3DMurk03 new cheats ?

Odd why didnt he decide to toss out Quake 3 when this happened with ATi??? Game or not.. its still the most used benchmark.

 

This Kyle character is so predictable. Did it really take a rocket scientist to figure out what he was going to say? Hell, almost the exact same words came out of my mouth the moment I said, "Man, I bet I know what that jerk is going to say when he catches wind of this story..."

 

Bullsh!tFX to the extreme. I can't believe they are this desperate. I'm looking forward to websites retesting all other popular benchmarks with the exe names changed.

 

Murk my words: no more quack...

 

Im telling you..

I would almost bet my paycheck that they were doing something with Doom-III as well. Especially as pre setup and Controled by them the whole thing was.

I wonder if anyone will be able to ever find out or prove one way or the other.

 

I think we already have some idea that nVidia was well armed going into that Doom benchmarking session. Remember, Carmack was quoted as saying that he didn't like the demo that nVidia had prepared...so it could have been even worse.

 

Updated the bias meter, wow if NV keeps their current track record, I will have to lenghten the meter.
Second, has anyone tested this with ati cards, I read that ATI caught this, but I wish TR would have also checked ATIs cards.

later,

 

I've no real problem with his opinion; he's entitled to it, wether I agree or not.

What I personally disagree with and do not think he is entitled to are propositions like this:

If the gentleman in question decides to NOT include Futuremark's applications from now on, then that's fine by me. What I don't understand is an uncalled motivation for all other reviewers/websites to follow his decision/opinion.

I usually don't like to get involved in silly personal sharades, but if some other respected IHV apart from NV or ATI has an open position for a specific PR spokesmen, it would be nice if they'd hire him; we might then finally get some peace of mind from what hardly can be called online journalism with at least a shred of integrity.

 

so, any preliminary benchmarks on a GFFX (any model) renaming quake3, splinter cell, UT2k3, etc?

 

hmm, i dont think nv is that stupid. then again....

later,

 

Ask Kyle the humble and all-knowing how reviewers are supposed to detect this kind of cheating without a software path (refrast, I think?) that can output the intended image? I guess he's cooler with the more real-world separate render paths, in which we must rely on reviewers to decide which IQ settings are comparable.

radar, please think before you post.

 

In the next Detonator readme ?

"Workaround for the Anisotropic Filtering bug :
If you have problem enabling Anisotropic Filtering in your games, rename the application .EXE file and try again, AF should now work correctly. Warning, this workaround can induce a small performance penalty."

 

did the same test on the 9800 pro...and well, there is no difference when renaming it...fps for all game tests remained the same, with the final score being 1 point off

 

Renaming UT2003's exe doesn't appear to work (the game won't launch at all) but I have asked Vogel and Sweeney if/how this may be possible. There appears to be some kind of file-linkings when ut2003.exe is launched or something. I'm trying. FYI, I have created a first-person UT2003 demo that I will be using in my forthcoming Albatron NV34 review (if anyone from Albatron visits our forums, I hope they understand why my review hasn't appeared afeter 2 weeks upon receiving their board!).

Renaming Spinter Cell allows the game to launch and run but I haven't done any benchmark investigation since I'm preparing to go on a long overdue holiday with the family during the weekend. Once I'm back, I'll be benchmarking SC with our own demo as well as a new demo I created not too long ago, using either the default named exe or a renamed exe.

 

quote:
--------------------------------------------------------------------------------
Originally posted by CorpesGrinder (my micron screenname was banned)
Kyle, are you going to start checking other games/benchmarks to see if they're affected?
--------------------------------------------------------------------------------


(from Kyle)

At this point I think it is bascially couter productive to spend a bunch of time on going back over bencharks that are years old, but I do think it is very much well worth our time making sure the tools we currently use are solid.

There will be a lot of things changing this year for sure. I am really counting on three games from here out that I think are going to be "must haves" on the benchmarking list. HL2, DOOM3, and XIII. We already have dialouges with these people so I feel confident we can get things done.

 

"We"?

 

some journalist,...he seems to contradict himself...but what do i know, i dont run a well-known website that could potentially influence a consumer's buying decision

 

He has a turd in his pocket....

 

LOL! That made my morning.
Thanks!

 

Slightly OT, but I was wondering if anyone could enlighten me if other drivers/IHV provide similar to the following:

With the Parhelia, the drivers allow you to set individual settings linked to any particular .exe you run (and you can scan for a load of common games/apps, and load up "recommended" settings which you are free to change). This applies to texture filtering, AA, and some other bits and bobs.

Separately they provide an app (the "Apptemizer" ) which scans for games and provides recommended .ini or registry settings (whatever the game uses) - this is more specifically targetted towards surround gaming though, which can occasionally be tricky to set up otherwise.

Both these routes are totally open and tell you exactly what they are doing to your games, and you have the choice to use them or not. This works really well for me.

What I am saying is that I am all in favour of application-specific settings. But only if I can see exactly what is going on, and have control over each aspect of it.

If Nvidia allowed you to see what it was detecting and optimising for, and also allowed you provide your own tweaks, and turn their "optimisations" off if you really wanted, then this would be an absolutely GREAT feature. But they don't allow you control and are not open about what they are doing, so it's not a great feature. It means that as a potential buyer with knowledge of all the recent shenanigans, I have no clue as to what the relative performance of the cards available would be in "Random Game X" would be at my preferred game settings (res, AA, AF).

Just my 2p's worth

Gnep