3DMark05 Score seems a bit low?

[sosuke]

im sorry i might have put this in the wrong spot, im interested in overclocking, but i only have 3dmark to base off of ...

Intel Pentium 4 Northwood 2.40GHz 533MHz Bus
Zalman CNPS7000A-AlCu CPU Fan
ABit IS7 (Intel i865-ICH5)
Corsair TWINX1024-3200C2PRO XMS3200 1024MB 2-3-3-6 (Running at: 2.5-3-3-7)
Xtasy ATI Radeon X800 XT

3DMark05: 4673

poking around in my bios i noticed my cpu multiplier was at 18, but i havent done enough research to say thats odd and feel like i know what im talking about

thanks for the help

 

[sosuke]

alright, so I did a minor overclock, i can't change the multiplier, so i raised the bus to 144MHz
Seems stable so far at 2.59 (144x18)

and running 3DMark05 again: 4957
wow, almost 300 points for 200MHz, is my CPU really that much of a bottle neck?

 

[Duvie]

3dmock loves the memory speed and bandwidth...So raising the fsb and subsequently the memory speed should help the score.....

My old 2.4b did 3.33ghz so keep trying.....

 

[sosuke]

good to know, im (hopefully) going to pick up a Pentium 3.2GHz 800MHz FSB, mainly because hyperthreading was so nice to work with under flash dev (faster compile time)

then my memory i got will be able to run much faster!

thanks again

 

[RampantAndroid]

Fest on this: I got a 5413 3Dmark, and 4474 CPU marks....eat that....haha, though some people have gotten 8000....they are doing some major OCing (and cheating to0 I think!) But check my sig, I have a powerful system, not OCed mind you....I want an FX-55!

 

[Stormgiant]

Originally posted by: RampantAndroid
Fest on this: I got a 5413 3Dmark, and 4474 CPU marks....eat that....haha, though some people have gotten 8000....they are doing some major OCing (and cheating to0 I think!) But check my sig, I have a powerful system, not OCed mind you....I want an FX-55!

Sorry. Don't get it...

Youre joking with him, while you have a MUCH better setup that he does, but only outpacing him with a 15% higher score ?!?!?

Why do you want a FX55 if you can't even put that that FX53 to work properly ???

hehehe

 

[RampantAndroid]

OK, lets see (lets put our brains to work!)

P4 Northwood, MUCH better than those stupid Prescot cores!

#2- The CPU Marks dont have any tangible impact (meaning, his X800 XT is near equal to my 6800 Ultra.....lets not start a video card arguement, lets assume they are equal) with equal videocards, the scores will be close. Were I to skip my CPU tests, my 3DMark05 Score would be the same as if I did do them!
#3- This was with programs running in the backround (I need to do a selective starup soon)
#4- I guess you haven't been around 3DMark05 long, but a 1000 pt difference in score is HUGE

I have a 3DMark01 SE of 26451
3DMark2003 14535

do some research....

EDIT: On top of all this, shall we do some real world tests, and see how diff computers fair in games like D3? With a res of 1280x1024 (max my LCD supports) AA x8, Aniso, and ULTRA HIGH quality settings, I easily maintain at least 30 FPS, and it is typically WELL above that.

 

[B Lo]

Hey guys! You think your PCmark05 score is low... mine is under 700 regardless of how much I OC. My setup is as follows:

P4 520 (2.8Ghz Prescott) running at 3.155Ghz
Abit DuraMax AA8 Mobo with latest BIOS revision (17, I believe)
2 x 256 Kingston HyperX PC2-5400
P.O.S. (Piece of Turd) PowerColor X300Se Radeon (slightly OC?ed from time to time)
500W Ultra Modular PSU
250GB Maxtor (DiamondMax Plus 9) HDD
Etc, etc, etc
ALL RUNNING ON STOCK VOLTAGES
Kingwin Case (426 if memory <no pun intended> serves me correctly)

 

[Hyperlite]

highest 3dmark05 score i have seen was 12k, from a dual opteron dual 6800 Ultra SLI rig. he also managed 28k in 3dmark03.

