Sign in with
Sign up | Sign in
Your question

3DM2k1 scores inconsistant?

Tags:
  • Graphics Cards
  • Graphics
Last response: in Graphics & Displays
Share
December 28, 2001 9:08:08 AM

i posted here as the other thread of 3Dmark2k1 was already WAY too long,

well as few posted there it seemed that 3dmark didnt have any consistancy in its scores(hacked scores?=/)

for any case and further coparison here's my OCed rig-(rigNo2-P3)
http://gamershq.madonion.com/compare2k1.shtml?2025006
-score:5297

now here are some cases in mind:
http://gamershq.madonion.com/compare2k1.shtml?2158270
-score :8581
-the rig is twice faster then mine-CPU wise.
-High Polygon Count (8 Lights) was .2 FPS faster then mine,all other tests seem to be 20%+ faster then mine.
over +3K points then my socre.

http://gamershq.madonion.com/compare2k1.shtml?2269934
-score :8608
-rig 2.5 times faster then mine CPU wise.
-RADEON 8500
-High Polygon Count (8 Lights) three times(!!)higher then mine, all other tests seem to be 30%+ higher then mine.

http://gamershq.madonion.com/compare2k1.shtml?2019671
-score 6836
-rig 2 times faster then mine CPU wise.
-Ti 200
-High Polygon Count (8 Lights) .6 FPS lower then mine.
-game scores 10% higher then mine
-Environment Bump Mapping/DOT3 Bump Mapping/Vertex Shader/Pixel Shader/Point Sprite, seem to give me 10% overhead.

http://gamershq.madonion.com/compare2k1.shtml?2322002
-rig 2 times faster then mine.
-High Polygon Count (8 Lights) .1 FPS faster then mine.
-only game scores are 20/30%+ faster then mine, all other are almost the same.
-----------------------------
-----------------------------
it seems that the radeon gave a REAL boost in the bottom part of the tests, specially a three(!!) times more plygon count.
i could say that the CPU is more of the factor at the bottom tests but the R8500 proved me wrong.

so what are the real factors -specially with the plygon count, to get the ultimate rig for 3Dmark2k1?

and what the hell does the picture quality test tests???
and why doesnt it have a score in the scores list?
is it to see if no one's cheating in their tweaking?
what if you would?it doesnt have any impact on the score..

<font color=green>
*******
*K.I.S.S*
*(k)eep (I)t (S)imple (S)tupid*
*******
</font color=green>

More about : 3dm2k1 scores inconsistant

December 28, 2001 11:04:27 AM

The bottom part of the test is T&L intensive and the Radeon 8500 has superior fixed and programmable T&L units.

AMD technology + Intel technology = Intel/AMD Pentathlon IV; the <b>ULTIMATE</b> PC processor
December 28, 2001 3:47:52 PM

When you go to the <A HREF="http://gamershq.madonion.com/products/orb/" target="_new">ORB</A> and click "matching your current project (approximate CPU speed)" link on the search and compare page will show all machines in your "class" based upon your current published project.

Make sure FSAA is not turned on, set your quality to fastest, turn off all background programs, defrag your hd and try again. Add ramsinks to your video card ram if none are present this is cheap and effective.

There is a BIOS update that will make an OEM card into a retail card. You have to search <A HREF="http://www.rage3d.com" target="_new">www.rage3d.com&lt;/A> for the link.

Windows 98SE will get you about 300 more points.

Some people will adjust the LODbias in the registry making all textures white or invisible. This will greatly increase your score by 1000+ I consider this method cheating since If they wanted you to run the benchmark with no/highly altered textures they would have made a checkbox.

Also there is a voltage mod for the brave, this will increase the cards voltage to be more stable at higher speeds and may also reduce the life of your card.

Adding peltiers with waterblock to the core is an option if you want to compete/reduce the life of your card. alot of people go this route before they voltage mod.

Im running both GF3 and R8500, GF3 wins at 3Dmark on my machine. The ATI drivers are still need time to mature. So far the 8500 is a poor overclocker, I have not done the voltage mod or pelt mod yet. I replaced ramsinks after hitting max speed of 285Mhz thats a +10Mhz overclock.
December 28, 2001 4:44:41 PM

i dont have a RADEON i have a geforce 3 card - Gladiac 920 to be specific, it does have ram sinks and it overclocks to its maximum specs working stable.

this comes to prove my point about the image quality test..
it just deosnt do anything..

<font color=green>
*******
*K.I.S.S*
*(k)eep (I)t (S)imple (S)tupid*
*******
</font color=green>
!