Graphics 3D Mark 2003 Scores

1605

1800XP @ 1725MHz (150x11.5)
512MB PC2100 DDR PNY-crap RAM
GeForce4 Ti-4400 @ 290/595
Audigy Platinum.
Epox 8k5a3+

Benchmarks are designed to kill every system out there to make people think they have to upgrade.

I am VERY disappointed with futuremark. To have Best Buy and Alienware as sponsors (notice those ads) and force you to pay to even adjust 16 bit color or 32 bit color depth, means they're TRYING to make you run to the sponsors stores to buy upgrades. And in the process, they're making it to where you can't SEE whether you'd be ok with 16 bit color or not, unless you pay THEM money.

********
That starst with a B
-Jason

 

Should have my new setup tomarow or the day after to improve on my 3954 score

 

Yeah I just oc'ed my cpu from 2.4 to 2.9 and it didn't even give it any real gain in pnts.... so i guess this is only gpu dependent and not like the previous 3dmark where I could gain 1,000 pnts or more by oc'ing the cpu.....

 

Grr, I get over 12000 points on 3dmark 2001, and to get 1615 on 03, it is humbling

I'm running a p4 2.4@2.6, with 256 megs ddr-400 and a GeForce 4 4600 Ti.

Time for some serious overclocking, could you hand me that LN2

 

Don't worry too much about your humbling experiences

There have been severe criticisms from nVidia about 3DMark03

Taken from http://www.hothardware.com/


NVIDIA:
"3DMark03 combines custom artwork with a custom rendering engine that creates a set of demo scenes that, while pretty, have very little to do with actual games. It is much better termed a demo than a benchmark. The examples included in this report illustrate that 3DMark03 does not represent games, can never be used as a stand-in for games, and should not be used as a gamers’ benchmark."

The statement above is a direct quote taken from NVIDIA's report on 3DMark2003. This is not a "new" argument, however. We here at HotHardware get statements similar to this one sent to us after virtually every video card review we write. There were, are and always will be a group of people that put very little stock in synthetic benchmarks. Throughout the report, NVIDIA makes a case against all the "game" modules that comprise 3DMark2003...

NVIDIA:
"Unfortunately, Futuremark chose a flight simulation scene for this test (game 1). This genre of games is not only a small fraction of the game market (approximately 1%), but utilizes a simplistic rendering style common to this genre. Further, the specific scene chosen is a high altitude flight simulation, which is indicative of only a small fraction of that 1%."

"For all intents and purposes game tests 2 and 3 are the same test. They use the same rendering paths and the same feature set. The sole difference in these tests appears to be the artwork. This fact alone raises some questions about breadth of game genres addressed by 3DMark03. --- These two tests attempt to duplicate the “Z-first” rendering style used in the upcoming first-person shooter game, “Doom 3”. They have a “Doom-like” look, but use a bizarre rendering method that is far from Doom 3 or any other known game application."

"Finally, the choice of pixel shaders in game tests 2 and 3 is also odd. These tests use ps1.4 for all the pixel shaders in the scenes. Fallback versions of the pixel shaders are provided in ps1.1 for hardware that doesn’t support ps1.4. Conspicuously absent from these scenes, however, is any ps1.3 pixel shaders. Current DirectX 8.0 (DX8) games, such as Tiger Woods and Unreal Tournament 2003, all use ps1.1 and ps1.3 pixel shaders. Few, if any, are using ps1.4."

"This year’s 3DMark has a new nature scene (game 4). It is intended to represent the new DirectX 9.0 (DX9) applications targeted for release this year. The key issue with this game scene is that it is barely DX9."

As you can see, NVIDIA takes issue with almost every aspect of Futuremark's latest benchmark. Why would they do this? I'm sure time will tell. Unfortunately for us, we haven't spent enough time with 3DMark2003 to have a truly educated opinion. My hypothesis would be that NVIDIA's legacy hardware lacks true support for ps1.4, which means performance will suffer in this benchmark as ps1.1 will be used as a fallback version rather than ps1.3...the version most NVIDIA hardware supports. The GeForceFX seems to perform well in 3DMark2003, but until we have a card of our own to test we can't really say for sure. It's the GeForce 4 an older product lines that 3DMark2003 may shine a negatice light on. After reading this report, we asked the folks at ATi what they thought of 3DMark2003, and they seemed pleased overall with Futuremark's latest.