Sosuke, one of my rigs is in the same boat you are, its a P4 1.8A OCed to 2.35ghz, but i have 1gig of ram and a 6800 @ 375/815. my highest score in '05 has been about 3200.

btw, i don't think you can change the multiplier on the P4 FSB, its locked at 18. =(

 

[Stormgiant]

Originally posted by: RampantAndroid
OK, lets see (lets put our brains to work!)

P4 Northwood, MUCH better than those stupid Prescot cores!

He HAS a P4 2.4B with 533 bus, not a P4C !!!!!!
Read better before answering...

Originally posted by: RampantAndroid
do some research....

LOL....
3dMark2005 is a benchmark for video cards. Period. Unless, the processor is the the bottleneck.
For instance, if your running a 6800ULTRA, with a FX53 the bottleneck will come much later in speed then if youre running a P4B. Get it ?

Now, go on and play some more with your bike...

 

[RampantAndroid]

LOL....
3dMark2005 is a benchmark for video cards. Period. Unless, the processor is the the bottleneck.
For instance, if your running a 6800ULTRA, with a FX53 the bottleneck will come much later in speed then if youre running a P4B. Get it ?

Now, go on and play some more with your bike...

You fool! I just said it was a benchmark for videocards! If this is what you truly believe, then why do you think that my score should be beating his by huge proportions??? You are contradicting yourself.


Oh, about the P4..READ!!!!!!!!!!!!!!!!! He says:

Intel Pentium 4 Northwood 2.40GHz 533MHz Bus

B Lo, I am willing ot be it is your videocard...the x300 isnt too powerful

 

[Stormgiant]

I'm not going to keep exchanging arguments with someone that cannot understand English.

You can read ok, but don't know what your reading

Peace,

 

[RampantAndroid]

OK, what don't I understand?

Yes, P4 Northwoods run at 800 MHz FSB, if thats what you mean, either he said the wrong core name or he said the wron FSB speed. As for CPU bottlenecks, a P4 2.4 won't be much of one, the issue here is that you think that b/c I have an FX-53, I should have a score of 488684774761007657 (would you like that in Binary, or hex perhaps?)

The difference between a 2.4 GHz P4 and a Athlon 64 FX-53 is not that big in most games. It is not the CPU which renders the video...all it does is send the raw data to the memoery, then it is routed to the videocard (or from proc directly to vid card if you have fast-writes enabled) Either way, there isnt much of a difference! Especially not in 3DMark benches.

This started because you thought I had a low score....look at some alienwares that are the same config as mine, they score the same! I am scoring above most of the Athlon 64 Procs the same speed as mine, because I know how to optimize a computer....I've been around long enough to figure this stuff out.

Perhaps you should go and try to fix your bike now, b/c it broke when you sat on it b/c you are to fat!

 

[Stormgiant]

LOL....

The difference between a 2.4 GHz P4 and a Athlon 64 FX-53 is not that big in most games.

This is getting better... Youre telling us that a 3year-old P4B 2400 can give your allmight FX53 a run for it's money in games ????? Good one.

I am scoring above most of the Athlon 64 Procs the same speed as mine, because I know how to optimize a computer....I've been around long enough to figure this stuff out.

You got 26xxx on 3dmark2001. wow, nice one... I did 23xxx almost a year ago with a P4C 3.0 and a 9800XT. Yay, way to go. You really know your stuff....

Fest on this: I got a 5413 3Dmark, and 4474 CPU marks....eat that....haha,

Sorry, but you were the first to throw the stone. Not me. I don't come here to grab about my machine in regards to other people.. YOU DID...

Ahhh.... and if your trying to tell us from the post above, talking about Binary and hexadecimal numbers, that you are any kind related to IT stuff... God bless us


Keep it cool dude... you are beaten to the ground...

