Elite BasT**** Chimes in on Nvidia/3dmark.

Wow, I followed the link looking to find something that might be taken with a grain of salt.. but instead found a hopeless and well crafted misinformation crusade.

It begins by trying to bemuddle the reader by creating a context for normal Z-clipping in games. This is trying to compare static clipping/cheat with dynamic clipping/normal.

This is done quite a bit in games from previous years and even today. You can clearly see this is many large world games like expansive RPG games (Morrowind, Gothic), city driving sims (Grand Theft Auto 3, Mafia) or racing games. In the distance, you can see things coming into view as you move.
Click to expand...

The article then creates a false pretext:
This would be a good sign of possible cheating if other reviews and previews showed the same results. The thing is, they don't.
Click to expand...

The other review sites still show a very largish improvement over both the 9800 and 9700 Pros. He also includes sites that have already indoctrinated that they will use the more balanced NVIDIA driver settings in past reviews versus driver defaults (like ExtremeTech has done).

More non-substatiated assumptions portrayed as if "fact":
In this Detonator FX review you can see that the GeForce FX 5200 and 5600 get performance increases across the board. So if the cheating was true in this driver review's case, then every benchmark used in it would have cheating happening.
Click to expand...

The best part of this misinformation crusade is the grandstanding of trying to create vapor theories to lead the reader into FUD-land:
How much of each variable has to do with these differences in score and Frames Per Second? There is the clock rates of the different cards, different architecture, different computer systems, different Windows installations, different settings and most of all, different testers. Yes, human error has a role to play in this.[/b]
Click to expand...

The whole article is nothing but a FUD crusade.. it makes no attempt to remain objective, asks no questions just makes statements of outlandish/unsupported assumption, and finishes off with my favorite line:
Here's me signing off and hoping you all keep an open mind, because you never know who is pulling the strings or your leg.
Click to expand...

It's pretty obvious who is doing the leg pulling here. :)
 
It's the internet. Basically anyone can write anything they want and present it as the truth. Oh wait...that happens everywhere... All the same, the point remains.
 
Vince said:
Ohh my, a well layed out opposing viewpoint to my own. By God, It just can't be!! :rolleyes:
Click to expand...

Um, an opposing viewpoint is one thing, one that is based on falsehoods, and/or backward logic is something else....something to laugh and shake your head at. ;)
 
Vince said:
Ohh my, a well layed out opposing viewpoint to my own. By God, It just can't be!! :rolleyes:
Click to expand...

Where "well layed out" would be the key phrase.
I think their argumentation was embarrasing rather than well layed out, which accounts for my post.
 
Suggesting to wait for more evidence before making a judgement... blasphemy!!!
Click to expand...

And just how much evidence do you need? I though Extremetech laid out a pretty good case. The only thing i've seen to oppose their view is....???
Do you expect Nvidia to come out and say "ok guys, you caught us." I'm beginning to wonder if that is what it would take for some people. Even if they did, I'm sure there would still be people that would refuse to believe it.
 
ninelven said:
Suggesting to wait for more evidence before making a judgement... blasphemy!!!

Ninelven
Click to expand...

Agreed, though basing your entire theory on circumstantial evidence (DetFX drivers show performance boost in other areas + rendering bug in 3DMark2001 = no cheating in 3DMark2003) isn't too slick, IMO. And trying to explain the clipping away as "aggressive culling" was amusing.
 
jjayb said:
And just how much evidence do you need?
Click to expand...

And just how do you know I haven't made a decision and what it is? ;) Snap judgement maybe? I just thought it was a valid viewpoint. Sure there were flaws in the guy's reasoning, but that really has no effect on the validity of the argument from a philosophical stance.

IMHO, it isn't neccessarily evidence that reveals the truth but time.... guess one could say I agree with Plato in this respect.
 
Well the first part is just a drawn out definition of cheating and would set the context quite nicely for the other side of the debate just as well.

The important fact here is that 3DMark 2003 scores from other websites do not show the Extreme Tech performance gap as seen here. There might be slight leads, but nothing like what their results show.
Click to expand...

If it's so important, they why have they linked sites that don't benchmark with the same settings as ET? The only site that comes close is bit-tech, who run 3DM2003 @1280x1024x32 w/4xFSAA and 8xAF and they actually get more of a performance gap between the 5900 Ultra and the 9800 Pro that ET - over 32%, lol.

"We would like to present Exhibit A to the jury...

R350: 1916
NV35: 2534

Umm... that can't be right, can it?" :rolleyes:

Of the good points they make (and there are quite a few), none are in any way original or back up their main argument. It just seems like a compilation of vague reasoning and shallow accusations padded out with accurate but redundant information.

The most notable problem IMO, is that they appear to try and suggest that the performance increases evident in DetFX have only been realised through cheating (not actual optimisations) and use this premise as a basis by which to argue that cheating is not occuring. It's absurd! :LOL:

Anyway... quite refreshing to see a site that at least *attempts* to discredit the accusations themselves instead of trying to explain how useless 3DMark is as an excuse for such cheats.

MuFu.

Edit - sp
 
They don't appear to appreciate that these issues are not related to the DetFX's though, which kinda breaks down their entire premise.
 
Yah, good point. It all kind of broke. Brain no-worky.
tardbang.gif


MuFu.
 
Dunno, I thought the point was this:

Quinn1981 said:
I hope they aren't cheating. I'll say that it's extremely possible they are cheating, but it's too early to tell.
Click to expand...

Sadly, the validity of the arguement has nothing to do with how well it is made(i.e. something very wrong can be argued very well and something true can be argued very poorly... doesn't change reality though). I guess, in the end, we all see what we want to believe. :(
 
You know Dave,

It saddens me that B3D isn't going to get directly involved in this. Don't get me wrong, I completely understand (or I think I do!) why you're not, and it's understandable. It just bugs me that one of the only sites that's actually capable investigating this with a thoughtful approach...can't. :cry:
 
ninelven said:
Sadly, the validity of the arguement has nothing to do with how well it is made(i.e. something very wrong can be argued very well and something true can be argued very poorly... doesn't change reality though). I guess, in the end, we all see what we want to believe. :(
Click to expand...

Reality also dictates that the informed opinion generally prevails.

MuFu.
 
You must log in or register to reply here.

Similar threads

H
6800GT review by some 'Hanners' guy...
2 3
Replies
53
Views
7K
Optimummind
O
R
Explain To Me The Benefits of SM 3.0 For Nvidia
2 3 4
Replies
63
Views
10K
Reverend
R
digitalwanderer
New article using Unwinder's anti-cheat script at EB
Replies
13
Views
3K
Neeyik
N
digitalwanderer
Something scary, I made the top 10 at B3D!
2
Replies
35
Views
4K
zurich
Z
Back
Top