some youbers raced that score, highest was somewhere around 17 fps.

That's nice 👌

https://youtu.be/lyvmn1CGQgM?is=U6-vTIARL9aS2rbV
This is my video on 200 mhz with peltier cooling 🙂

Here 2 pictures from the cache module.

Leo1976 wrote on 2026-04-09, 19:27:

Does anybody now what is the highest score in Quake timedemo on a 486 with an amd x5 133 mhz cpu overclocked?

When you ask about the highest overclocked DOS Quake timedemo score on an Am5x86-133, do all you care about is one successful run of DOS Quake timedemo, or does the system need to be fully stable with the vast majority of software from that era? For example, if someone records 23 fps at 180/200 MHz, but the system won't run Windows 9x, does the Quake score still count? Or, what if the system will boot Windows 95, but there are some software titles which crash the system, yet the afflicted software ran fine at a lower frequency - do you still care about the Quake score? For my past benchmarking, I only gave credit to scores on a system which was broadly stable when run on a hot summer day.

Pentium POD is not the (486) better cpu for quake?? I remember qake was eavely optimized for Pentium cpu

Leo1976 wrote on 2026-05-27, 08:38:

Here 2 pictures from the cache module. […]
Show full quote

Here 2 pictures from the cache module.

Do you have gerber files? I'd love to build one. Which chips did you use? I can't quite make them out from the photos.
Did you do any mods to the clocker or anything else on the board?

AlessandroB wrote on 2026-05-27, 15:53:

Pentium POD is not the (486) better cpu for quake?? I remember qake was eavely optimized for Pentium cpu

Correct, the Pentium's strong FPU gives it a huge advantage over most 486s; IIRC, the POD is closer to a P54C than the original 5V P5 Pentium.

That said, I've no idea how well it would compete against a 200MHz clocked 5x86.

feipoa wrote on 2026-05-27, 11:58:

Leo1976 wrote on 2026-04-09, 19:27:

Does anybody now what is the highest score in Quake timedemo on a 486 with an amd x5 133 mhz cpu overclocked?

When you ask about the highest overclocked DOS Quake timedemo score on an Am5x86-133, do all you care about is one successful run of DOS Quake timedemo, or does the system need to be fully stable with the vast majority of software from that era? For example, if someone records 23 fps at 180/200 MHz, but the system won't run Windows 9x, does the Quake score still count? Or, what if the system will boot Windows 95, but there are some software titles which crash the system, yet the afflicted software ran fine at a lower frequency - do you still care about the Quake score? For my past benchmarking, I only gave credit to scores on a system which was broadly stable when run on a hot summer day.

My goal is to run all dos benchmarks and windows 98 with GLquake and maybe another game. I have no plans for using it for long term.

DangerMouse wrote on 2026-05-27, 21:11:

Leo1976 wrote on 2026-05-27, 08:38:

Here 2 pictures from the cache module. […]
Show full quote

Here 2 pictures from the cache module.

Do you have gerber files? I'd love to build one. Which chips did you use? I can't quite make them out from the photos.
Did you do any mods to the clocker or anything else on the board?

I have no gerber files. A good friend of my made the stick. Chips he used are: IDT 71V124 SA10Y K0150M.
I didn't do any mods to the board. I put the voltage at 4v. and the fsb and pci clock to 66 mhz.

AlessandroB wrote on 2026-05-27, 15:53:

Pentium POD is not the (486) better cpu for quake?? I remember qake was eavely optimized for Pentium cpu

That's true. Only its not a real 486 cpu.
And indeed quake is making use of the floating point unit (FPU). That's why the POD is faster.

Dan386DX wrote on 2026-05-27, 21:35:

AlessandroB wrote on 2026-05-27, 15:53:

Pentium POD is not the (486) better cpu for quake?? I remember qake was eavely optimized for Pentium cpu

Correct, the Pentium's strong FPU gives it a huge advantage over most 486s; IIRC, the POD is closer to a P54C than the original 5V P5 Pentium.

That said, I've no idea how well it would compete against a 200MHz clocked 5x86.

POD is for sure a bit faster.

Dan386DX wrote on 2026-05-27, 21:35:

Correct, the Pentium's strong FPU gives it a huge advantage over most 486s; IIRC, the POD is closer to a P54C than the original 5V P5 Pentium.

I guess that is why it was much delayed and the spec changed, someone finally figured out it wasn't a good idea to sic a 15W P5 core onto 486 voltage regulators designed for 5W CPUs. ... thus implemented from the lower power one.

Leo1976 wrote on 2026-05-28, 14:59:

Dan386DX wrote on 2026-05-27, 21:35:

AlessandroB wrote on 2026-05-27, 15:53:

Pentium POD is not the (486) better cpu for quake?? I remember qake was eavely optimized for Pentium cpu

Correct, the Pentium's strong FPU gives it a huge advantage over most 486s; IIRC, the POD is closer to a P54C than the original 5V P5 Pentium.