 

[RampantAndroid]

Listen you stupid dipf&amp;^%. Can't you read????????? He said it was a fricken P4 Northwood WHICH IS A P4C!!!!!!!!!!!!!!!!!!!!!!!! We DON'T KNOW which one it is. Either way, my old P4C with a 5900 Ultra scored a a 22xxx something in 2001SE....go on ORB....check and see what the highest scores are....the extreme ones (LN cooling, going above 4 GHz) are in the 28xxx range! There is a point when you just can't make an old benchmark use the new hardware....there is a point when the upper score mean NOTHING.

Mind you, before I was getting a 24xxx in 2001 SE b/c of a bad PS, but as soon as I bought a new one, my score jumped.

The proc doesnt affect games NEAR as much as the video card!!! Go on HARDOCP's website and look at their graphs of FPS in games....if equally eqipped, the difference between a P4C 2.6 and a P4C 3.2 is MINSCULE!

The Athlon 64 Series is NATIVE 64 bit! It wasn't designed to work in 32-bit all the time, yet it handles it better that the P4EE. Wait until 64-bit Win XP and games come out, then we'll compare our computers again...wait, yours cant handle 64-bit!!!!!!!!

Yes, I do know IT stuff VERY well. I have been working on computers since I was 12, and I am now a computer engineer, so yes, I know more than you could ever hope to know!!!!!

So crawl back into your hole and shut the &amp;%* up! May god bless your ugly A$$.

 

[Tyrant222]

ok mind you im no computer engineer,

I believe it depends on the engine/game.

many games, the cpu plays a factor.

just look at any a64 review

here

1 minute of work to show that cpu does in fact play a very BIG ASS role in gaming.

 

[RampantAndroid]

out of curosity, where did you go for your degree, and what was your concentration? (will you be getting a masters or no?) I found, that the degree itself means nothing (you may share that opinion or not) Its more about practical knowledge and learning stuff in the field.

Also, yes, it does depend on the game engine. Many don't use the CPU much, especially if the game has an outdated physics engine (thus, Far Cry, D3, HL2 etc are very dependant on the CPU...I should have been clearer on what I meant.

The point is, it isnt like my score should be humungeous just because I have a FX-53...the CPU doesnt play a large role in the score in '05. I can go without doing any tests other than the first 3 game tests, and I get a score.

 

[RampantAndroid]

OK, so I went to your link...as I just said, physics engine heavy games, there is a diff, but look at battlefront....the differences are pitiful....vitually none at all. Same with BF:V.

The 3D mark benches are rendering only...they do not require the CPU to run the physics (the CPU tests do that)

 

[Stormgiant]

Hi again

- First, I'm terrible sorry since you do seem to have some kind of high degree, but can't maintain one discussion without calling names to the other person.
- Second, i'm also sorry that you can't understand that NOT ALL NORTHWOOD's ARE 800FSB AND HT ENABLE. I wrote it in big letters so that maybe you can now read ok...
- Third, i don't know from what rock you came down of, but saying that top of 2001se is 28xxx, is just big and plain BS... You don't have any idea of what YOU are talking about. Top in ORB right now is 42xxx, and i have 3 friends that passed the 40000 barrier, and yes, they do have a FX55
- Fourth. don't know where did you dig that 32/64bits war. BUT, if you don't mind, just look below to my sig and see that i also have a A64... You do know what a signature is, don't you ?
- Fifh, now go on, and go to your street friends and show how much of IT stuff you seem to demonstrate

Peace dude,

 

[boshuter]

Originally posted by: RampantAndroid
Fest on this: I got a 5413 3Dmark, and 4474 CPU marks....eat that....haha, though some people have gotten 8000....they are doing some major OCing (and cheating to0 I think!) But check my sig, I have a powerful system, not OCed mind you....I want an FX-55!

 

[boshuter]

The point is, it isnt like my score should be humungeous just because I have a FX-53...the CPU doesnt play a large role in the score in '05. I can go without doing any tests other than the first 3 game tests, and I get a score.

