VOGONS


Reply 41 of 52, by GL1zdA

User metadata
Rank Oldbie
Rank
Oldbie
VooDooMan wrote:

You are welcome 😀 What other games from 1999-2002 are good benchmarks as well?

I'm a fan of 3D GameGauge - they specified the games they've used. The games rotated in and out as they were release.

3D GameGauge 2.0 in 1999: https://www.tgdaily.com/hardware-brief/19581- … asures-3d-games
Madden' 99, Unreal, Motorhead, Descent 3 and Half-Life.

3D GameGauge 2.5.1 in 2001: https://www.extremetech.com/computing/50468-a … 07500-preview/6
3D GameGauge 2.5.1 is a suite of seven games, one OpenGL (Quake 3), and six Direct3D (UT, Re-Volt, Expendable, Panzer Elite, NFL Fever and Jane’s USAF).

3D GameGauge 3.0 was released in 2002: https://www.extremetech.com/computing/51156-e … gauge-30-debuts
Commanche 4, Dungeon Siege, IL-2 Sturmovik, Jedi Knight II, NASCAR 2002, NHL 2002, Serious Sam SE, Unreal Performance Test

getquake.gif | InfoWorld/PC Magazine Indices

Reply 42 of 52, by slivercr

User metadata
Rank Member
Rank
Member
VooDooMan wrote:

That sound AWESOME! Will you be able to make those tests with the same CPU that I used? 😜 Don't worry about the time, it is the least important factor right now.

I actually have a couple of SL5XL in my stash, I'll look around for SL6BY. The test will get done eventually though 😉

Outrigger: an ongoing adventure with the OR840
QuForce FX 5800: turn your Quadro into a GeForce

Reply 43 of 52, by silikone

User metadata
Rank Member
Rank
Member
VooDooMan wrote:

You are welcome 😀 What other games from 1999-2002 are good benchmarks as well?

Perhaps that Crytek X-Isle demo? Certainly a mouthful for a CPU when there's no hardware T&L.
Assuming it can run with that hardware at all. If it doesn't work on Ati, then there sure is some impressive backwards compatibility since I can run it on a GTX 760 right now.

Do not refrain from refusing to stop hindering yourself from the opposite of watching nothing other than that which is by no means porn.

Reply 44 of 52, by matcarfer

User metadata
Rank Newbie
Rank
Newbie

Wow, impressive. I currently have a Soyo SY-7VBA133 (VIA Apollo Pro133A) and have stability issues using some AGP cards. Im currently aiming for a MSI 815E Pro (Intel 815, with AGP 3,3V and ISA!). Thanks!

Reply 45 of 52, by matcarfer

User metadata
Rank Newbie
Rank
Newbie
GL1zdA wrote:
I'm a fan of 3D GameGauge - they specified the games they've used. The games rotated in and out as they were release. […]
Show full quote
VooDooMan wrote:

You are welcome 😀 What other games from 1999-2002 are good benchmarks as well?

I'm a fan of 3D GameGauge - they specified the games they've used. The games rotated in and out as they were release.

3D GameGauge 2.0 in 1999: https://www.tgdaily.com/hardware-brief/19581- … asures-3d-games
Madden' 99, Unreal, Motorhead, Descent 3 and Half-Life.

3D GameGauge 2.5.1 in 2001: https://www.extremetech.com/computing/50468-a … 07500-preview/6
3D GameGauge 2.5.1 is a suite of seven games, one OpenGL (Quake 3), and six Direct3D (UT, Re-Volt, Expendable, Panzer Elite, NFL Fever and Jane’s USAF).

3D GameGauge 3.0 was released in 2002: https://www.extremetech.com/computing/51156-e … gauge-30-debuts
Commanche 4, Dungeon Siege, IL-2 Sturmovik, Jedi Knight II, NASCAR 2002, NHL 2002, Serious Sam SE, Unreal Performance Test

Where to download GameGauge? It seems impossible.

Reply 46 of 52, by pshipkov

User metadata
Rank l33t
Rank
l33t

Good stuff.
Shared info describes pretty well the perf-to-clock ratio between the different chipsets and motherboards.
Some of the mobos can OC way past stock frequency which will change results a lot tho ...

retro bits and bytes

Reply 47 of 52, by biessea

User metadata
Rank Member
Rank
Member

Hi there VoodooMan,

great project and great great work you do! Congratulations! I really appreciated!

My question is:

Do you use the ASUS CUV266 with a Tualatin model where the VIA chipset VT8633 isn't native for that processors?
I ask you this cause I really need to know if I can upgrade my Pentium 3 that actually have, a 1GHZ version with a Tualatin one, not just for the speed but expecially for the cache doubled; I think I will have a great gain in games and in normal use on my Windows98SE Voodoo2 SLI project.