That said, I've no idea how well it would compete against a 200MHz clocked 5x86.

POD is for sure a bit faster.

Yes and I completely forgot, thanks to people here, there's proof, it's top of the shop in Quake with 20.8FPS. It does lose out to some of the fastest 5x86s in integer performance though.

It's all academic as you say, it's not a 486 anyway.

21.2 on mine @ 180mhz. Score goes up a tad at 200 but since I normally use it at 180 that's the "official" score for me

M919, 1mb L2, matrox g200

Leo1976 wrote on 2026-05-28, 14:47:

feipoa wrote on 2026-05-27, 11:58:

Leo1976 wrote on 2026-04-09, 19:27:

Does anybody now what is the highest score in Quake timedemo on a 486 with an amd x5 133 mhz cpu overclocked?

When you ask about the highest overclocked DOS Quake timedemo score on an Am5x86-133, do all you care about is one successful run of DOS Quake timedemo, or does the system need to be fully stable with the vast majority of software from that era? For example, if someone records 23 fps at 180/200 MHz, but the system won't run Windows 9x, does the Quake score still count? Or, what if the system will boot Windows 95, but there are some software titles which crash the system, yet the afflicted software ran fine at a lower frequency - do you still care about the Quake score? For my past benchmarking, I only gave credit to scores on a system which was broadly stable when run on a hot summer day.

My goal is to run all dos benchmarks and windows 98 with GLquake and maybe another game. I have no plans for using it for long term.

I have surprisingly good results with long term stability at 180mhz under windows 98 with whatever I can throw at it. You just need to find a good cpu - if you are able to buy a lot of 20 (or better yet 30) you will probably find one, or hey you may get lucky and get a gooder 2-3 in.

bertrammatrix wrote on 2026-05-28, 19:59:

Leo1976 wrote on 2026-05-28, 14:47:

feipoa wrote on 2026-05-27, 11:58:

When you ask about the highest overclocked DOS Quake timedemo score on an Am5x86-133, do all you care about is one successful run of DOS Quake timedemo, or does the system need to be fully stable with the vast majority of software from that era? For example, if someone records 23 fps at 180/200 MHz, but the system won't run Windows 9x, does the Quake score still count? Or, what if the system will boot Windows 95, but there are some software titles which crash the system, yet the afflicted software ran fine at a lower frequency - do you still care about the Quake score? For my past benchmarking, I only gave credit to scores on a system which was broadly stable when run on a hot summer day.

My goal is to run all dos benchmarks and windows 98 with GLquake and maybe another game. I have no plans for using it for long term.

I have surprisingly good results with long term stability at 180mhz under windows 98 with whatever I can throw at it. You just need to find a good cpu - if you are able to buy a lot of 20 (or better yet 30) you will probably find one, or hey you may get lucky and get a gooder 2-3 in.

21.8 fps is the best long-term stable I could achieve with an Am5x86. Leo1976 shows a video indicating 23.7 fps in Quake, which is beyond any results I've seen so far. I was wondering if there's a timer malfunction. I doubt his system is fully stable. Didn't you get 23.4 fps, but the system wasn't fully stable?

feipoa wrote on 2026-05-28, 21:01:

bertrammatrix wrote on 2026-05-28, 19:59:

Leo1976 wrote on 2026-05-28, 14:47:

My goal is to run all dos benchmarks and windows 98 with GLquake and maybe another game. I have no plans for using it for long term.

I have surprisingly good results with long term stability at 180mhz under windows 98 with whatever I can throw at it. You just need to find a good cpu - if you are able to buy a lot of 20 (or better yet 30) you will probably find one, or hey you may get lucky and get a gooder 2-3 in.

21.8 fps is the best long-term stable I could achieve with an Am5x86. Leo1976 shows a video indicating 23.7 fps in Quake, which is beyond any results I've seen so far. I was wondering if there's a timer malfunction. I doubt his system is fully stable. Didn't you get 23.4 fps, but the system wasn't fully stable?

Yeah I think it was 23.4, I have a test setup sifting on top of that particular system ATM so I don't feel like moving it to switch it to 200. I may have a photo somewhere. It would hang at windows shutdowns at 200 but seemed stable otherwise.

feipoa wrote on 2026-05-28, 21:01:

bertrammatrix wrote on 2026-05-28, 19:59:

Leo1976 wrote on 2026-05-28, 14:47:

My goal is to run all dos benchmarks and windows 98 with GLquake and maybe another game. I have no plans for using it for long term.

I have surprisingly good results with long term stability at 180mhz under windows 98 with whatever I can throw at it. You just need to find a good cpu - if you are able to buy a lot of 20 (or better yet 30) you will probably find one, or hey you may get lucky and get a gooder 2-3 in.

21.8 fps is the best long-term stable I could achieve with an Am5x86.

In that case, have I misunderstood your ultimate comparison results where the POD achieved what looked like the best scores at 20.8 stock + 24.4 @ 100?

Or did you manage the 21.8 a bit later?