yhea, thats the IBM utility I can get running but when I change any values to enable a feature and then hit f3 everything just resets. the Peter Moss utility crashes on me when i try and run it

mine
speed: 3x clock
stepping level: 0
revision: 5

I'm running at 3.3 volts. been stable thus far.

I had issues with the IBM utility too. Initially it worked but after a while the changes I made would not save. I probably should have put more effort into diagnosing the problem but I moved on....

Are you bypassing your autoexec.bat and config.sys? Try the Peter Moss utility from the command line with my / switch options listed above. I would guess your crashing because your voltage is so low but I have never used your board/bios. I also don't see a list of your hardware in this thread. I would rip out your network card and sound card and run a bare bones system with only 1 RAM module being used.

the motherboard is kinda crap. i mean it has a lot of features but its a very picky board. unfortunately PCI 486 boards are pretty rare and this is the only one Ive ever come across and I haven't been able to stomach dropping 100+ on a better one on Ebay.

only hardware I have running is the video card and an awe32 for sound.

I would suggest never running 4 volts through this cool chip you have. Alternatively keep your eyes open for an interposer board similar to the one below that will give you 3.6V. The other alternative might be to install a voltage trimmer mod like the one on my board below. This means I can set the voltage to basically anything I like. Good motherboards are expensive 🙁

+1 vote for the Peter Moss util.
To my best knowledge, it does not work as a TSR (just sets the registers then quits) - so it is pointless trying loading it high.

I have confirmed that LSSER=off is the same as LSSER=0 and is the optimal setting.

I'm running at 3.3 volts. been stable thus far.

Then you should leave it there
No need to fuss with voltage mods or anything else
I figured it would work at 3.3

okay guys. got the Peter Moss utility working and even auto executing on boot. just place the line in the autoexec.bat file

I used the same settings as rgart suggested and got a small speed increase. I also went back and benched the AMD 5x86 133mhz to compare. results are mixed

AMD 5x86 @ 133mhz

3DBENCH = 75.2
PCPbench = 19.7
Doom = 42.58
Quake = 12.6

Cyrix 5x86 @ 120mhz

3DBENCH = 73.6
PCPbench = 22.3
Doom = 35.08
Quake = 14.9

so half my benches the AMD does better and half the Cyrix runs better. some are pretty close though. Doom really seems to dislike the lower speed the video card is getting with the Cyrix.

also Rgart I looked at the back of my chip. its marked 1995 Cyrix g5fb602h

soviet conscript wrote:

okay guys. got the Peter Moss utility working and even auto executing on boot. just place the line in the autoexec.bat file […]
Show full quote

okay guys. got the Peter Moss utility working and even auto executing on boot. just place the line in the autoexec.bat file

I used the same settings as rgart suggested and got a small speed increase. I also went back and benched the AMD 5x86 133mhz to compare. results are mixed

AMD 5x86 @ 133mhz

3DBENCH = 75.2
PCPbench = 19.7
Doom = 42.58
Quake = 12.6

Cyrix 5x86 @ 120mhz

3DBENCH = 73.6
PCPbench = 22.3
Doom = 35.08
Quake = 14.9

Good work! What was the issue with the Peter Moss utility crashing? You should add speedsys to your benchmarks, it is an amazing little program ( google "speedsys 4.78")

<my cyrix 120> - with Matrox Mystique and ECS UM8810 board.

3DBENCH (v1.0): 96.0
PCPbench /vgamode v1.97 (320x200): 20.7
Doom 2 Full Version v1.666 FPS: 39.8
Quake Share Ware v1.06 FPS (320x200): 15.4

soviet, did you get your PCI bus at 40 MHz? Or is it still running at 27 MHz? You can check it with a scope if you have one.

it was human error, 🤣. I just didn't know what I was doing.

my speedsys score is 68.02

also I use the same pcpbench ver but I use 3Dbench 1.0C.

I tried replacing my ark card with a maxtrox mystique just to compare. my results were mostly the same . actually were very slightly lower in a few tests.