NVIDIA:
"So, where do you find a true gamers’ benchmark? How about running actual games? Most popular games include a benchmark mode for just this purpose. Doom3, Unreal Tournament 2003, and Serious Sam Second Encounter are all far better indicators of current and upcoming game performance."

Pretty strong words, huh? It seems to us that this is a bit of damage control on NVIDIA's part, but we're sure there are going to be some that will totally agree with NVIDIA's stance on this situation. Why wouldn't 3DMark2003 fallback to ps1.3, if support for ps1.4 wasn't present? Interesting question.


The benchmarks does put nVidia in a bad light as even their Ti4600 cards are not scoring above 2K. Only their GFX cards are slightly ahead of the R9700 Pro.

However, I do agree with some of the criticisms - even HardOCP also said they would not be using this as a video benchmark

HardOCP has also made similar observations
http://www.hardocp.com/article.html?art=NDI4LDQ=


"3DMark03, a DirectX 9 Benchmark?" This is one question that we really must ask ourselves. Game 1 is a very simple DX7 test that is not representative of any current games. Game Tests 2 and 3 are DX8.1 but use Pixel Shader 1.4, which is not used by any games we are aware of and will not be to our knowledge. Game 4 is a hybrid of DX8/DX9. It is these four tests that determine the overall score. Only one game test in this benchmark DirectX 9 and then only partially.

Is 3DMark03 really a good indication of what “Real World” gaming is? Is 3DMark03 really "forward looking"?

With all that we have seen so far it does not appear to actually give us any indication how video cards are going to compare in any real games. It produces an overall 3DMark which is taken from unbalanced game tests. Furthermore as we have seen directly above in the benchmarks, drivers can be optimized to run the game tests faster. The problem is this is just a benchmark and not based on any real world gaming engines. Therefore while you may get a higher 3DMark number you will not see ANY increase in performance in any games that are out there.

One issue we have been discussing as of late with amoungst ourselves and with ATI and NVIDIA, is the implementation of real world game benchmarking versus synthetic benchmarking. We do feel that synthetic benchmarks do have their place if used properly. Actual games or gaming engines should remain the primary focus though. When it comes right down to it we would rather see improvements and gains in real world games rather than in synthetic benchmarks.

In closing, Kyle has informed me that [H]ard|OCP will not be using the overall 3DMark03 score to evaluate video cards.


Another reader at Hothardware has made an interesting observation

3DMark03 is no longer using a real game engine, instead it is using wrappers into DX. No game is ever written like this, so any score from this benchmark means nothing. The reason for low scores with the benchmark is not because your hardware is slow but because the wrappers are. Example Game Test #1 is DX7, there is nothing in it to really cause such slow framerate on current hardware. Its just using the old T&L DX7 code paths which are so slow no one ever used them. This benchmark is a joke

I should think that is a valid observation since why should this simplistic test using low polygon should have such low fps?

 

Its looking more and more from the scores that 3dmark 2003 is purely graphics card dependent.

You could use the old one to get a general idea of how fast the whole system was running, small changes to different aspects of your pc made a difference to the score.

Ahh well it will be tonight before I get a chance to run the test - heres waiting.

 

1713

 

1155 is the best I can do for now - it's def. GPU/fill rate limited

Compare

I will have to try it out on my work system (P4 2.0, GF4 4200)

 

I bet my work computer gets a zero

 

718...

My rig:

Pentium 3 850MHz (BX board)
512Mb RAM
Matrox Parhelia

Need to try this on my Dad's PC later on...

Hopefully the DX9 drivers for Parhelia will be out soon...

 

4855 here, need to tweak a lot One of my mates at home is gettin 5.7k...

 

1021 on my 8500 running on a Abit-kt7aRaid with 768mb sdram and a XP 2000+ chip. Seems like you NEED a dx9 card for this or it's pointless although I can see what they are trying to acheive with all the tests (Comparing accelrated against cpu only, and with sould etc.)

 

I hate it when things like this are released, as they are not "really" any use to people as to how "good" their computer is at running games... I wonder how many of you will go out and buy a new graphics card, *just* to get a higher 3dMark score...

I still run an original GeForce3, which I bought in March/April 2001, it runs everything I throw at it with a high level of competency... I'm sure futuremark/madonion are getting a cut from the graphics card sales that they generate with their benchmarking program that (as stated above) is not actually true to gaming.

