3DMark03 scores

[Juggalo_Dan]

Grand Theft Auto: Vice City for the pc uses dx 9 :D
Im just about to reboot and test my system...
Motherboard: Trigem Cognac: wopping 100mhz fsb :D
RAM: PC-100 320mb (256+64=320)
Video: 3D Prophet 9000 PCI 64mb
Sound: Hercules Muse LT
CPU: Wopping 733mhz celeron @ 906.5 mhz (note the .5 [very important :D]makes all the difference)

[Juggalo_Dan]

ok i got a wopping 544 marks lmfao
BTW the core and ram clocks are locked on my video card FROM THE DAMN CORE so i cant o/c it :cry: 250/200

[MFaction]

I got 3315. xp1600+ w/ radeon 9700 and 512 elixer ddr

[MFaction]

i changed some **** in my bios and got 4131. with 4x AA and 16xAF

[RDR]

I just tried the following setup:

P4 3.06
Asus P4C800 i875
1024MB DDR333
Radeon9700Pro

4933 with no OC'ing

unfortunately i875 will not let you run the memory at DDR400 with a 533FSB

....just tried the same setup with a Visiontek 9800pro

5563 with no OC'ing

:D :D :D

[weta]

3DMark03 (build 330) Info



3DMark03 v3.3.0 is an updated version of 3DMark03. Hardware review sites discovered deliberate cheats in some drivers. These drivers identify 3DMark03 build 3.2.0, and render the tests differently than 3DMark03 instructs, in order to gain additional performance. Build 3.3.0 has been changed so that the test results remain the same, but the questionable drivers do not identify 3DMark03 anymore. The drivers now think 3DMark is a 3D application among others, and render the tests like 3DMark instructs. This produces a result that is genuinely comparable to other hardware. There is a command line option "-allowdocumentedcheats", that runs 3DMark03 like it used to, and enables the cheats in the questionable drivers. You can try this switch to see the performance difference the cheats give.

Futuremark encourages everyone to download the latest patch (330) to ensure comparable results.

Note: 3DMark03 results from the original version (313) are identical with the results of this new build (330), and those of the first patch (320), except when using certain drivers with built in cheats.

3DMark03 Patch (build 330) Improvements

Fixes & Updates

Graphics card core and memory speed detection corrected under Win98/SE/ME.
Updated graphics card core and memory speed detection

Important Notes

Benchmark results from the original release of 3DMark03 (build 313), 3DMark03 (build 320) and 3DMark03 (build 330) are fully comparable.
3DMark03 (build 330) is currently only available as a patch version.

The registration codes for 3DMark03 (build 313 and build 320) also work for 3DMark03 (build 330).

Download 3DMark03 build 330 patch

[Anti]

edit:
Next time I went back it was no longer greyed out.

_____________________
Epox 8RDA+
XP1800+ @1713MHz 11x155
2xKHX3000/256@333
Gigabyte 9000pro 64mg OEM @ 317/315

[MFaction]

DOES THIS EXPLAIN why I: 1st ran 3dmark2k3 and got around 4000, then i tweaked and got like 4200, a few days later I GOT 1500... wtf i tweaked and got 1560ish. then i updated catalyst to the new release and got 2400. wtf what happend to my 4k score?
im going to format and start from scratch this weekend

[weta]

On May 15, ExtremeTech presented the test results of its investigation of nVidia's performance on 3DMark2003. We suspected nVidia was falsely optimizing for the benchmark with its new GeForceFX 5900 Ultra product. On May 23, Futuremark confirmed our suspicions with a detailed report on extensive "driver cheats" by nVidia along with some ATI anomalies. At the same time, Futuremark released a new version of the benchmark that defeats nVidia's detection/special-casing techniques.

Based on that information, we quickly took Futuremark's latest version of 3DMark2003, which is designed to negate those cheats, and put the products to our own tests. This story details our independent analysis of all eight false nVidia optimizations revealed by Futuremark, along with ATI's and nVidia's scores using this latest 3.30 version of the 3Dmark2003 benchmark. We believe, finally, that this is an accurate representation of relative performance between nVidia's top-of-the-line GeForceFX 5900 Ultra and ATI's Radeon 9700 Pro. Read on for the whole truth.
ExtremeTech

ExtremeTech article

[weta]

<div align="center"><img src="http://forums.tweaktown.com/attachment.php?s=&postid=152200"></div>

Futuremark

"Recently, there have been questions and some confusion regarding 3DMark 03 results obtained with certain Nvidia" products, Futuremark said in the statement. "We have now established that Nvidia's Detonator FX drivers contain certain detection mechanisms that cause an artificially high score when using 3DMark 03."

Nvidia, (un-named employee)

"Since Nvidia is not part of the Futuremark beta program (a program which costs of hundreds of thousands of dollars to participate in), we do not get a chance to work with Futuremark on writing the shaders like we would with a real applications developer," the representative said. "We don't know what they did, but it looks like they have intentionally tried to create a scenario that makes our products look bad."

