Interesting 3Dmark 2005 scores

Asus

Newbie
Joined
Aug 22, 2003
Messages
10,346
Reaction score
0
Link
WE HAVE SOME numbers that we would like to share with you before Futuremark finally introduces this long awaited test.

With current drivers, Nvidia will score 5000 points with the 6800 Ultra while the X800XT PR will score only 4500 with 4.9 drivers. With the new ATI driver codenamed 8.07, ATI score will jump to an incredible 5800. All these scores are on a 3.4GHz Pentium 4 machine. The ATI driver will be introduced as the Catalyst 4.10 and we learned it should get its WHQL very soon.

The Geforce FX 5950 Ultra will score a modest 1270 while the Radeon 9600XT will provide 1570 marks. The mainstream card kicks the former high end in its performance nuts.

The Geforce 6800 GT will score 4500.

You will be able to see many more scores on the 29th of this month at 3PM CET when Futuremark lifts the non disclosure agreement (NDA) veil. µ
 
B.Calhoun said:
Fek me. Fek my 9800pro.
Your 9800pro would be better than the 9600xt...
Remember the FX is terrible at DX9.
 
time to go back to 4 digit scores for a while.

wonder what 9800pros will get. must get my 6800gt soon
 
That new catalyst driver boosts teh X800XT PE's score from 4500 to 5800! And that's just a beta version! Wow. I'm really impressed with that. Is that the driver where they were revising the OpenGL codepath?
 
um no.
3Dmark is a DirectX benchmark. ;)
 
About that Cat 4.10 driver, will you get "lower" graphics when you have that AI thing on? Or will graphics remain the same but just get you more fps?
 
Should read the reviews. Same graphics, same quality.
 
Asus said:
um no.
3Dmark is a DirectX benchmark. ;)

Yeah I knew that but I was just wondering if the drivers also had the OpenGL thing included as well as the AI. Wow if these drivers really are that good maybe I'll end up getting an X800 pro for christmas.
 
Nex321 said:
About that Cat 4.10 driver, will you get "lower" graphics when you have that AI thing on? Or will graphics remain the same but just get you more fps?

:| Catalyst AI increases performance quite a lot IMO. I think it uses a method to switch back and forth between Bilinear and Trilinear filtering, correct me if I am wrong. Safe to say there's no image quality loss, cause well with the naked eye, it's hard to tell.

Anyways, 4.10's sound freakin amazing. Let's hope that the OpenGL improves with the new 4.10's. I cannot get over that man a while 1300 points with just a driver update, remarkable.
 
well, more reason why 3dmark is a crap test.

No offense to the driver writers at ATi, but unless the new driver opens up new pipelines, there shouldn't be a ~30% increase in performance of the card.

I can almost guarrentee you that you won't see a 30% increase in performance in your games with the new drivers, regarless of what 3dmark05 says.
 
I thought the drivers for the x800's were very mature concidering ati has had a year or so to refine them because they are using roughly the same card they were last generation. But from what I have heard that people are experiencing less glossy textures with ati's new drivers in doom 3, but on the upside it runs faster.
 
Well you know what, screw 3DMark, let's wait until they officially are released and are benched in all the latest games. If there's no loss in IQ and ATi isn't cheating it's sure looking good for them...
 
Couple of things to end confusion.

- Catalyst AI will not affect synthetic benchmarks like 3D Mark 05.
- Even though the X800 series is similar to the 9800 series, they are not as closely related as you expect them to be. The 6800 series is as closely related to the FX series as are the respective ATI cards. You should still expect driver increases from the X800 hardware as well as 6800 hardware.
- Synthetic benchmarks have to make an assumption on what future games will be like. 3D Mark 03, hedged their bets with Doom 3 like technologies (stencil shadows, etc) and that explains why Nvidia cards did extremely well in 3D Mark 03. So don't take performance numbers from 3D Mark and expect them to equate to the same real world performance.

And why 'o why didn't they give an advanced score for the X800 Pro!
 
:) Probably because the guys at ATi couldn't get the new coding to work with the x800 Pro.
 
blahblahblah said:
- Synthetic benchmarks have to make an assumption on what future games will be like. 3D Mark 03, hedged their bets with Doom 3 like technologies (stencil shadows, etc) and that explains why Nvidia cards did extremely well in 3D Mark 03. So don't take performance numbers from 3D Mark and expect them to equate to the same real world performance.

And why 'o why didn't they give an advanced score for the X800 Pro!


which benchmarks are you looking at? you must be talking about the 6800 series, and you must been "doom 3 like technologies" in a broad sense.... ogl and dx are not so similar....
 
SidewinderX said:
which benchmarks are you looking at? you must be talking about the 6800 series, and you must been "doom 3 like technologies" in a broad sense.... ogl and dx are not so similar....

The fundamental graphical features are the same, regardless of OpenGL or DX. A normal map is a normal map, regardless if the engine is written in openGL or DX. The same goes for stencil shadows, dynamic lighting, etc.

3D Mark 03 used stencil shadows (something that the FX series and 6800 series can do in spades because of its architecture) in 2 of the 4 tests. Stencil shadows is one of the key graphical features that is touted in Doom 3. Another thing 3D Mark 03 focused on was dynamic lights, again another Doom 3 touted feature. Only one test (mother nature) used a respectable amount of shaders, but that test used primarily DX 8.1 shaders (running under the pretense of DX 9). The people at futuremark did not expect shaders to be heavily used in games like Far Cry and HL2 so they did not anticipate that.

It doesn't matter what 3D Mark 03 was written in, it still tests the graphic cards ability to render different things.
 
one reason why the 6800u beats the x800xt is the fact that SM3.0 tests are included in 3dmark05
 
[Matt] said:
one reason why the 6800u beats the x800xt is the fact that SM3.0 tests are included in 3dmark05
Ah but since you read it you know that the X800XT will beat the 6800U by a good margin after the updated drivers. ;)
 
ailevation said:
:| Catalyst AI increases performance quite a lot IMO. I think it uses a method to switch back and forth between Bilinear and Trilinear filtering, correct me if I am wrong. Safe to say there's no image quality loss, cause well with the naked eye, it's hard to tell.

Anyways, 4.10's sound freakin amazing. Let's hope that the OpenGL improves with the new 4.10's. I cannot get over that man a while 1300 points with just a driver update, remarkable.

Yupp that's right. It's strange that nobody has thought of that method before. It's gonna kick Nvidia's ass though (at least until nvidia starts using a similar technique)

I'm looking forward to the release :) A few extra fps in HL2 doesn't hurt anybody.
 
[Matt] said:
Optimisations ;)

*cough*cheats*cough*
hehe
If only that were the case.
Although ATI's optimizations are not used in 3Dmark. And futuremark is a stickler when it comes to driver certification.

Plus nvidia's current drivers, the drivers I'm running, have yet to be WHQL certified. :)
Common nvidia...
 
Wait, is this system supposed to get 4800 points:

X800 XT PE
1024 Mb ram
3.4 ghz

How come I only got 4175?

with this system:

1024 mb ram
X800 XT PE
[email protected] ghz

I sure can overclock my cpu to 3.4 ghz but thats not going to help to get 625 extra points. Whats up with this?
 
The never said what beast the were running and we don't know how 3Dmark2005 scales with system hardware. I'd expect a good number of reviews in the comming weeks.

I'd like to see what blahblahblah gets as well.
I'll have to downloud and run it when I get home.
 
My question right now is, do these new drivers bring the X800 pro above the 6800GT?
 
Back
Top