Thanks everyone 😀 I've just updated my previous post with the missing 'stock clock' results.

Unfortunately I don't have Forsaken but I'll see if I can get hold of a copy and test it with this rig 😀

Regarding UT, I guess all of us are using the default intro flyby to bench the game?, right? I suggest that we timedemo UTBench instead to get more real-world results of how our systems will perform under a realistic gaming scenario with a reasonable amount of players blasting each other across an entire map 😀

Last week I got Tualatin P3-S 1.4GHz, now I wish I had Voodoo 5 😁

subhuman@xgtx wrote:

Unfortunately I don't have Forsaken

It's d3d only so you not lost anything.

Hey Phil, do you have the diamond drivers for that diamond monster sound vortex 2? I did some quick searches and couldn't find any except reference drivers.
I happened to find the same card in a PC I gutted.

Sorry for the utter crap quality, these are my two main retro systems. In the center I've got the Tualatin, on which I did the tests, and on the right the K6-III+ with a Voodoo3. There's also an Athlon 64 system on the left that I fool around with sometimes. Maybe I'll open up a thread with much higher res and in-depth photos at some point.

raymangold wrote:

Hey Phil, do you have the diamond drivers for that diamond monster sound vortex 2? I did some quick searches and couldn't find any except reference drivers.
I happened to find the same card in a PC I gutted.

They are on the Vogons driver archive 😀

But for this system, I used the latest reference drivers. 2048.

Work in progress for the challenge:

Pentium III-S 1400MHz (Tualatin-512k) SL6BY
512MB SD133 2-3-2-8
Asus TUSL2-C
Voodoo 5 5500 64MB AGP 2x

Great machine!

Phil, can you do a full CPU scaling write up like you did for the voodoo^2?

blank001 wrote:

Phil, can you do a full CPU scaling write up like you did for the voodoo^2?

😊 Sorry, but no. That paper was a once in a life project 😵

philscomputerlab wrote:

blank001 wrote:

Phil, can you do a full CPU scaling write up like you did for the voodoo^2?

😊 Sorry, but no. That paper was a once in a life project 😵

A very nice one, that made me buy a powerleap adapter and put a Tualatin Celeron 1.1GHz into an ASUS P2B (i440bx), because it'd push the Voodoo2 SLI much more than the PIII 500MHz it had originally installed. 😉

philscomputerlab wrote:

blank001 wrote:

Phil, can you do a full CPU scaling write up like you did for the voodoo^2?

😊 Sorry, but no. That paper was a once in a life project 😵

You need to find local apprentices to do the grunt work of swapping CPU's and boards and such. Then you can just write up and analyze the data.

Well, the thing is, I love that grunt work 😊

Just not THAT MUCH 🤣

I can't remember how long it took me, but towards the end it felt like work and couldn't wait for it to be finished. With other projects I'm a bit gutted when it's done because it means I got to find a new one. But that's ok 😀

In terms of V5 scaling, I believe we can all see the trend. You don't have a V5 and play at 640 x 480. Interesting to see differences in processing power, but most will play at 1024 x 768 minimum. At that resolution, I find a Pentium with 1 GHz or so to be fast enough. But a Tualatin, IMO, is THE perfect match for a V5, even though Athlon is faster, it just doesn't feel "right" 🤣

This topic is an oooold one 😁
Phil that research that you did on the Voodoo 2 scaling was badass! Kudos to you!

This is what I dug up:

it was made in 2004 by VoodooAlert.de forum member Friday_13th

Reading all the posts and seeing all the nice data just makes me wish that my missing parts were here already. I spent a small fortune on it, but being a Voodoo 5 owner is something special and I'm so happy I can be a part of it.

I just unboxed a V5 that I had purchased last year, so here are the benchmarks from my unexotic but stable SE440BX-2 system.