Can you tell me if it is worked or you used that processors in a 266T motherboard that isn't the CUV266 that you photographed?

Please dissipate my doubts, cause if only I know that a Tualatin can be compatibile with my rare motherboard I will sure bought it!

THANKS A LOT!

Computer lover since 1992.
Love retro-computing, retro-gaming, high-end systems and all about computer-tech.
Love beer, too.

Reply 48 of 52, by pshipkov

User metadata
Rank l33t
Rank
l33t

Small contribution to this excellent thread.
262 fps in Q3A on Asus P2B-D 1575/150 with Quadro FX4000 (details here).

Last edited by pshipkov on 2022-10-08, 05:47. Edited 2 times in total.

retro bits and bytes

Reply 49 of 52, by Doornkaat

User metadata
Rank l33t
Rank
l33t
biessea wrote on 2021-02-18, 09:37:
Hi there VoodooMan, […]
Show full quote

Hi there VoodooMan,

great project and great great work you do! Congratulations! I really appreciated!

My question is:

Do you use the ASUS CUV266 with a Tualatin model where the VIA chipset VT8633 isn't native for that processors?
I ask you this cause I really need to know if I can upgrade my Pentium 3 that actually have, a 1GHZ version with a Tualatin one, not just for the speed but expecially for the cache doubled; I think I will have a great gain in games and in normal use on my Windows98SE Voodoo2 SLI project.

Can you tell me if it is worked or you used that processors in a 266T motherboard that isn't the CUV266 that you photographed?

Please dissipate my doubts, cause if only I know that a Tualatin can be compatibile with my rare motherboard I will sure bought it!

THANKS A LOT!

He's using the board with an adaptor as he states in his post.
A Voodoo2 SLI system isn't going to gain much performance from a Tualatin chip though, so you're not really missing out.

Reply 50 of 52, by biessea

User metadata
Rank Member
Rank
Member
Doornkaat wrote on 2021-02-23, 05:19:
biessea wrote on 2021-02-18, 09:37:
Hi there VoodooMan, […]
Show full quote

Hi there VoodooMan,

great project and great great work you do! Congratulations! I really appreciated!

My question is:

Do you use the ASUS CUV266 with a Tualatin model where the VIA chipset VT8633 isn't native for that processors?
I ask you this cause I really need to know if I can upgrade my Pentium 3 that actually have, a 1GHZ version with a Tualatin one, not just for the speed but expecially for the cache doubled; I think I will have a great gain in games and in normal use on my Windows98SE Voodoo2 SLI project.

Can you tell me if it is worked or you used that processors in a 266T motherboard that isn't the CUV266 that you photographed?

Please dissipate my doubts, cause if only I know that a Tualatin can be compatibile with my rare motherboard I will sure bought it!

THANKS A LOT!

He's using the board with an adaptor as he states in his post.
A Voodoo2 SLI system isn't going to gain much performance from a Tualatin chip though, so you're not really missing out.

Ok, perfect.
Thank you a lot.

Computer lover since 1992.
Love retro-computing, retro-gaming, high-end systems and all about computer-tech.
Love beer, too.

Reply 51 of 52, by overdrive333

User metadata
Rank Newbie
Rank
Newbie

Great project, thank you very much. As expected, 440bx is king.
But there is an opinion that the best chipset for P3 is... nvidia NV2A.
Dual-channel ddr memory operating at 400mhz (like nforce 2 ultra 400!), integrated gf4 ti4200 level graphics (at low frequencies). It is a pity that we did not see this logic in desktops.

Reply 52 of 52, by dismuter

User metadata
Rank Newbie
Rank
Newbie

Much respect for the dedication. 4 years later, still people like me stumbling upon that 😀

fitzpatr wrote:

I have a personal preference for the y-axis to start at zero so that differences aren't exaggerated, but that's a personal thing.

Actually, it's not a personal thing. Using a 0-based Y axis is objectively the only way that benchmark information comparing different systems should be presented to be meaningful.

VooDooMan wrote on 2019-01-01, 20:31:

No problem for me to make all charts with y-axis starting from 0, but what is the point? The differences in tests would be almost unnoticeable...

You've already done a lot of work, don't worry about it.

But the almost unnoticeable aspect of the differences is valuable information that you removed. It means that the differences could be within the margins of error, or that they should be of little concern.
The added value of a bar chart compared to a sorted table for comparison of results from different systems, is precisely to give at first glance an idea of the scale of difference between results.
If you arbitrarily set the start of the Y-axis, you present the highest values as being 2, 3, 4... potentially 1000 times greater than the lowest ones even when there could be only 0.1% difference, and that not only defeats the only purpose of the chart, but also makes it misleading.