07-03-03, 09:12 AM
You know, I'm depressed.

There is no way to create a shader benchmark that is tamper-proof. Completely impossible.

Blah.

So much for DX9 performance.

Sazar
07-03-03, 09:20 AM
my question is... why not use an r3xx v/s nv3x comparison ?

would that perhaps not be a little more representative ?

article was underwhelming.. from the rumour mill there was supposedly going to be an attempt to identify some app detections et al... ah well...

can anyone verify the ps/vs version used in nature in 2k1 ? and what version nvidia and ati are both running ?

DivotMaker
07-03-03, 09:22 AM
Originally posted by Deathlike2
Have you noticed all these cheats are "tree based"?

Apparently ATI and NVidia don't like the way the trees are done.... LOL.. maybe they aren't tree lovers.. hehe

I can tell you from experience that animated 3D tree rendering in todays hardware is easily one of the MOST taxing things to render in an outdoor environment. EA's developr for Tiger Woods 2003, Headgate Studios, has worked extremely hard at making convincing trees perform on a wide range of video cards. Still, it is the toughest part of their job...and in 2004 it gets tougher because in addition to trees that are fully animated, they are now adding animated 3D grass for the rough....and so far it looks incredible!

GlowStick
07-03-03, 09:39 AM
That was a pritty good article accpt

where were the geforce fx's, where was 3dmark03, where was ut2k3?
???

?

quake3?

----------------Edit

Oh, also im pritty sure that you would need a diffrent Anti-Detection script for each diffrent version of drivers.

i get the feeling the riva tuner script just goes to certin bits and changes something... so a diffrent driver that would not do the same thing.............

extreme_dB
07-03-03, 09:44 AM
Originally posted by Deathlike2
I guess all this reaffirms the cheats in 3DMark03....

Unfortunately Unwinder found more in ATI... ow...

Where did you get that from?

ATI boosted their score in the Nature test by 30%, while Nvidia boosted their score by 57% in Nature, and by a substantial amount in Dragothic as well.

The funny thing is that ATI was the first to cheat in Nature. Nvidia countered with even more effective cheats, and were probably so pleased with the results that they went crazy with optimizations in other applications. :)

Sazar
07-03-03, 09:45 AM
Originally posted by GlowStick
That was a pritty good article accpt

where were the geforce fx's, where was 3dmark03, where was ut2k3?
???

?

quake3?

:)

and where were the r3xx's ?

I thought there would be more to the article... especially after all the furor...

decent work all around... but fairly limited in its scope..

GlowStick
07-03-03, 10:05 AM
Originally posted by Sazar
:)

and where were the r3xx's ?

I thought there would be more to the article... especially after all the furor...

decent work all around... but fairly limited in its scope..
RADEON 9500 64MB video adapter with the RADEON 9500PRO capabilities unlocked:

they used a 9500,

i can understand that alot of people cannot affoard to buy 1 9800pro 256mb, and one 5900 ultra 256mb, but i thought they get em for free : O

BigFish7
07-03-03, 10:19 AM
Originally posted by GlowStick
RADEON 9500 64MB video adapter with the RADEON 9500PRO capabilities unlocked:

they used a 9500,

i can understand that alot of people cannot affoard to buy 1 9800pro 256mb, and one 5900 ultra 256mb, but i thought they get em for free : O
sure they can get them for free whenever they want to reveiw it and show how good these products are yet whenever you want to show how cheaty the are ... Sigh...
:rofl you can get your hands on them for free :D

Sazar
07-03-03, 10:20 AM
Originally posted by GlowStick
RADEON 9500 64MB video adapter with the RADEON 9500PRO capabilities unlocked:

they used a 9500,

i can understand that alot of people cannot affoard to buy 1 9800pro 256mb, and one 5900 ultra 256mb, but i thought they get em for free : O

ah... k... I only saw references to r200 in the article.. musta missed the line with the 9500 info...

my bad..

mikechai
07-03-03, 11:19 AM
Ok guys, let's face it. No wonder drivers exist.
:mad: