POST YOUR 3DMARK 2003 SCORES HERE!!!!

[Rabbitrunner51]

Mr Burns... good info on the timings as I 'm pretty positive that my sytem is running in dual channel.( checked the bios ) ,but you bring up a interesting point about the memory being in the same mode....
Is there another setting in the bios that i should recheck? My memory is running at the standard timings for what they are set for.....More memory might up this some also..When I get my new HD next week sometime everything will go up also...with the speed and a cleaner system...due to a totally new re-install,and the fact that it will jump from 5400 speed to 7200 speed.

Overall there are so many issues and bottlenecks that may or may not occur in ones system...this is where a total benchmark shows the whole picture..and can definitely affect the basic graphical benchmark score.....

It is very intersting to me..that there are so many different benchmarks for all of us with similar systems and other factors...alot of fun to compare, but we shouldn't get so caught up in this to a point of fixiation....there is always better and best and worse.....

[dalleman]

I just reran 3DMark03 and got the following:

7232 SCORE.

Setup was this:

Video bench only (NO SOUND OR CPU)

Resolution: 800x600

Anti-Aliasing: NONE

Texture Filtering: Optimal

Max Anistrophy: 4

Vertex Shaders: Optimal

Repeat test: OFF

Fixed Frame Rate: OFF

[TiGgErDbC]

3d mark03: 5595 3dMarks
3d mark2001: 16905 3dmarks
AquaMark3: 39593

with the setup:

Mobile XP 2600 @ 12 X 200 1.68v
GA-7N400 Pro2 rev 2
PR3ACH3R GA-7N400 Pro2R2 FB
WhisperRock IV
Arctic Silver V Thermal Compound
Sapphire 9800 Prow/ VGA Silencer
Antec P-160 Aluminum Case W/ 2 120mm Fans
Antec True Blue 480W PSU
512 Corsair XMS LL
WD 36.7 Raptor

and NAV and background programs running still


Still wanna to overclock some and maybe speed up the 9800

not bad

i have a 2500+@189 fsb locked multi
gigabyte 7n400pro2 rev2
using ac5 as well
sapphire 9800 NP@ 425/329 with new VGA silencer and OCZ bga copper ram sinks
antec case with a enermax 465
512 OCZ el @ 2-3-3-6
WD raptor 36.6 gig(waiting for second one to come back from rma for my raid 0)

i just hit 5859, i'm going to try to break 6000... wish me luck i'm so close, all my vid card settings are default. problem is my ram doesnt want to break 330. i get artifacts at 330(was 315 before i put my new ramsinks on). also i have a crappy 2500 to over clock.. i wanted to hit 200 fsb.. i'll be trying to sell my 2500 and get myself a mobile sometime. then my "dream" system will finaly be compleat, and i'll be able to start on my new one

*just hit 5961 at 430/329

[MrBurns]

@dalleman: You should use standard settings for your benchmark (1024x768, no anisotropy).

@Rabbitrunner51: the best timings you can get normaly are CAS-TRCD-RAS-TRAS=2-2-2-5. It depends on your memory if you can get so low and also on the clock speed of your memory.

[Rabbitrunner51]

The timings for the product state 2-3-3-6 for the Corsair PC 3200 Platinum....are these good timings..I don't know...

[MrBurns]

Originally posted by Rabbitrunner51
The timings for the product state 2-3-3-6 for the Corsair PC 3200 Platinum....are these good timings..I don't know...

Yes, but you can try to reduce them further. But use a program like prime95 to tests stability. Run it for at least 24 hours w/ your final timing. In addition you can use the memory testing programs memtest86 (at least 2 full tests) and GoldMemory (the same).

[Rabbitrunner51]

MrBurns.....thanks alot for the info....this sounds very useful....

Another general info and question is I have my Aperture grahics setting at 256 now ( use to leave it at 64..until I learned of the Mobos default ..) and everyting runs fine at this setting and also...what should the PCI Latency timer be set to? I uped mine to read 128.... Is this OK or what is the best..and what does this do and mean?

[MrBurns]

Do you have an ATI or NVIDIA GPU?

If you have a NVIDIA GPU, the best setting for the AGP aperture size is 50% of your system memory.

If you have a Radeon GPU, the only way to find it out is to make a benchmark for every possible seeting from the minimum to 50% of your RAM.

[mobo57]

MrBurns, check out my totally unofficial and unscientific test results:
http://www.sysopt.com/forum/showthre...49#post1114849

[MrBurns]

The reason why I posted, taht when you have a Radeon you should test it yourself is, that the results are different on every system. Some systems have the best results w/ 32MB or 64MB (which is usually the minimum), some have the best results w/ 512MB (which is the maximum). And some have trhe best results somewhere between, i.e. 128MB or 256MB.

[mobo57]

What about the results for the TI? Seems lower is better. Maybe it is the system memory is slower than on card?? Weird..

[MrBurns]

Originally posted by mobo57
What about the results for the TI? Seems lower is better. Maybe it is the system memory is slower than on card?? Weird..

I have a GeForce 4 Ti 4600 w/ 8xAGP and 128MB. The best results I have are w/ the max. of 512MB. Here are my results:

Aperture Size /3DMark01/3DMark03/Aquamark3
32MB/14029/1779/21740
64MB/14234/1808/22192
128MB/14210/1803/22206
256MB/14215/1894/22208
512MB/14241/1896/22208

I also made a graphic w/ excel:

[mobo57]

Ok let's figure this out,
From 32 meg, you got a boost of 1.5% for 01, 2% for Aqua and 1% for 03 until 256 meg and then a jump to 6.5% and same for 512 meg.
Looks like overall anything over 32 megs except for 03 is about the same....
Me thinks we gots to get others to post here:
http://sysopt.com/forum/showthread.p...hreadid=162116
I'll compile the results and see what we come up with.
Also, think we need to do some with UT 01-03, Halo etc.

[MrBurns]

For me these 3 benchmarks are enough, because they basically all show the same result (512MB is alllways the best).

[mobo57]

Only shows that 03 gets a significant boost. Real world, UT shows a lot less. Seems that overall, it does not matter that much, assuming previous and current gen cards.
hypo 1:
Game programmers know the physical limitations of systems (read hardware).
hypo 2:
Game programmers set their limits to those systems (hardware).
hypo 3:
most games today in standard settings will not tax top of the line cards, ie 9800 and 5950. Only cards that seem to really benefit are older cards with lower performance specs.
Therefore, looks like a lot of hype for not much in performance. Or????