Bit P*ss*d With Benchmark (3dmark03)

Kintarō Hattori

Over the weekend I upgraded myself from a Socket A XP2400 with an AGP 6800TD LE to a brand new Athlon64 setup and I got virtually the same score with the new setup as I did in the old one: around the 8500 mark.

At the moment I'm running the following:

Athlon64 Skt 939 3000
1GB (2x512) Dual Channel DDR400 (Generic with Copper Heat Spreaders)
Zalman CNPS 7000 Alcu Cooler
Sapphire X800GT 256MB PCI-E
Abit AN8-V Mobo
Hiper Modular 580W PSU

2 x 120mm fans in exhaust airflow setup. 1 x 80mm fan at top of case. I have it in a CMSTACKER, so there is plenty of room and everything is tidied pretty neatly, so airflow shouldn't really be a problem.

There is plenty of fans and cooling going on in the case, so I dunno what it is. Surely I should have seen a greater increase in performance with this system than I had in my old Skt A system?? Afterall I had one of the leper 462 chips!

Old System:

AthlonXP 2400 with Artic Cooling Heatsink and Fan
Asus A7N8X Mobo
Asus 6800TD LE 128MB
Same Memory and Case as above
Sharkoon 350W PSU

Jammer

what benchmark?

if its a 3D benchmark, then doesnt sound too wrong, as 3dmark would mostly bench your gfx card, and they would be in or around the same performence i'd say...

Shoulda seen me when my 7800GT benched at under 5k in 3DMark05 - nearly died. Formatted windows and its closer to the 8k now

Kintarō Hattori

Hmnn,

Fair enough, but I thought for the simple fact that I just upgraded to 64bit I'd see even a tiny increase. Also as it's just been upgraded, Windows was completely fresh- I don't even have any games installed yet!!! :eek:

unkel

Jammer wrote:

3dmark would mostly bench your gfx card, and they would be in or around the same performence i'd say...

My guess too

unkel

eo980 wrote:

I just upgraded to 64bit

Upgraded to 64 bit what? Just using a 64bit CPU does not work

First of all, your operating system (OS) needs to be 64 bit. Then all your hardware and your software needs to be 64 bit too

Kintarō Hattori

My point is that I upgraded from an old Socket A chip, the 2400 to a new 64bit 939 chip with PCI-E. I was expecting a SLIGHT performance increase over what I had, but it's practically identical.

tba

check the game preformance difference, it sould be a bit better

WizZard

But in fairness, the difference between old and new graphics card are practically non-existent. Apart from one being AGP anfd on PCI-E their performance is pretty much similar.
Did you research your parts before buying or just assume that cos the X800GT is new it would be better than a 6800 ?

stevenmu

2dmark03 is farily old at this stage and probably would have been more GPU limited than CPU limited by both your systems, and as mentioned I don't think there'd be a huge difference between your graphics cards. If you'd tried the 05 versions on both systems there'd probably have been a huge difference.

Kintarō Hattori

Fair enough guys, fair enough. Wizzard I appreciate your input. I've been on the auld computer scene a while and I didn't think that just because it was new it would be better.
I did assume I would see some sort of increase, small as it may be, but I guess I was wrong to assume that. I wasn't looking for a huge difference, even just a few hundred points in 3dMark, but fair enough there is none to be had.

Joeface

Tis is a little Bit off topic but do benchmarks really matter, i mean Ur system specs ar s**t hot, Havnt Run 3Dmark03 on my own only 05 and aquamark3
And I only get around 2500 with 3Dmark05 and 46000 with Aquamark3 and in searching the net I have found I am way down on everyone else, yet my system Plays everygame I have perfectly, ( HL2 , BF2, FarCry, DOOM) all setting to high for all of them.

me system is only : P4 2.8Gig HT (478), 2 Gig of Ram , 9800pro 128Mb all on IS7

Kintarō Hattori

Hey Joeface,

Yeah I agree with you, that in a way it shouldn't matter. I haven't installed any games yet but I know it will play them all fine, I was playing Farcry on my Skt A system with everything maxed out. Even Doom 3 was perfectly playable on it.

It's just nice to know if you've forked out for a new system that it'll score a bit higher than your old system, especially when you had a chip such as the XP2400 and not the 2500!!!

Khannie

Try turning off AA and AF in your advanced settings if you have them on. For bench purposes it'll slow you. The cards are similar enough allright, but I would have expected a reasonable performance increase because of the CPU (hell, it's well over 20% faster people!!).

Make sure you have the latest chipset drivers as well as gfx etc.

WizZard

Joeface wrote:

Tis is a little Bit off topic but do benchmarks really matter,

Really, no.*
Just because my system can get a few more 3dMarks than yours doesn't mean mine is inherently better than yours. It's down to luck a lot of the time. Run the benchmark a couple of times to get your average score. I've noticed differences of a couple of hundred marks between consecutive runs of 3dMark.

It matters what's best for you, and if you are happy with the real world performance (which is miles away from synthetic benchmarks), then that's all that should matter.

*I can say this because my 3DMark05 record is as yet unbroken on boards

Joeface

Hmmmmmmm so what is ur 3Dmark05 score......

WizZard

10,579 3DMark05

Khannie

*cums*

DemonOfTheFall

Did you have the latest GFX and motherboard drivers installed, and installed in the rightr order?

If you run your generic WinXP motherboard drivers expect crippled performance

maidhc

eo980 wrote:

It's just nice to know if you've forked out for a new system that it'll score a bit higher than your old system, especially when you had a chip such as the XP2400 and not the 2500!!!

Please dont say you upgraded what most people would consider a sh1t hot computer and are now looking for a justification for doing so. Surely one should upgrade when you notice things start to lag, games cant play at the resolution you want etc? (I am not a gamer, so I may be wrong...)

DemonOfTheFall

An XP2400+ isnt really a **** hot computer, in fact it would struggle with a lot of things. he should have seen a fairly whopping increase in system performance overall by upgrading to an A64 3000+ though. It's probably mainly a case of 3mark03 only taxing the graphics card and not the processor though.

Run a CPU benchmark and it should annihilate your old machine.