Beyond3D, (Dave Baumann)

Despite still being a full Futuremark Beta member, ATI did not make it out of the report entirely unscathed either. There is a performance difference of about 8% in Game Test 4, that accounts for about a 2% difference in the final 3DMark03 score, between the new and old versions, indicating that although not visually different something was occurring on this particular test.

ATI, (Chris Evenden, PR Director)

The 1.9% performance gain comes from optimization of the two DX9 shaders (water and sky) in Game Test 4 . We render the scene exactly as intended by Futuremark, in full-precision floating point. Our shaders are mathematically and functionally identical to Futuremark's and there are no visual artifacts; we simply shuffle instructions to take advantage of our architecture. These are exactly the sort of optimizations that work in games to improve frame rates without reducing image quality and as such, are a realistic approach to a benchmark intended to measure in-game performance. However, we recognize that these can be used by some people to call into question the legitimacy of benchmark results, and so we are removing them from our driver as soon as is physically possible. We expect them to be gone by the next release of CATALYST.

Tim Sweeney, (the main man behind the Unreal Engine)
The difference between Driver cheating, and Driver optimizations.

Pixel shaders are functions, taking textures, constants, and texture coordinates as inputs, and producing colors as outputs. Computer scientists have this notion of extensional equality that says, if you have two functions, and they return the same results for all combinations of parameters, then they represent the same function -- even if their implementaions differ, for example in instruction usage or performance.
Therefore, any code optimization performed on a function that does not change the resulting value of the function for any argument, is uncontroversially considered a valid optimization. Therefore, techniques such as instruction selection, instruction scheduling, dead code elimination, and load/store reordering are all acceptable. These techniques change the performance profile of the function, without affecting its extensional meaning.

ExtremeTech, (3DMark03 [Build 330] test conclusion)

The real irony of all this "cheating" is that our basic impressions of ATI's and nVidia's latest cards hasn't really changed: Both cards are essentially neck-and-neck in performance, and until we have real shader-based games, our opinions likely won't change.

John Carmack (Main programmer, ID Software) NEW

Rewriting shaders behind an application's back in a way that changes the output under non-controlled circumstances is absolutely, positively wrong and indefensible.

Rewriting a shader so that it does exactly the same thing, but in a more efficient way, is generally acceptable compiler optimization, but there is a range of defensibility from completely generic instruction scheduling that helps almost everyone, to exact shader comparisons that only help one specific application. Full shader comparisons are morally grungy, but not deeply evil.

The significant issue that clouds current ATI / Nvidia comparisons is fragment shader precision. Nvidia can work at 12 bit integer, 16 bit float, and 32 bit float. ATI works only at 24 bit float. There isn't actually a mode where they can be exactly compared. DX9 and ARB_fragment_program assume 32 bit float operation, and ATI just converts everything to 24 bit. For just about any given set of operations, the Nvidia card operating at 16 bit float will be faster than the ATI, while the Nvidia operating at 32 bit float will be slower. When DOOM runs the NV30 specific fragment shader, it is faster than the ATI, while if they both run the ARB2 shader, the ATI is faster.

When the output goes to a normal 32 bit framebuffer, as all current tests do, it is possible for Nvidia to analyze data flow from textures, constants, and attributes, and change many 32 bit operations to 16 or even 12 bit operations with absolutely no loss of quality or functionality. This is completely acceptable, and will benefit all applications, but will almost certainly induce hard to find bugs in the shader compiler. You can really go overboard with this -- if you wanted every last possible precision savings, you would need to examine texture dimensions and track vertex buffer data ranges for each shader binding. That would be a really poor architectural decision, but benchmark pressure pushes vendors to such lengths if they avoid outright cheating. If really aggressive compiler optimizations are implemented, I hope they include a hint or pragma for "debug mode" that skips all the optimizations.

Futuremark, (Patric Ojala, 3DMark Producer) Responding to earlier comments by an "un-named" Nvidia employee.

First of all, you can join the Futuremark beta program, if you are in some way a significant player in the PC industry and you pay the membership fee of $5000 USD / year.

And second, Nvidia was a member of our beta program, got pre-release builds, shader source code, and participated actively in the development of 3DMark03. After they left our beta program, there was mostly just bug hunting and product quality assurance left.

I don't see how they could 'not know what we did, in order to intentionally try to create a scenario that makes their products look bad'.

[AgentX]

Whoa where did whoever get that 10k score from?? Is it really real the most ive ever seen b4 is around 6k

[maximus7001]

Well i wonder why it thinks i have a pentium II?????
From what i have seen, 1245 is a decent score and thats not at full oc yet.compare 3d mark 03

[weta]

Cool movie poster inspired by the Ang Lee film"Crouching Tiger Hidden Dragon" See my previous post

[minibubba]

Cool movie poster inspired by the Ang Lee film"Crouching Tiger Hidden Dragon" See my previous post

ROFL
ya gotta love that :thumb:

[romi427]

Got a nice score with my setup :p

3DMARK03: 4953

Barton xp2500+ @ xp3200+ 200fsb//11multi
Asus A7N8X
Radeon 9700Pro
Seagate Barracuda SATA 120gig