PLEASE READ!! V. IMPORTANT!!!!!! DURANGO!!!! ANT!!! KRUZIN WHATEVER!!!!!!!

[Snark]

Ok... I've posted a few topics already, but nobody seems to reply. If you are reading this, please help.

I am losing faith in Matrox. This is 4th month I am using the G400 as my graphics card, but there probably have been 4000 problems I have faced. No I am not in the mood for "faces" cause I am seriously quite fed up.

Anyway, I'll discard all the problems for a bit later and now ask for some basic test results to see if its MY G400 or it's just simply that Matrox graphics cards just suck to the core.

I ran NVidia's Tree Mark program to benchmark my G400 and here are the following results for the simple.bat file in the package.

NVidia Tree Mark Test Results...
================================
frames per second: 6.31273
Screen Resolution 1024x768
Frame count is 1000
leaves: 1273
branches: 1272
polygons per frame: 35820


This is where I need help. May anyone who reads this please test this benchmark on your computer and post the results in this forum. I need to be convinced that my computer REALLY just sucks to the core.

So sorry, can't help with links as I got it from a friend, but I do believe you can get it straight from NVidia's website at HTTP://www.NVidia.com .


Thanks for your help.

--------------------
PIII 500 Mhz (no oc but running at 516)
SB Live (Full)
Millenium G400 32 mb DH (no MAX) ( no tweakers either)
256 mb SDRam (No ECC)
ABit BE-6 mb
17 GB Seagate ATA 66 HDD
SoundWorks FPS 2000
US Robotics modem
40x NEC CD-Rom
17" Philips 107MB monitor

[Stringy]

That TreeMark is an nVidia invention...
It is coded to take advanage of the Hardware T&L while giving those computers without hardware T&L not even Software assistance, like in DX....

So, IF this program was D3D based or included SSE you'd see a HUGE increase!...
Try looking at 3Dmark2k for their Complex Poly with X Lights ...
Check out this link and look how nVidia's own TNT keeps up with the geForce once you get to 4 Lights, and Doubles the Performance of a GeForce with 8 lights....

Just goes to show you that it DOES matter who's benchmark it is...

Craig

[Borg]

or try to run MATROX G400 TECH-DEMO on a GE-FORCE...muahahahahahah !..the same shit, nothings wrong with your G400..dont PANIC & wait till TL-800 LoL

------------------
PIII450@465,P2b,128mb,Matrox Mill-G400 32SH,SB.Live !(value),IntelliMouse Explorer (yahoooo! no fu..hm hm. ball any more ...)

[Maggi]

Your system doesn't look that slow when you compare your scores to mine ...

frames per second: 5.88824
Screen Resolution 1024x768
Frame count is 1000
leaves: 1273
branches: 1272
polygons per frame: 35820

...

Those were done a couple of weeks ago on my working rig (specs below), but I don't remember which drivers I used.


------------------
Cheers,
Maggi
________________________

Working Rig:
Asus P2B-DS @ 103MHz FSB
Double Pentium III-450 @ 464 MHz
4 x 128MB CAS2 SDRAM
Matrox Millennium G400 32MB DualHead
Eye-Q 777 (22" with 127kHz) on primary VGA
Nokia 445Xi (21") on secondary VGA


Home Rig:
Asus P2B-S Bios 1010 @ 100MHz FSB
Celeron 333A @ 500MHz
2 x 128MB CAS2 SDRAM
Matrox Millennium G400 32MB DualHead @ 150/200MHz
CTX VL710T (17")
and a brand new Pioneer 303S SCSI-DVD

[Jorden]

I think when I tested it I got about 3fps out of my P3-450, so don't sweat it... will test again tonight

Jorden.

[Jorden]

P3-450, 256Mb PC-100, Asus P2B (1011), G400 32Mb DH (OC'ed to 120.1%) PD4.51, DX 7.0
-----------
<u>Complex Demo:</u>
frames per second: 5.62778
Screen Resolution 1024x768
Frame count is 1000
leaves: 1273
branches: 1272
polygons per frame: 35820

I'm not even that far behind Maggi

Jorden.

[Kruzin]

Your system doesn't suck, that demo is simply not meant for your card. It's an NVida demo for their cards with T&L. Don't bother using that one to bench your system.

At least G400 will run it. GeForce won't even run the Matrox demo

[xortam]

Forgot my specs (see sig). Couldn't edit/delete last message.

