My Benchmarks after upgrading graphics card

[Spitfire_x86]

Many people in this forum may know I've recently bought a graphics card (Radeon 9000). Here I'm sharing benchmarks with you.
---------------------------------------------------------

Specs (Older):-
===============

CPU:- AMD Duron 1 GHz
Mobo:- MSI K7N420 Pro (nForce 420-D chipset)
RAM:- 224 MB(2 x 128 MB (-32 MB) PC2100 DDR, @ PC2100 2.0-2-2-6, Dual Channel mode, Async with FSB (200/266), IGP Enabled, 32 MB shared VRAM for IGP
Graphics card:- Integrated GeForce2 MX
HDD:- Maxtor D740X 40 GB



Current Specs:-
================

CPU:- AMD Duron 1 GHz
Mobo:- MSI K7N420 Pro (nForce 420-D chipset)
RAM:- 256 MB PC2100 CL2.0 DDR @ PC1600 2.0-2-2-6, Dual Channel mode, Sync with FSB (200/200)
Graphics card:- Gigabyte Radeon 9000 64 MB
HDD:- Maxtor D740X 40 GB



OS:-
=====

Windows 2000 Professional (Service Pack 3)



Drivers:-
==========

Chipset:- Nvidia reference nForce drivers 2.03 (WHQL)
Integrated GeForce2 MX:- Nvidia reference 41.13 (WHQL)
Gigabyte Radeon 9000:- ATI reference 7.84 (Catalyst 3.2, WHQL)



Benchmarks:-
============


3DMark 2000 (1.1)
==================

Older specs (Integrated GeForce2 MX):- 4409
Current specs (Radeon 9000 Pro):- 6272

Result:- <b> + 42.25% </b>



3DMark 2001SE (build 330)
==========================

Older specs (Integrated GeForce2 MX):- 2274
Current specs (Radeon 9000 Pro):- 5441

Result:- <b> + 139.27% </b>


3DMark 2003
============

Older specs (Integrated GeForce2 MX):- 120
Current specs (Radeon 9000 Pro):- 977

Result:- <b> + 714.17% </b>


This is end of gaming benchmarks. Since IGP was limiting my system performance by sharing memory bandwidth, and forcing to run the memory async with FSB (for better 3D Peformance), I've got a good boost performance in other areas except gaming.



Other becnhmark:-
==================


GoGo NoCoda Encoder
====================

[Converting 67 MB 44100 Hz, 16 bit PCM WAV file to 128 kbps MP3 File]

Older specs (IGP in use, Async (200/266) memory operation):- 49.00x (/1x CD-ROM speed)
Current Specs (IGP not in Use, Sync (200/200) memory operation):- 67.652x (/1x CD-ROM speed)
Current Specs (IGP not in Use, Async (200/266) memory operation):- 61.726x (/1x CD-ROM speed)

Result:- <b> + 38.06% </b> (previous vs. current 200/200), <b> + 25.97% </b> (previous vs. current 200/266)

---------------------------------------------------------


With this kind of performance boost, I should be very happy. But I feel that good, reason is described in this thread <A HREF="http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=343380#343380" target="_new">http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=343380#343380</A>

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>

 

[chuck232]

Well, it has improved gaming quite a bit I see...

Have you tried playing any games yet instead of just benching 3D Mark?

...And all the King's horses and all the King's men couldn't put my computer back together again...

 

[Spitfire_x86]

Of course, I've played all my favorite games for sometime with this new card. With fraps, they show performance boost like 3DMark 2001SE (except games that are quite dated, about 2 years). I wanted to play MOHAA (for 4th time) and Max Payne (for 2nd time) again with this new card. But since I've these games before, I can't find them much enjoyable now. :tongue:

BTW, the 2nd game test of 3DMark 2003 runs @ 6.3 fps and the 3rd game test runs @ 7.4 fps. Haven't watched such ultra-fast slide shows before! And more, Vertrex shader test runs @ 5.0 fps and Ragtroll runs @ 3.7 fps. They are bit closer to real-world slide show.

