Intel486dx33 wrote on 2021-09-02, 17:39:

Chadti99 wrote on 2021-08-07, 13:10:

Face palm you guys, plugged an EDO stick into the LS486E and now I’m hitting 19.4, same as the 8433UUD under stable settings.

I know it’s not news to anyone but here are the speedsys results between FPM and EDO on the LS486.

Can you post a photo of your motherboard jumper setting ?
Mine is running at 160mhz.

I’m quite sure these were it.

I ran some initial tests with an IBM 5x86c-100HF at 150 MHz on the Biostar MB-8433UUD motherboard while using a few different peltier pads. I started with conservative DRAM and SRAM timings. The 51 W peltier at 12 V was insufficient. I upped it to 75 W at 12 V and was able to complete DOS Quake with the Cyrix register features disabled. Unfortunately, when I enabled all the features, Quake crashed half-way through the timedemo. Thus, it would seem that one or more of the Cyrix features isn't stable at 150 MHz. It would take quite a bit of time to narrow down which feature(s), but I decided that I did not want to risk frying the CPU and have ceased testing.

Vcc to the CPU was 3.75 V. I tried it up to 3.82 V, but there is still some Cyrix feature which doesn't cooperate with that operating frequency. For the record, using conservative DRAM/SRAM settings with with all of the Cyrix features disabled, the Quake score was 16.8 fps at 150 MHz. PCI was run at 2/3*50=33MHz.

By contrast, I was able to use the 51 W peltier to run an AMD DX2-66-WB at 180 MHz with just 3.45 V to the CPU. DOS Quake scores of the Am5x86-180 and the Cyrix 5x86-133/2x are the same at 19.2 fps w/256K.

EDIT: There seems to be some variation in the power at 12V for these peltiers. Some quote the max wattage of heat dissipated, while others are quoting the electrical current needed to run the peltier at max current. For clarity, I used a TEC1-12708 for the IBM 5x86c and a TEC1-12704 for the AMD at 180 MHz.

So it was similar to @120(2x60)MHz with Cyrix register features enabled. I had 16.9 fps in Quake with Matrox card and PCI 60 MHz (?).

IMO you just want your windows stable enhancements turned on.

I was wondering if anyone tried the Cx5x86 on a MB-8433UUD-A motherboard with a 3D accelerator other than Voodoo 1 such as the Rage Pro, Riva128 or i740?

I mean, I know that a Riva128 PCI works in DOS (I use it in my system) flawlessly but I wanted to know if it would be worth installing Win95c and nVidia drivers to see how this system would fare in early Direct3D/OpenGL games like GLQuake, Outlaws and the like..

with a 3D accelerator other than Voodoo 1 such as the Rage Pro, Riva128 or i740?

Terribly, if you can't disable Vsync, which is mostly double buffered, so performance will drop in big increments.

The Serpent Rider wrote on 2021-09-22, 12:43:

with a 3D accelerator other than Voodoo 1 such as the Rage Pro, Riva128 or i740?

Terribly, if you can't disable Vsync, which is mostly double buffered, so performance will drop in big increments.

RivaTuner should be able to disable VSync for pretty much everything IIRC.

RivaTuner is not designed to work with RIva 128 or early pre-detonator drivers for TNT.

Wanted to share that I was experimenting with a Tec1-12715 Peltier and my AMD5X86. This particular CPU is stable at 180MHz (60x3) with just a heat sink and fan, no thermal compound needed. It’s also fine at 3.45volts at that speed. I’m actually able to post at 200MHz (50x4) on this same setup so I’m hopeful I can achieve stable 200MHz operation at lower Peltier voltages. This does require 4v to the CPU.

Running the 12715 at 5v allows me to get into Dosbench and run a couple tests but none of the graphics tests.

If I run the peltier at 12V I think I’m heat soaking the heat sink and peltier, things get absurdly hot. So I’m looking for something bigger to cool this with and a way to dial in something in between 5-12v.

Notice how the Speedsys "score" of the Am5x86 at 200 MHz is still lower than an IBM 5x86c at 133 MHz.

feipoa wrote on 2021-10-18, 11:29:

Notice how the Speedsys "score" of the Am5x86 at 200 MHz is still lower than an IBM 5x86c at 133 MHz.

Cool

Trying to find it in the thread, I see one at 75@133 what did it hit at 150?

download/file.php?id=15677&mode=view
84.12

I can appreciate the worlds fastest 486 with a 586 processor?
but going up to 586 cpu means it still has 33mhz FSB like the 486.
Pentium overdrive had simular bottleneck fast cpu but fsb that wasn't sufficient to fully appreciate the power.
The Intel multimedia extensions were great on pentium class and that started getting interesting on socket 7.
Then they brought out the super 7 socket 7 mobo with 100mhz fsb that made socket 7 more comfortable.
Fastest isn't always the best.
Overclocking such old stuff is just suicide for old silicon. I mean look at a window or mirror after say 30 years they're made of sand too and they warp imagine silicon chips its gotta be worse than bad.

feipoa wrote on 2021-10-18, 14:00:

download/file.php?id=15677&mode=view
84.12

That is a nice score!

Feipoa, which version of the Cyrix 5x86 do you recommend for pushing the limits?

Whatever higher binned you can find within your budget. It's not something you can buy in bulk to find the perfect sample, like AMD 5x86.

I recommend the QFP variant of the IBM 5x86c. From my experience, they overclock a lot better than the ceramic variant.

Bumped up the voltage on the peltier to 7v, wiring positive to yellow and ground to red on a molex connector. Used some thermal paste between peltier and cooler, instead of thermal tape. And now we are talking! Things got frosty! Was able to pull off a Doom max run but no go with Quake so far.

Good stuff Chadti99.
All the right moves.
Thermal paste, 4 rods heatsink, high-enough voltage to the Peltier for freeze.
However the Doom score is too low. You should be over 70 fps already.
I think this is LuckyStar revision C2 motherboard that comes with 15ns L2 cache chips.
That most likely won't be enough for tight BIOS timings = lowered perf.
Do you have 12 or 10 ns L2 cache chips ?

However the Doom score is too low.

Doom needs overclocked PCI bus to achieve better scores.