I'll run the benchmark later, but, I doubt I'll get anything more than 1.5k, is there actually any games out that are DX9 at the moment? nope so 3Dmark2001SE is probably a better indication of how fast your PC is for games of "today"

Buy a new graphics card when a new game comes out that won't run on your current one, not just when you're upgrading for the sake of getting a better 3Dmark score. </rant>

 

YESSSS!!!

Update to new drivers and got.... wait for it...... an increase of 3 (yes, three) points!
I ROCK!


As for all this whinging about the damn benchmarking. I see it like this;
3D Mark 2003 may not be (as stated above) the most reliable benchmark for an indication of computer performance for real gaming. BUT, look at it this way. It is a 'standard'. It may not represent today's, or true, gaming very well, but it does give everyone a uniform test, and result. So, if you look at it like that, then yes, it does give a fair test across all hardware, because all 9700 users are getting roughly the same, all Ti4200 users roughly the same, etc etc. So it is a benchmark that can be compared, and hence, an evaluation can be made. It seems this one is very processor independant, so gives a clearer indication of graphics card comparisons even though it may not represent how well your card will perform in real games.

Just my two bob.

 

Now what you want is a benchmark that

Has completely seperate and highly optimized full game engines for Direct X 3 | 4 | 5 | 6.1 | 7 | 8.1 | 9 and Open GL 1.2 | 1.3 | 1.4 | 2.0

That ran through at least 5 tests for each of those, the tests would stress test each component in the system and make sure the system was taken as a performance whole

That ran through resolutions in each test
320x240 | 640x480 | 1024x768 | 1280x1024 | 1600x1200 | 2048x1536

That tested the card at its highest possible quality settings, and also tested out all its AA and Ansio settings at each of those resolutions.

========================

Stopping at the systems limit, the more tests it is able to run and the faster it runs them the better the score... Could be called something like Pure Game 24, because it would probably take 24 hours to run it, ! Certainly put pay to those LN2 cooled systems getting ridiculously high results for 5 mins then blowing up.

 

4918 Not too bad for starters
http://service.futuremark.com/compare?2k3=13436

I think that puts me in 3rd behind Ole Drunkenmaster and DaveMustaine
That was my first and only run as I was busy last night, so no tweaking yet though, I will have it over 5K when I get back tonight, then I'll be happy!

 

946 on the first run then o/ced my gf3 a bit and d/led new drivers and got 1256, interesting

XP2200 @ 2000 (166*12)
Abit KX7-333
512mb DDR333
GF3 o/c'ed quite a bit but i cant remember what to

 

I havent slept in 3 days now and last night I was supposed to catch up, then I had to d/l this. I was up till 3:30 waiting for it (I didnt see B-T was mirroring it... =\ I waited for it to finish at 18k/s while I coulda used all the speed B-T could serve up). I couldn't keep my eyes open for the test =(. I do know that I got a 4284 on my first and only run. I woke up with my keyboard imprinted upon my face and saw that as I rubbed my eyes. Based on the other scores floating around... doesnt seem to bad. Ill do some tweaking when I get home but im stuck in school ATM.


What I do find quite interesting is this...

I dont score to great ok 2001 for some reason... I think I get about 11-12k. I think this shows very little correlation between 2001 and 2003. This makes me wonder how much that "Game 4" nature test counts for. It really appears to be quite alot.

Anyhow... Im going to upgrade my drivers, tweak some memory settings, and then repost. Overclocking is after that with my final post. I unfortunatley am too poor for my watercooling (still), so that last part wont help too much.

Specs:

Athalon XP 1700 (stock speed)
512 MB DDR PC2100
Abit KG7-Lite
Radeon 9700 (Stock for now)
WD1200JB (I cant wait to overclock this thing! )

 

I always got great 3dmark 2001 scores, I could have pushed to around 12500 or more if I had wanted to, but I didn't, as the card was getting quite hot being OC'ed to where it was already.

I have come to the comclusion that ATI must have put their money in the right pockets, because this benchmark's results are skewed towards ATI, as Nvidia does not yet have a DX9 card on the market.

If you look at the score diffrence between me and a similar system with a ATI card, the score is doubled, if not trippled!

I have no doubt that it will be a useful tool for the future benchmarking of graphics cards, but right now, it is completely usless to me

 

1029 with a ti4200

seemed to jutter and stay on 2-3 fps thru out everything but the first test