nVidia release new all singing all dancing dets.

[walkndude]

I agree with you to a certain extent Brit but there is no arguing both jk2 and rtcw are far more cpu bound at lower resolutions or fsaa-less settings than q3.

benching jk2 or rtcw at 1024*768 and noting no improvement doesnt really mean a thing. Theyre both too cpubound. Testing at a resolution/setting which start to incurr a performance hit due to the video subsystem is where to look for any improvements.

Check out matts comparison over at 3dgpu.com, 50 fps improvement in antialiasing performance.

 

[Hellbinder]

sorry guys.. i meant D3D drivers not openGL drivers...

At any rate... what about it? they almost have to be cheating somehow right? the hardware just does not seem like it sould be able to pull off numbers like that. You should also then see big FPS games in other shader enabled games.. The numbers dont show it.

 

[Bjorn]

]sorry guys.. i meant D3D drivers not openGL drivers...

At any rate... what about it? they almost have to be cheating somehow right? the hardware just does not seem like it sould be able to pull off numbers like that. You should also then see big FPS games in other shader enabled games.. The numbers dont show it.

I agree. They have to be cheating.

Just like Ati

http://www.beyond3d.com/forum/viewtopic.php?t=1922&highlight=catalyst+improvement

I swear it, they're all evil. Eeeeeeviiiil aaaaarrrgggh...

 

[Typedef Enum]

Yeah...Great logic.

The only way possible to improve performance is to cheat.

I'm pretty sure if you were to take any 1st/2nd year Computer Science course, you probably wouldn't dial into any course work which would discuss such topics as...oh, I don't know...complexity, Big-O notation, algorithms, etc.

Quite frankly, I know for a fact that none of the hardware/software guys who browse this forum have ever stumbled across some chunk of code that just screamed inefficiency...fixed it up, and ended up yielding a good performance increase...Or, do that across several areas which, in total, gave a nice boost.

Anyhow, I'm going to concur...there's no other logical explanation as to the performance increase other than cheating.

 

[Hellbinder]

Read my other post that this one is refering to..

I am refering to nature specifically. It is only a question and it *is* logical despite what you sneering Nvnew guys think suddenly getting a 40 FPS increase in Nature *only* and in FSAA in Quale 3 *only* is *very* fishy.

The went from 115 with 4x FSAA in Q3 at 1024x768x32 to over 160 yet non of their other games incresed by more than 4 FPS.

Tell me again how they are not outright cheating somehow in these coupple of benchmarks??

 

[Kristof]

Could we be careful with the word "cheating" and maybe think of it as "game specific optimizations" ? Are there visual issues, are they cutting corners, are things missing ? Or did they just optimise the driver/hardware path for some popular apps ?

Online articles should specify their findings and educate people on how exciting these new drivers are...

K~

 

[Murakami]

Using SeriousSam SE I see no signigicant difference in performance or IQ using the new drivers in anisotropic filtering performance. Here are the results using a static screen shot in the Techdemo of SS-SE. GF3 Ti200 at 240/500 core/mem. This is without MSAA. Below images you can compare the FPS (upper right corner) and IQ from the different in game AF settings. System specs below.

NoAF

2xAF

4xAF

8xAF

In Summary

From:

1. NoAF - 2xAF, 58 - 42 FPS -> 28% drop
   NoAF - 4xAF, 58 - 29 FPS -> 50% deep drop
   NoAF - 8xAF, 58 - 20 FPS -> 66% Plunder!

A pretty much non T&L game like SeriousSam SE shows no significant difference between the new and old drivers, at least for me. Another note is that 3dMark2001se does not predict nor indicate the performance change or lack of performance change in this program. To believe that a change in 3dMark2001se means a change will happen in other programs is obviously a fallacy. You must know what caused the change in 3dMark2001se to be able to determine if another program will reflect similar result changes. To say that 3dMark2001se predicts 3d performance I would say is streching the truth.

SS:SE "is" a TL game..

