Very small jump in 3D05 being at 3.1Ghz

[WA261]

3.131 is all I can get stable at in 05. I can run anything else at 3.245. This only bumped my score from 7568 to 7714 on an X800XT/PE. I would have thought more. The 7568 was on a 2.8 clawhamer. I guess 05 is very GPU/VPU oriented.

 

[NokiaDude]

No shit, 3d05 is a GRAPHICS CARD test.

 

[imported_DaveA]

3dmark05/03 are gpu limitted. to see a difference use 3dmark01.

 

[fbrdphreak]

I've been out of the Venice seen, 3.2GHz!!! :Q :Q :Q :Q :Q I can't imagine a chip clocked up there w/such a high IPC. Nice!! :thumbsup: Go AMD

 

[imported_Kensai]

Phase change will take anything up very high. Pentium 4's can reach 6GHz with cascaded phase change.

 

[ddogg]

Originally posted by: NokiaDude
No shit, 3d05 is a GRAPHICS CARD test.

yup

 

[Lonyo]

Originally posted by: NokiaDude
No shit, 3d05 is a GRAPHICS CARD test.

Wrong.

3DMark05 is a premium benchmark for evaluating the latest generation of gaming hardware

Hardware means RAM, CPU, GPU etc.
The fact that it FAILS to be a hardware benchmark and is only a graphics benchmark is its major flaw, but it didn't set out to be a graphics card benchmark. It just sucks major ass.

So yes, your score barely changed because 3D Mark sucks and isn't a true representation of overall system performance, unless you use 01.

 

[Munky]

Originally posted by: Lonyo

Originally posted by: NokiaDude
No shit, 3d05 is a GRAPHICS CARD test.

Wrong.

3DMark05 is a premium benchmark for evaluating the latest generation of gaming hardware

Hardware means RAM, CPU, GPU etc.
The fact that it FAILS to be a hardware benchmark and is only a graphics benchmark is its major flaw, but it didn't set out to be a graphics card benchmark. It just sucks major ass.

So yes, your score barely changed because 3D Mark sucks and isn't a true representation of overall system performance, unless you use 01.

Yeah, but for gaming the video card still plays the biggest role as far as performance goes, unless you're running a 1.6ghz celeron, in which case you're hopeless. The latest games tend to use high polygon models, high res textures, lots of shaders, and additional effects like normal mapping, shadows, and such. All this puts a big strain on the gpu, and given that you have a decent A64 or P4 system, the video card will be the deciding factor if you can use all the eye candy or not. Even a 2+ ghz Barton can still run the latest games at max settings if you have a high end video card. On the other hand, you could have a 3 ghz A64, but if your video card is a gf4 mx, you will not run games very well at high settings.

The point is, modern games tend to need a high end gpu to be played at high settings, and so it's appropriate to have a benchmark that stresses the gpu the most.

 

[WA261]

^ Agree.

I will try 01 and see what happens. How about 03?

 

[jdogg707]

3DMark 05 and 3DMark 03 were both much less dependent on the overall speed of the machine opposed to the speed of the grpahics card itself. 3DMark 01 on the other hand was more of an overall system benchmark, as an overclock in the CPU speed could increase the score by a very wide margin, wider even than overclocking your video card. If you OC the card in 03 or 05, you will notice a much more significant change opposed to an OC in RAM or Processor speed.

 

[Lonyo]

Originally posted by: WA261
^ Agree.

I will try 01 and see what happens. How about 03?

03 and 05 both suck as overall system benchmarks. 01 is the only vaguely useful one.

 

[Lonyo]

Originally posted by: munky

Originally posted by: Lonyo

Originally posted by: NokiaDude
No shit, 3d05 is a GRAPHICS CARD test.

Wrong.

3DMark05 is a premium benchmark for evaluating the latest generation of gaming hardware

Hardware means RAM, CPU, GPU etc.
The fact that it FAILS to be a hardware benchmark and is only a graphics benchmark is its major flaw, but it didn't set out to be a graphics card benchmark. It just sucks major ass.

So yes, your score barely changed because 3D Mark sucks and isn't a true representation of overall system performance, unless you use 01.

Yeah, but for gaming the video card still plays the biggest role as far as performance goes,
The point is, modern games tend to need a high end gpu to be played at high settings, and so it's appropriate to have a benchmark that stresses the gpu the most.

With most LCD's being 1280x1024, that becomes debatable.
Here, Half Life 2 is CPU bound at 1024x768 w/4xAA8xAF
BUT: at 1600x1200 with same sparkles, it's GPU bound. In between, at the 1280x1024 native resolution of most 17/19" LCD's, it will probably vary, but both will come into play I would assume. And remember, those tests were run with an FX55 and a P4 EE, so with "average" CPU's like 3+GHz P4's or 3200+ Athlons, it will be even more CPU bound, even at moderate resultions with bells and whistles.

 

[bjc112]

Originally posted by: WA261
3.131 is all I can get stable at in 05. I can run anything else at 3.245. This only bumped my score from 7568 to 7714 on an X800XT/PE. I would have thought more. The 7568 was on a 2.8 clawhamer. I guess 05 is very GPU/VPU oriented.

What chip do you have?

What voltage?

 

[fstime]

ATI runs 3dmark so well.

You almost have 8K in 05 LOL.

I have never seen a 6800gt do that.

If you go to XS.org they have x850PEs with 18K in 03 and ~8k in 05.

Then again, we're talking about heavily oced cards.

Your system is running great, 6.5K seems to be avg for that card, 7.5 is reachable with a little ocing =).

 

[slash196]

Your OC is a good 800 mhz above mine. I want to kill you and steal your system.

 

[WA261]

Originally posted by: bjc112

Originally posted by: WA261
3.131 is all I can get stable at in 05. I can run anything else at 3.245. This only bumped my score from 7568 to 7714 on an X800XT/PE. I would have thought more. The 7568 was on a 2.8 clawhamer. I guess 05 is very GPU/VPU oriented.

What chip do you have?

What voltage?

3500 Venice voltage 1.55+18%.