LOL, you just don't get it:roll: The point is; you score is pitiful for the video card you have. If you aren't breaking 6000 with a 6800 ultra, then you obviously don't have a clue how to set up your system. You have an FX-53 and a 6800 Ultra and you can only manage 26k in 3dmark01?? You should be asking why your scores are so low, not answering questions on subjects you know absolutely nothing about.

As far as the top scores in 3dmark01 being in the 28's...... you must be on crack! I can break 28k with my P4 3.6 and X800XT (not PE), and I get just over 6K in 3dmark05.

 

[RampantAndroid]

Alright...

First, no, I don't doubt that people have easily beaten 28xxx.....however, they are heavily overclocked....an FX-53 (or 55) running at 3.6 GHz is.......rediculous......I don't count OCed procs when I look at 3D mark scores, since I don't OC myself. When I bench, I look to see how mine compares to its own kin. I don't understand why people would OC a perfectly good proc and void the warranty, and possibly waste 800 smackers. I apoligize, I was limiting my searches in the wrong way when searching ORB.

Secondly, for some reason I thought that the northwoods were only P4Cs....just looking, you are right, they arent (forgetful me....I pay too much attention to AMD)

When I bring up the 64-bit war, I bring up your oh-so-wonderful P4 which scored 22xxx.

Fifth, I am sorry, but if you look, you'll see that computer engineering is currently expected to DOUBLE (increase by 100%) by 2010. Starting wages for people just out of college (with a B.S.....Bachelor of Science....I mean Bull......lol) is average 75,000.....once in the field a few years, you see a dramatic increase. I do not know about you, but i fully understand both programming and electronics, on the IC level and in the microchip level, as computer engineering is a wonderful combination of comp sci and elec eng. I do not know about you, but I see no need to demonstrate my magic to my "street friends." Rather, I do my work and get paid for it. Do you have a job my dear sir, or do you spend your days and nights living with mommy and daddy playing computer games?

and Boshuter, well, looking on ORB, I'd say you'd be wrong...the setups that I have are around 5400, and 26-27k....if I werent in line with other procs the same speed as mine (Athlons 64s...) I would be worried....in fact, there are a few that I am able to devestate, who have the same clock speeds as me in procs and videocards. Also, the X800XT fares better in 3D mark benches, thats been show by numerous reviews.

Finally, I am using the FREE versions of these benchmarks...ALSO....currently, my sound card is in the process of RMA....I am using onboard audio (yeah, my sound test was low in 2001SE) So once I get that back, I'll ru it again.

Finally, I always have backround progs running while I do benchmarks...to much to restart constantly.

 

[RampantAndroid]

Also, take note, a good number of those 40000+ scores are using DirectX 8.1..........

 

[RampantAndroid]

Last one for now....Just looked at 05 scores in ORB....all above 6000 fall into these categories:

OCed P4s (around 4GHz) and 64s running at 2500+ (typically)
majority x800s
videocards where OCed
SLI rigs

So what, are you telling me that your non-oced p4 can hit 6000 fine? I mean, cool if you can, but I didn't see anything in my quick glance through that looks like your system in there.

and finally, I'll put this one in: these are synthetic benches...after the snaffu with the 5900s last year, I think most people don't pay attn anymore, as I know I don't buy based on benchmarks. TO me, they mean nothing...they depend on how well the bench was written, how good the drivers are, and too many other conditions that can't be listed or accounted for. I do not take these benchmarks to mean how good my computer is, I take them to mean that I am in line with like computers, and that my system is stable (with my new PSU it is) Finally, I take the fact that I can run D3 in Ultra high mode at 1280x1024 with 8x Antialising to be more of a testament to its power. (yes, I kept a VERY healthly frame rate) Now, I'm going to go and use a gift certificate to get HL2....so you people get busy typing away, and I'll be off enjoying my vacation.

 

[Stormgiant]

Good. Now that you are more calm, we can talk like adults.

I just came home back from WORK, and in the way here I brought a 6600GT to make some tests.

I'll let you know how it fares againts your 6800ultra in a couple of minutes.
And yes, I Overclock my PC's alot, but i also play many hours of games, not only for benchmarks