Graphics 3D Mark 2003 Scores

System:
MSI K7N2 Delta-L
Barton 2500+ @ stock
1GB Crucial PC-3200 @ 333 MHz (2-2-2-5)
Radeon 9600XT @ stock

Scored 3730

 

:edit:

Oops, Ignore me, I'm an idiot.

:edit again:

Finally managed to find a decent speed server with it on

3541 3DMarks.

Asus A7M266, 512Mb PC2100 DDR, Athlon XP 2400+, Sapphire Radeon 9600XT 256Mb with 256Mb AGP Aperture set, Creative Audigy.

 

ummm.. your project linky is the bork

 

Hey all, New guy to the forum here

Heres my score, im pretty pleased with it


3DMark2k3 Score:5727

 

My ATI 9600 Pro

My cheap little 9600 Pro got good scores overclocked till it almost broke!

(dont worry - it survived )

4451 3DMark03

14660 3DMark01

P42.4C at 3Ghz, 1Gig RAM

Guess thats pretty good for a 9600 Pro (03 score is good, 01's not the best cos my mem timings are holding me back - DDR400 1:1 3-4-4-8 at 250FSB dual channel)

 

Your 2k1 score is still good mike

Your 9600 Pro is doing damn good !
When i first got my 9800 pro i was benching under that. Managed with a little tweaking of my crappy ram to get just over 16000 in 2k1 (2.2.2.6)

 

Erm why are u guys getting such low scores bow down to my 5,500 something score forget and i didn't even finish all the test !!! :O

 

I got about 7.5k, couldn't publish tho because I haven't bothered downloading the latest version on stinky 56k

 

What are you talking about?

ATI 9600 Pros get 3900 ish stock and djmike is getting 4451 which is great for that card.

If you look at j3mZ he is getting 5727 on a ATI 9800 Pro wich is also a good score.

 

What you have to remember is 3D Mark 03 is not totally GPU dependent.

Game Test 1 "Wings of Fury" is CPU dependent, this does effect the final score. (8 % score change is possible, even more on highly OC'ed top end systems).


For 3D Mark 03 the tests are as follows.

Game Test 1 - "Wings of Fury" is DX7.
Game Test 2 - "Battle Of Proxycon" is DX8.1.
Game Test 3 - "Trolls Lair" is DX8.1
Game Test 4 - "Mother Nature" is mostly (not all) DX9.

The games tests are not weighted the same.

Game Test 4 comes out as 20% of the final score 80% is split evenly to between the other three tests.

This show even the semi-DX9 game test 4 is not given an appropate 25% weighting.

If you want to know more read this very good technical article.

http://www.beyond3d.com/articles/3dmark03/intro/

 

3Dmark03 Score of
5207


2.4 GHz Celeron / Intel D865GLC board
512 MB Crucial PC2700
EVGA 5900se gfx card

 

Hmmm... seems a bit pointless this, as no standad is set... If I run the tests with the card's default settings, I get scores of around 5800, but with everything maxed, it drops to around 2000. Maybe you need to specify what settings the card should use, or publish results from ALL settings?

Mr G

 

okay guys i wanna upgrade asap someone tell me what card would i be best with and if poss the best possible prices (uk) thanks

 

If you are in the UK then the www.overclockers.co.uk Sapphire Atlantis or PowerColor ATI 9800 Pro are approx £178.

Both cards are the same, both are modeled on the ATI reference board.

 

Erm, it's a benchmark, so if you're looking for a number, you leave everything on default settings. That's what everyone else does.

 

I would say think of a benchmark like this:

It's there for you to get an idea of how your system performs and how changes to your system will increase/decrease that performance. It's not a competition, because you will never win. Someone will always go faster. Basically, you know how your system performs right now in games, video editing, whatever. So you benchmark it and see what your score is and how it compares to similar setups (not to try to beat them but to make sure you are getting the performance you should). Then you want to get a little more oompf out of it. So you tweak and then re-bench it.

I see no point in doing what some folks do and trying to get the absolute highest score regardless of the consequences. People will turn their display colors down to 16 colors. Turn off everything. Stop every service possible. And then crank up the cpu, ram and video card to their absolute maximum (which probably isn't fully stable by the way). Then they will run the benchmark with a crappy picture (which may even have artifacts) just to get 9000 points in 3dMark03....but they would never be able to actually play games or anything with the system like that. So what's the point?

Use a benchmark tool as just that...a tool to measure changes in your setup. And use it to test your system how you plan to use it.

</rant> Sorry for that.

 

3D Mark 01 and 3d Mark 03 has an adopted standerd, you should use these to compare you system to others:

1024x768x32 BPP.

No Antialiasing.
No Anistropic filtering.

Direct 3D Vertical SYNC is Off.

 

3745 at stock

http://service.futuremark.com/compare?2k3=2043683

 

Well, thought what the hell and just tried to softmod my Herc 9800se AIW..... successfully!!!

Date: 2004-02-11
CPU: AMD Athlon(tm) XP/MP/4 1402 MHz
GPU: ATI RADEON 9800 PRO @ 378 MHz / 338 MHz
OS: Microsoft Windows XP
Res: 1024x768@32 bit

Score: 5134.



Damn I'm pleased.
Score obviously being dragged down a bit by my slow-ass CPU, but might review that situation in the not too distant future.

Still not clocked the card yet either, so will try that in a bit, see what sort of scores I can get. I ran the test before I used the softmod drivers and only got 1209marks - bit worrying - but I realised after installing the new ones I had 6xAA and 4xAF forced on.

 

Look Ma, I wanna look like Michael Jackson!

check that ****, thats how its really done

http://www.btinternet.com/~xray_ted/pc.JPG

shamone

0wned

(EDIT: Now I owned you, next time please make images of that size a link)