 

[Nappe1]

My own designed drivers always CHEAT!!!111. they do nothing. I can even release sources, look:

//Nappe's drivers 1.0
//development version
//destroy before reading

uses 
  OP_DDK, Graph;

var 
  Dummy1, Dummy2, Dummy3 : Integer;

//umm... what next?? maybe some structures:
type 
  TDriverzModifierz of Record;
    Key : LongString;
    Value : Boolean;
  end

//Great! now some real code! :)

begin
  // naaah... I code this later. Now some Guake!!111
end.

hey, come on guys... DoomTrooper, HellBinder and alexsok... these are just god damn graphics cards.

(on moments like this, I am really happy that my fav. company haven't delivered.)

 

[no_way]

has anyone benched Codecreatures with those thingies yet ?
I have a naggin feeling, that tis boy will show some massive performance increases as well.

Edit:
well, not really.
http://www.icrontic.com/index.php?page=public/articles&articleId=233
*shrug*, whats so special about the Nature scene then ?
DX8.1 SDK samples could be checked too, if any single one of them show increase in FPS ?

 

[Fuz]

My own designed drivers always CHEAT!!!111. they do nothing. I can even release sources, look:

//Nappe's drivers 1.0
//development version
//destroy before reading

uses 
  OP_DDK, Graph;

var 
  Dummy1, Dummy2, Dummy3 : Integer;

//umm... what next?? maybe some structures:
type 
  TDriverzModifierz of Record;
    Key : LongString;
    Value : Boolean;
  end

//Great! now some real code! :)

begin
  // naaah... I code this later. Now some Guake!!111
end.

hey, come on guys... DoomTrooper, HellBinder and alexsok... these are just god damn graphics cards.

(on moments like this, I am really happy that my fav. company haven't delivered.)

LOL. Let me try those drivers Nappe1.

Fuz

 

[OpenGL guy]

*shrug*, whats so special about the Nature scene then ?

It's worth 20 pts per frame in 3D Mark?

 

[Ailuros]

Wolfenstein is not CPU limited...

It isn't?

 

[K.I.L.E.R]

Is it just me or has the quality of FSAA increased substantially?

 

[BoardBonobo]

Some FSAA stuff levered directly from the drivers:

4X Mixedsampling (skewed grid, 8-tap filter)
4X Mixedsampling (skewed grid)
4X Multisampling (Gaussian)
4X Multisampling
4X Supersampling (skewed grid, 8-tap filter)
4X Supersampling (skewed grid)
4X Supersampling (Gaussian)
4X Supersampling
4X Supersampling (LOD bias)
2X Supersampling (Quincunx)
2X Quincunx
2X Multisampling
2X Supersampling (vertical)
2X Supersampling (horizontal)

Supersampling and Multisampling galore!

 

[ChrisK]

Bah...

Nappe1,

you shouldn't use Pascal to develop your drivers, I mean, really. Visual Basic is sooo much more 1337

Chris

 

[Nappe1]

Re: Bah...

Nappe1,

you shouldn't use Pascal to develop your drivers, I mean, really. Visual Basic is sooo much more 1337

Chris

better than that, I skip that one and go directly to Most Advanced GFX developing language: PC-LOGO.

you can guess it will be lightningly fast.
(if anyone can even recal anymore what was PC-LOGO language.)

 

[madshi]

Re: Bah...

Nappe1,

you shouldn't use Pascal to develop your drivers, I mean, really. Visual Basic is sooo much more 1337

Chris

That should be Delphi, my favorite language.

 

[nevidimka]

could this driver update be a 1st true ex 3dfx programmers input???

 

[noko]

Murakami wrote:

SS:SE "is" a TL game..

That is not what Croteam says. . .

 

[nevidimka]

my god.. look at some of the features.. from the hacked tabs in the new drivers.. most of them u can find on 3dfx tools. i really am thingking theres a 3dfx input in this.. example the texture sharpening which must be Lod bias.. in the AA tab.