My sys vs A64 2.6ghz

OblivionLord

Limp Gawd
Joined
Apr 16, 2004
Messages
168
I cant find any ingaming benchmarks for A64s vs P4 anywhere. I can only find timedemos and botmatches which mean squat to ingaming. Anyone know a site that does compare a simular system like mine to a A64 @ 2.6ghz doing ingame benching?

My 3dmark01 score is 23483 for stock xt-pe speeds and 23906 @ x850xt-pe speeds (540/590). On furturemark there are alot of people with my card at stock speed on oced A64's at 2.6ghz getting on the 30k mark in 01. Basicly thats 6k diffrence. I'm sure this means a great deal for ingame benching.

If noone can find a site that has ingame benching comparisons then I'm willing to download game demos to see results for a fair comparison. Problem is somone with an A64 at 2.6ghz with my card will have to be willing to bench with me.

Thanks
 

smercer

n00b
Joined
Jan 2, 2005
Messages
25
I have a 64 939 3500, I am currently running it at 2.4ghz but have oc'd it closed to 2.6(2.56) the only prob witht that oc'd ws that i am only using air so the load temps went a little to high to be safely running it. But i could run my system at it for a little bit to test it.
I have the Saphire x800xt pe. Just tell me the test that you want to run and i will get them and see what the results are.
 

OblivionLord

Limp Gawd
Joined
Apr 16, 2004
Messages
168
Awesome.

I have UT2k4 which is very cpu dependant. If you dont have the game and are willing to dl the demo then Im all for it.

I also have Farcry but, im willing to download HL2 and Doom3 demos. Just get back with me and we can go from there. If you want you can prvt msg me.

Just wondering ... what are your load temps at 2.6ghz and what HSF are you using to accomplish this?
 

arentol

2[H]4U
Joined
Jun 15, 2004
Messages
2,712

Absentee

[H]F Junkie
Joined
Aug 27, 2004
Messages
9,412
OblivionLord said:
Awesome.

I have UT2k4 which is very cpu dependant. If you dont have the game and are willing to dl the demo then Im all for it.

I also have Farcry but, im willing to download HL2 and Doom3 demos. Just get back with me and we can go from there. If you want you can prvt msg me.

Just wondering ... what are your load temps at 2.6ghz and what HSF are you using to accomplish this?

the graphics in the ut2k4 demo are toned down a bit...no lights on cars. etc
 

OblivionLord

Limp Gawd
Joined
Apr 16, 2004
Messages
168
Timedemos are prerecorded like you said which are basicly scripted benchmarks without any variances. Same applies with botmatches. Therefore neither will show the same results as ingame performance. Example.. when you have 32bots on screen with variable calculations, its going to put a hell of alot more strain on the cpu than a timedemo with the same number of bots since all of it is predetermined. Timedemos and botmatches are nothing more than synthetic benchmark results just like 3dmark. No variances or nothing.

Sucks that the demo doesnt offer alot but, even at its highest settings it still can show a diffrence between systems.
 

smercer

n00b
Joined
Jan 2, 2005
Messages
25
I got a 5880 score in 3d mark 05 ( sorry i have been really busy to do the test but as soon as I can i ill give you the results) By the way what is your score?
 

OblivionLord

Limp Gawd
Joined
Apr 16, 2004
Messages
168
3dmark05 is heavily GPU intensive which is how I got around the same score as you

05 = 5855

This UT2k4 ingame bench will deffinently show a diffrence since its heavily CPU bound. I sent you a prvt msg about how to performn the ingame test. If you could Id like to see results. Thanks
 

chipmonk010

Weaksauce
Joined
Oct 31, 2004
Messages
86
why not use sandras arithmatic benchmark since it just tests the cpu....i get

10931 MIPS
3894/5043 MFLOPS

with my s939 90nm 3000+ @ 2.4ghz
 

DryFire

2[H]4U
Joined
Jan 19, 2003
Messages
2,777
maybe i can get my s754 3000+ to 2.5-2.6ish when i get my wc'ing in (should of had the remaing parts today :( ) I doubt my ram will go that far though.

woudln't super pi or something be a better test?
 

chipmonk010

Weaksauce
Joined
Oct 31, 2004
Messages
86
3d mark is more of a graphics card test....if someone has a better video card and a slower cpu they could still outscore u. if you want to know if your cpu is running fast or slower then arithmatic benching is the only way to go. btw 3dmark is a synthetic benchmark....
super pi is also a good one it realies alot on ram aswell tho not just cpu.
 

OblivionLord

Limp Gawd
Joined
Apr 16, 2004
Messages
168
Im mainly interested in ingame benching since I do alot of gamming. There just isnt any luck for me to find sites that does ingame benching but, only botmatches and timedemos.
 

Stoopid Guy

Limp Gawd
Joined
Oct 1, 2004
Messages
240
The difficulty with 'ingame benching' is that unless it IS a demo, like CS:S's, there is just about no way of doing EXACTLY the same thing on two computers, unless I am being stoopid here :D
 
Top