CPU Chart 3DMark2005 stats vs My CPUs stats

[Arch00]

I own an AMD X2 4400+, my top score with it on 3dmark2005 is 5644.

On the Tomshardware CPU chart, they get a score of 6164. About 500 points higher using the same CPU.

I'm running on an A8N-SLI Deluxe motherboard, 2 gigs of ram, with slightly higher timing than the one used in the test project. (2.5 CAS instead of 2.0).

What exactly could be limiting my score so much, and should I be worried about it? I have extremely few start up programs.. just the one for my Razer mouse so I doubt i'm using up extra CPU cycles that would limit my score.

 

[fishmahn]

Isn't 3dmark05 pretty GPU oriented more than CPU oriented?

And the RAM timings could make some difference, though I doubt that much either.

Mike.

 

[Arch00]

These are the CPU scores, not the total score which is quite heavy on the GPU.

The CPU tests run off the CPU only as far as I know (which explains the extremely low framerates on the CPU tests)

 

[fishmahn]

These are the CPU scores, not the total score which is quite heavy on the GPU.

Ok, that took care of that possibility...

In that case I don't know... Hopefully someone else will have some possibilities - I'm not much into benchmarks except to see if the PC works.

Mike.

 

[Arch00]

I actually haven't ran it yet, I'll give it a try.

 

[Arch00]

I got a 1686 on 2006's CPU test.

 

[MadModMike]

I got a 1686 on 2006's CPU test.

I get 2011

~~Mad Mod Mike, pimpin' the world 1 rig at a time

 

[Omerta]

I got a 1686 on 2006's CPU test.

I get 2011

~~Mad Mod Mike, pimpin' the world 1 rig at a time

2086 for me.

http://service.futuremark.com/compare?3dm06=222608

 

[TabrisDarkPeace]

The CPU score is about 80% CPU and 20% GPU / drivers.

The 3DMark documentation indicates this.

I only get 2545 in '06:
http://service.futuremark.com/compare?3dm06=84767

However, only the 2nd CPU test uses four cores, the 1st one uses only two. (Task Manager graphs confirm this).

 

[Arch00]

The CPU score is about 80% CPU and 20% GPU / drivers.

The 3DMark documentation indicates this.

I only get 2545 in '06:
http://service.futuremark.com/compare?3dm06=84767

However, only the 2nd CPU test uses four cores, the 1st one uses only two. (Task Manager graphs confirm this).

http://service.futuremark.com/compare?3dm05=1894183

But I have dual 7800 GTXs! And they only used a single 6800GT in the tests.

 

[TabrisDarkPeace]

Maybe the (more recent - GeForce 7000 series) drivers offload some functionality to the CPU(s), or 2nd core, to boost frame rate, detail, etc and that affects the CPU score a little.

Wouldn't surprise me at all.

ATI have Catalyst AI, and it has 3 settings, it can boost frame rates a good 20% or so at times. 8O

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

Just wait until Friday afternoon

My laptop (Presario V2400) will have far more 3D performance than my Opteron 270 (quad-core)... :lol:

As I am taking the Opteron 270 back to the onboad ATI Rage XL - 8 MB. (The system memory is faster than the onboard video memory ). Good thing I ain't losing that 8 MB to system memory... I'd really miss it. :roll:

(Laptop has ATI Radeon Xpress 200M single channel, shared with system memory 8O )

So soon my score in most of the 3DMark series will be zero, as the GPU won't let any of the tests run.

Less gaming = More coding. 8)

PS: If we are looking at '05 my CPU result is: 7374

 

[TabrisDarkPeace]

ATI chipset drivers are the first thing that comes to my mind looking at your board.

A mainboards chipset performance / drivers will affect 3DMark scores, and even general performance, significantly. Maybe not significantly enough to notice unless you sit at each given configuration for 4+ hours each, but it's there.

Which Video Card are you running ?

If it is an ATI card, What ATI Catalyst Control Centre settings are in use ? (eg: CATALYST AI level, Is Anisotropic filtering forced, Is [Temporal] FSAA forced, Is Gemetry Instancing enabled ?, etc, etc).

 

[TabrisDarkPeace]

What ATI Catalyst Control Centre settings are in use ? (eg: CATALYST AI level, Is Anisotropic filtering forced, Is [Temporal] FSAA forced, Is Gemetry Instancing enabled ?, etc, etc).

Try forcing Texture / MipMap quality to High Performance.

Try settings CATALYST AI to Standard or Advanced.

Confirm latest mainboard chipset and video card drivers are installed correctly.

...

 

[TabrisDarkPeace]

You forgot to mention Geometry Instancing and High-Quality (transparency) filtering.

 

[TabrisDarkPeace]

nVidia video drivers don't hit the CPU as hard (as the ATI drivers do) during those CPU tests.

Whack in a GeForce 6800 GT/GS and see for yourself.

ATI have done very interesting things in their driver, since day one.

Catalyst AI is what sparks my suspicions, try turning it off, see if CPU score rises more (with same settings).

 

[Arch00]

I actually used to run dual 6800 GT's in SLI.

http://service.futuremark.com/compare?3dm05=1885790

actually got a lower CPU score.

 

[MadModMike]

I used to run 6800GS's in SLI and they beat a stock x1900XT when they're @ 515/1250 each.

~~Mad Mod Mike, pimpin' the world 1 rig at a time