VOGONS


Improving Quake performance on POD83.

Topic actions

Reply 20 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

Thanks I’ve been enjoying it!

I think we have a winner by removing one simm module I’ve been running stable at 100 MHz for two hours or more now!

Currently at 22.6 fps!

Curious if 50ns memory would allow me to drop the dram wait state back to 0ms.

Also curious if anything would best the tseng et6000. For pure dos quake anyway.

Reply 21 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

Alright, installed windows 98 SE and a Voodoo2 and now I’m hitting 39.1 FPS at 640x480!

Attachments

Reply 22 of 67, by Joseph_Joestar

User metadata
Rank Oldbie
Rank
Oldbie

It might be worthwhile to try out a few different versions of 3DFX MiniGL.
EDIT - it seems that the latest MiniGL versions are optimized for an Athlon CPU, so something older might work better for you.

Installing Windows 95 instead of 98SE might also provide a slight boost on your system.

Using Audigy drivers with a Sound Blaster Live
Installing DOS drivers on an Audigy2 ZS
OPL3 vs. ESFM vs. CQM vs. SBLive
OPTi 82C930 review

Reply 23 of 67, by Chadti99

User metadata
Rank Member
Rank
Member
Joseph_Joestar wrote on 2020-07-31, 12:39:

It might be worthwhile to try out a few different versions of 3DFX MiniGL.
EDIT - it seems that the latest MiniGL versions are optimized for an Athlon CPU, so something older might work better for you.

Installing Windows 95 instead of 98SE might also provide a slight boost on your system.

Thanks for the recommendations I’ll def test these out!

Reply 24 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

So 800x600 runs about the same, at least in timedemo 1, maybe 1fps lower.

I went ahead and tried SLI, why not right? I’d read here there isn’t any performance improvements and with glQuake I can confirm, maybe even hurting performance slightly. I’m sure the “POD100” is struggling too feed both cards fast enough. One interesting note is that I can start GLQuake in 1024x768, with SLI, but the whole system shuts off after a minute or less. I’m thinking the two Voodoo’s are drawing too many amps over the PCI bus, causing the shutdown, but I’m not sure.

Reply 25 of 67, by DosFreak

User metadata
Rank l33t++
Rank
l33t++

Any difference with QDOS vs orginal and glide vs software?
http://dk.toastednet.org/QDOS/

DOSBox Compilation Guides
DosBox Feature Request Thread
PC Game Compatibility List
How To Ask Questions The Smart Way
Running DRM games offline

Reply 26 of 67, by Chadti99

User metadata
Rank Member
Rank
Member
DosFreak wrote on 2020-08-01, 00:55:

Any difference with QDOS vs orginal and glide vs software?
http://dk.toastednet.org/QDOS/

Can I read anymore info around QDOS somewhere? Maybe a tutorial? Sounds like something I’d like to try!

Not sure what you mean exactly by glide vs software. In DOS I was able to get ~22fps with a tseng labs et6000 and the POD83 clocked to 100 MHz. With GLQuake and a Voodoo 2 I’m getting 39fps with the same overclocked POD.

Reply 27 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

Seemed to have found a stable combo of jumpers and memory timings to reach 22.5fps in DOS Quake on the tseng labs et6000 with the POD at 100 MHz.

Acquired a Rendition Vérité based GPU to test out VQuake, something I’ve wanted to do since seeing the magazine ads. Hitting 23.7 FPS on the POD100. Vquake has its own unique look and it’s interesting to see the gulf of performance between Vérité and 3DFX.

Attachments

Reply 29 of 67, by Chadti99

User metadata
Rank Member
Rank
Member
Socket3 wrote on 2020-08-10, 16:41:
Try adding a voodoo 2 card :P […]
Show full quote

Try adding a voodoo 2 card 😜

TAn1wz8l.jpg

Sound enabled of course. I'm not running the most relaxed timings either. In software mode I get 19.2fps.