INTEL SE440BX-2
PIII-1000 SLOT 1 COPPERMINE
VOODOO 5 5500 AGP (1.04 DRIVERS)
AUREAL VORTEX 2 (AU8830)
256 MB GENERIC PC100 SDRAM
160GB WD CAVIAR SE HDD
WINDOWS 98SE

UNREAL GOLD (framerates taken from the third cycle)
800x600 : 113.7 (105.9)*
1024x768 : 99.1 (95.0)*
1600x1200 : 51.9 (52.5)*

* I forgot I had turned on the "Use3DHardware" option for the A3D in unreal.ini as well as bumping the sample rate to 48000Hz. These are the scores in parens.

UT99 GOTY (framerates taken from the third cycle)
800x600 : 73.25
1024x768 : 71.85
1600x1200 : 52.14

QUAKE 2 3.20 default OpenGL (framerates taken from third cycle)
800x600 : 163.6
1024x768 : 152.0
1600x1200 : 66.3

GLQuake 1.09/0.97 (framerates taken from third cycle)
800x600 : 194.9
1024x786 : 127.3
1600x1200 : 57.5

Enabling A3D and boosting the sampling rate is very good, but I didn't do this for this project. Actually I forgot about it 😊 Everything is default, so to make it easier to compare.

Nice system BTW.

philscomputerlab wrote:

Enabling A3D and boosting the sampling rate is very good, but I didn't do this for this project. Actually I forgot about it 😊 Everything is default, so to make it easier to compare.

Nice system BTW.

I thought I would include those rates since I had to rerun the tests with the default settings anyway. Love A3D. Have a Quadzilla on the way.

I will post another update once I get my Tualeron 1300 set up with the Powerleap adapter.

boxpressed wrote:

I just unboxed a V5 that I had purchased last year, so here are the benchmarks from my unexotic but stable SE440BX-2 system. […]
Show full quote

I just unboxed a V5 that I had purchased last year, so here are the benchmarks from my unexotic but stable SE440BX-2 system.

INTEL SE440BX-2
PIII-1000 SLOT 1 COPPERMINE
VOODOO 5 5500 AGP (1.04 DRIVERS)
AUREAL VORTEX 2 (AU8830)
256 MB GENERIC PC100 SDRAM
160GB WD CAVIAR SE HDD
WINDOWS 98SE

UNREAL GOLD (framerates taken from the third cycle)
800x600 : 113.7 (105.9)*
1024x768 : 99.1 (95.0)*
1600x1200 : 51.9 (52.5)*

* I forgot I had turned on the "Use3DHardware" option for the A3D in unreal.ini as well as bumping the sample rate to 48000Hz. These are the scores in parens.

UT99 GOTY (framerates taken from the third cycle)
800x600 : 73.25
1024x768 : 71.85
1600x1200 : 52.14

QUAKE 2 3.20 default OpenGL (framerates taken from third cycle)
800x600 : 163.6
1024x768 : 152.0
1600x1200 : 66.3

GLQuake 1.09/0.97 (framerates taken from third cycle)
800x600 : 194.9
1024x786 : 127.3
1600x1200 : 57.5

OK, I finally got my Celeron Tualatin 1300 working with the Powerleap and SE440BX-2, so here are the results from that combo:

UNREAL GOLD (framerates taken from the third cycle)
800x600 : 126.8
1024x768 : 106.4
1600x1200 : 51.9

UT99 GOTY (framerates taken from the third cycle)
800x600 : 83.55
1024x768 : 81.36
1600x1200 : 53.49

QUAKE 2 3.20 default OpenGL (framerates taken from third cycle)
800x600 : 189.2
1024x768 : 157.9
1600x1200 : 66.8

GLQuake 1.09/0.97 (framerates taken from third cycle)
800x600 : 194.7
1024x768 : 127.3
1600x1200 : 57.9

With the exception of GLQuake, gains at lower resolutions and nothing at 1600x1200. Not sure why GLQuake stayed the same at every resolution.

Cool!

I think you should send me all your 1+TB drives. I'll gladly send you lots of 80/160GB SATAs 😁