I have a led on my machine that still says 133mhz and i don't know how to change it. between that and the annoying video card speed thing my board apparently does at 40mhz I really want a 133mhz c5x86 or more likely the IBM version that I can OC

feipoa wrote:

soviet, did you get your PCI bus at 40 MHz? Or is it still running at 27 MHz? You can check it with a scope if you have one.

na, still running at 27mhz

Your motherboard is holding you back. Try the jumper switch trick noted earlier in the thread.

soviet conscript wrote:

I have a led on my machine that still says 133mhz and i don't know how to change it. between that and the annoying video card speed thing my board apparently does at 40mhz I really want a 133mhz c5x86 or more likely the IBM version that I can OC

Most likely a rather inconveniently placed lot of jumpers behind the front LED panel. Yeah those chips are not easy to come by!

feipoa wrote:

Your motherboard is holding you back. Try the jumper switch trick noted earlier in the thread.

i'm a little unsure how to do this and I don't want to damage the MB or the chip so let me see if I have this correctly.

so I think my case has a turbo switch. I should set this to the jumper that current off (jp3a). I boot the machine in 33mhz mode, my cpu will be 100mhz at this point since its 33x3. after i'm booted I press the turbo button and that should set the machine to 40mhz thus kicking the cpu to 120mhz 40x3 but my PCI bus will remain at 33mhz?

The PCI divider will stay the same (PCI = FSB instead of PCI = FSB/2) so you will get 40mhz instead of 20mhz.
That's what I quoted: "It seems the PCI speed is 20 MHz after a 160 boot and 33 MHz when booted in the 133 mode. HOT switching to 160 MHz must keep the dividers disabled and you end up with a actual 40 MHz PCI bus speed."

Admittedly that's a bit high, in Pentium boards 37.5mhz is often done when overclocking to 75mhz FSB. I don't this might damage anything, but it is very well possible that not all PCI cards work with that speed. It might be necessary to use a different video card. Beware of data corruption when writing to your hard drive!

Let me walk you through it in a little more detail. Your turbo switch on the case has a connector on the end of the cable. It is probably a 3-pin connector. You only need to use two of those pins. Pick any two adjacent pins. The switch connects and disconnects those pins at the end of the switch when you push the button. The button's pushed in position will either be with the pins connected or disconnected, depending on which two adjacent pins you select.

On the M919 motherboard, when JP3A has a jumper on it, the FSB is at 33 MHz. When the jumper is removed, the FSB goes to 40 MHz. The turbo switch can take the place of the jumper. Be sure to leave JP3B and JP3C jumpered. Determine if the out or in button position is with the pins are connected or open. What idspispopd was suggesting is to boot the computer with JP3A switched closed (BIOS should detect an AMD 5x86-133 on startup). Then once you past the POST screen (and presumably the computer is idle), hit the switch and the FSB will run at 40 MHz. You should now have a PCI bus at 40 MHz because you have now passed the CMOS's startup routine which was trying to automatically adjust your PCI bus.

idspispopd, on this motherboard, the PCI bus does not get divided in half, it is multiplied by 2 and divided by 3.

success! I jumpered the turbo switch to jp3a. with the button undepressed it booted at 100mhz when I got to my c:\ prompt I pressed my turbo switched and ran some checks to confirm CPU was now running at 120mhz. seems to have worked. new benchmark results

3dbench= 95.0
pcpbench= 22.3
doom= 48.63
quake= 15.0

very minor or no change with both quake and pcpbench but the 2 benches that were pretty low before shot up. Doom bench is now finally beating the AMD 133mhz

Congrats soviet conscript. Nice bench results!

Congratulations! Now that's a turbo button that deserves it's name. It makes sense that the benchmarks that have higher fps receive a bigger boost since they are more dependant on transfer speed.
As I said, look out for instability. I don't think anything could be damaged, but 40 MHz PCI is quite an overclock. I don't know what onboard components could be affected. Also check if ISA clock is affected, too.
Even on 440BX boards with 133MHz FSB the PCI bus is usually not overclocked, only AGP.