Yes the performance delta is incredible, the main reason 3DFX gained such popularity. I’m hitting 32fps at 800x600 with sound on with a POD100. I tried two in SLI but as other have mentioned on here it didn’t make a difference, in GlQuake anyway.

Attachments

Reply 30 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

Discovered a -nodma option For vQuake and picked up another few FPS!

Attachments

Reply 32 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

This 50ns EDO stick has proven stable, no crashes yet, was not expecting that on a socket 3 board. Pure software Quake is up to 23.9! Almost playable 🤣.

GLQuake on a Voodoo 2 is hitting 41fps, quite playable! Although a bit slower with sound on.

Attachments

Reply 33 of 67, by Chadti99

User metadata
Rank Member
Rank
Member

Moved everything over to a Biostar socket 3 board and squeezed out another half frame in software Quake.

Attachments

Reply 35 of 67, by vetz

User metadata
Rank l33t
Rank
l33t

You've beaten my Quake score with the POD@100mhz which has been a record holder for a long time (POD builds). I got 23,6 fps in a VLB system, but with 60ns FPM RAM and version 1.06 of Quake.

https://docs.google.com/spreadsheets/d/1lvF9n … wpNU/edit#gid=7 (row 27)

3D Accelerated Games List (Proprietary APIs - No 3DFX/Direct3D)
3D Acceleration Comparison Episodes

Reply 36 of 67, by Chadti99

User metadata
Rank Member
Rank
Member
chrismeyer6 wrote on 2020-09-01, 13:01:

How is the actual game play on your system?

It’s playable but GLQuake runs so much better I’ve been playing that mostly on this system. I imagine the last level of episode 3 would be a struggle in the large outdoor area.

Last edited by Chadti99 on 2020-09-02, 03:57. Edited 1 time in total.

Reply 37 of 67, by Chadti99

User metadata
Rank Member
Rank
Member
vetz wrote on 2020-09-01, 13:37:

You've beaten my Quake score with the POD@100mhz which has been a record holder for a long time (POD builds). I got 23,6 fps in a VLB system, but with 60ns FPM RAM and version 1.06 of Quake.

https://docs.google.com/spreadsheets/d/1lvF9n … wpNU/edit#gid=7 (row 27)

That’s cool to know! For some reason I thought folks were getting 25fps on these POD builds. Have you tried 1.08?

Reply 38 of 67, by PC-Engineer

User metadata
Rank Member
Rank
Member

I have got 26.8fps with V1.08. with the system in my signature. With SIS496 PCI setup i have got 26.6fps and with a UMC8881 setup 26.3fps - all with 1024kB cache.

But the SV2GX4 had DMA issues with L1WB and some Windows 95 issues with transparent mode - so i switched back to L1WT and synchronize mode. With this setup i have got 25.1fps

The S3 864 was a little bit faster than the ET4000 - 26.8fps vs. 26.6fps. Maybe with a ARK VLB card or a Diamond Trio64 VLB you can gain a little bit higher values.

AND the version 1.08 has a improvement from appr. 7% over the 1.06. So the 26.8fps in V1.08 should result in 25.1fps in V1.06

1994/1995 - Socket3 - ASUS SV2GX4 / POD 100MHz / 64MB / SCSI - Windows 95

Reply 39 of 67, by feipoa

User metadata
Rank l33t++
Rank
l33t++

The variance in DOS results is due to a few factors.

a) Some people use v1.08 instead of the v1.06 demo version. v1.08 returns higher results.
b) Some people are using Phils benchmark pack, which if I recall right, has made some adjustments to settings which alters results.
c) Other installed hardware, particularly USB, and to a lesser extent network, cards may reduce the benchmark results.

Before Phils pack, most people would benchmark using v1.06 demo at 320x200 with sound disabled. After Phils pack, there has been more uncertainty in what people are using. I prefer the original approach, that is v1.06 demo with sound off, booting directly into DOS with few drivers loaded, no EMM386, or mouse drivers, etc.

Ultimate 486 Benchmark | Ultimate 686 Benchmark | Cyrix 5x86 Enhancements | 486 Overkill Graphics | Worlds Fastest 486