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>

 

[chuck232]

Yeah, even with my Ti4200, the 2nd and 3rd tests run like crap. It's mainly cause PS1.4 (i think) almost halves the polys that need to be rendered. Since my Ti4200 is only PS 1.2 (or 1.3) I have to do almost 2x the work... kinda sucks cause I was almost getting seizures from it... :lol:

...And all the King's horses and all the King's men couldn't put my computer back together again...

 

[jaythaman]

Does 3dmark 2000 depend much on the proc too?? Coz when i tested it on my new system i believe i got close to 7000 marks on a gf4 440 mx which is lower than the 9000 (is it?)

My computer NEVER cras...DOH!.

 

[Spitfire_x86]

It's mainly cause PS1.4 (i think) almost halves the polys that need to be rendered

Radeon 9000 has pixel shader 1.4, and Dual Pixel/Vertrex shader like R8500/GF4 Ti series.

How much fps you get in these tests?

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>

 

[Spitfire_x86]

Post deleted by Spitfire_x86

 

[phial]

SPITFIRE!!!!

you amaze me EVERY time you post just about


first off he said 3dmark2000, not 3dmark03 FFS



2nd, the 9000pro only has ONE, read 1, vertex/pixel shader.
HALF the amount of a 8500 or GF4TI

MAN get your facts right. your confusing alot of people probably

 

[jaythaman]

YAY!!! My 440 mx kicks 9700 pros arse!!!!!
ok Now somebody answer my question

My computer NEVER cras...DOH!.

 

[Spitfire_x86]

Damn, what a mistake I've made! I was in a hurry to go to college.

2nd, the 9000pro only has ONE, read 1, vertex/pixel shader.

THG Radeon 9000 review
<A HREF="http://www6.tomshardware.com/graphic/200207181/index.html " target="_new">http://www6.tomshardware.com/graphic/200207181/index.html </A>

In the chart they say that R9000 has two vertrex shaders like Radeon 8500, they never said that it has only 1 pixel /vertrex shader

More from the article:-

<i> The second change has to do with the vertex shader. It wasn't till now that ATI has admitted that the Radeon 8500 contains two vertex shader units, similar to the GeForce4 Ti. The reason for keeping quiet on this was more because of marketing concerns rather than understatement. The R8500 had two shader units more than NVIDIA's GeForce3. With the R9000, these units were re-worked and optimized. If the Canadian PR department is to be believed, then the new vertex shaders contain many optimizations that are also found in the R300 design. </i>

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>

 

[phial]

ive read in quite a few articles, a few at THG i think , that states the 9000pro only has 1 pixel shader. PIXEL shader...vertex shaders can be done in software (xabre,gf2) cant remember where i read it tho

besides, that article is part of the reason people think the 9000pro is better than an 8500

 

[eden]

THG tends to often make errors in their charts. Sometimes they give wrong clocks, sometimes wrong amount of pipelines, shaders.

BTW Pixel Shader amount is never specified, something I never understood why. For all we know, one Pixel Shader is used despite 4 VS in the R300.

--
This post is brought to you by Eden, on a Via Eden, in the garden of Eden.

 

[chuck232]

Well, haven't yet tested my R9700 Pro yet, but with my Ti4200, I was getting ~7 FPS on average I'd say in the 2nd and 3rd games. It was so bad I quit, so I didn't get a final score.

...And all the King's horses and all the King's men couldn't put my computer back together again...

 

[phial]

yeah the GF4 series are just realy fast dx7 cards, with dx8 features thrown in

 

[Spitfire_x86]

I know that vertrex shaders can be done in software. But then, THG won't say it has 2 vertrex shaders.

Pixel shader benches and Nature test of 3DMark2001 indicates that it has equal number of pixel shaders.

Now give me a link that clearly says R9000 has only 1 pixel/vertrex shaders.

besides, that article is part of the reason people think the 9000pro is better than an 8500

IMO, this article has nothing that can make people think that R9000 Pro is better than R8500. THG said about much weaker Multi-texturing fillrate and 3DMark2001 fillrate tests proved that. They also said about using newer driver with R9000 Pro. So nothing is confusing in this article

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>

 

[Spitfire_x86]

bump

Submit your opinion <A HREF="http://forumz.tomshardware.com/community/modules.php?name=Forums&file=viewtopic&p=28537#28537" target="_new"> Should Tom Fire Omid? </A>