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

• ASUS P2B-S, PIII 450MHz, Award ACPI BIOS v1010, 128 MB RAM
• MYLEX FlashPoint RAID+ (BIOS v2.02N) running RAID 0 on two 9 GB IBM DDRS 39130D Disks
• Diamond MX300 sound card, now with MX25 S/PDIF output
• Matrox Millennium G400 Max Dual Head - English
• NEC 5FG monitor
• YAMAHA CRW4416S and NEC Multispin 3x CDs
• 3Com Fast EtherLink XL 10/100Mb TX NIC
• US Robotics 56K Voice FaxModem Pro
• Pioneer DVD-303S SCSI
• Note--All SCSI devices (except disk drives on RAID) are connected to onboard AIC7890 U2W SCSI
• Mainly running Win98 v4.10.1998

[DuRaNgO]

yup, they're right
now the most important thing to remember is that you won't find any programs that need T&A yet, but you can find MANY that use EMBM
jim

[xortam]

Sorry if this is a double post but I seem to be having problems deleting my post. Again, with my sig ...

I'll have to agree with Kruzin in that TreeMark is a biased benchmark; most vendor benchmarks are. Also, any benchmark needs to be well designed and run in a controlled environment in order to accurately measure the performance of a system or sub-system. There are also issues of workload characterization, real-world application, etc. I think that Ziff-Davis Benchmark Operations does a decent job of this. The poorly designed or intentionally biased benchmark still can provide some insight into comparing two systems if you can understand and control the biases.

With that in mind, I'll provide my results of 5.54723 FPS when running the simple test on my non-o/c'ed system (PIII/450). Curious that my number is somewhat lower than Maggi's and Jorden's. Maggi, How do you measure you're bus speed? Are you adjusting the speed and how? Anyone else have a PIII/450 and Max that can offer their results? It would be interesting to eliminate some variances between our systems to see what is causing the difference (I assume the I/O sub-system isn't tested). Any interest in exploring this further? What would really be nice is if MURC could manage a database where we all could compare our results from various benchmarks: Controlling the test environments would be key.



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

• ASUS P2B-S, PIII 450MHz, Award ACPI BIOS v1010, 128 MB RAM
• MYLEX FlashPoint RAID+ (BIOS v2.02N) running RAID 0 on two 9 GB IBM DDRS 39130D Disks
• Diamond MX300 sound card, now with MX25 S/PDIF output
• Matrox Millennium G400 Max Dual Head - English
• NEC 5FG monitor
• YAMAHA CRW4416S and NEC Multispin 3x CDs
• 3Com Fast EtherLink XL 10/100Mb TX NIC
• US Robotics 56K Voice FaxModem Pro
• Pioneer DVD-303S SCSI
• Note--All SCSI devices (except disk drives on RAID) are connected to onboard AIC7890 U2W SCSI
• Mainly running Win98 v4.10.1998

[Mark F]

PIII-450 @ 560 -128MB -G400Max not OCed

Simple :
frames per second: 7.12555
Screen Resolution 1024x768
Frame count is 1000
leaves: 1273
branches: 1272
polygons per frame: 35820
Mark F.


------------------
OH NO, my retractable cup holder swallowed a CD



[This message has been edited by Mark F (edited 13 December 1999).]

[Jorden]

Everyone should try the Complex Demo for comparissons

Why?? Because the Simple Test gave me these numbers:
frames per second: 5.47076
Screen Resolution 1024x768
Frame count is 1000
leaves: 1273
branches: 1272
polygons per frame: 35820

I feel framed when a Complex demo gives a (slightly) better framerate than a Simple Demo

Jorden, only OC'ing the G400

[Greebe]

My Eclipse'd Athlon850 with a Max PD5.41

TreeFarq 10x7 simple.bat
11.4364

Max OC'd 20%
11.4364

Acts more like a poor cpu test than anything!

[Snark]

Hurm...... so WHICH benchmarking proggie should I use for a 'fair' test?

And regarding Jorden: My computer churned out something like 4.0148 frames for complex.

Um... sorry guys got a bit worked up with this whole affair and what with this 40ºC fever roasting me... Did'nt mean to slam the G400 as I did before, apologies.



------------------
My CPU: PIII 500 Mhz (no oc)
SB Live (Full)
Millenium G400 32 mb DH (no MAX)
256 mb SDRam (No ECC)
ABit BE-6 mb
17 GB Seagate ATA 66 HDD
SoundWorks FPS 2000
US Robotics modem
40x NEC CD-Rom
17" Philips 107MB monitor

--Everyone wants a snarky... ...YEAH RIGHT--

[Greebe]

At the moment I'll have to agree with DuRaNgO, 3Dmark 99 Max seems to do a good job.