VOGONS


First post, by Crinale0

User metadata
Rank Newbie
Rank
Newbie

I went to fit my Pentium MMX CPU (200mhz over a 166Mhz standard Pentium) and the little switches to set the FSB looked like they had never been touched. I carefully tried to set it up but...

20240824-143237.jpg

All but two of the banks snapped (why i carried on after god only knows but here we are)

Initially, how it was set would now make my Pentium 166Mhz run at 100Mhz and the Pentium MMX run at 233Mhz

20240824-145606.jpg

I don't want to overclock the CPU at all so i got a sharp edged knife and tried to move the banks that way. They were REALLY stiff and wouldn't move but i managed to coax it into running somewhat close to spec.

20240824-151416.jpg

How it's come up with this number I have no idea as there is no setting for 190/193Mhz in the manual but hey, its within spec so I've left it as is.

I did notice though that now it doesn't seem to be using a multiplyer at all but instead it's using the Time Stamp Counter?? What is this?

20240824-151512.jpg

There is a performance increase so, it does seem to be working as it should

20240821-183140.jpg

20240824-151606.jpg

But yeah, is it ok to be run like this? It's going to be running DOS games and I want to be able to use Setmul and other such tools to slow the machine down will i still be able to do this?

Should of just left it alone but nooo I had to have MMX like i did when i was a kid. But hey, here we are so... yeah, thoughts? 🙁

Athlon XP 2400+ @2Ghz|512MB DDR RAM|QDI 7X/400|PNY Geforce 4 Ti 4200 128MB w/ AGP 8x|Sound Blaster Live

Reply 1 of 5, by analog_programmer

User metadata
Rank Oldbie
Rank
Oldbie

Maybe you've set multiplier of 1.5 which is interpreted by MMX CPUs as 3.5 multi and the FSB is set for the "weird" frequency of 55 MHz (supported by some s.7 chipsets), resulting in 3.5 multi x 55 MHz FSB = 193 MHz CPU.

The word Idiot refers to a person with many ideas, especially stupid and harmful ideas.
This world goes south since everything's run by financiers and economists.
This isn't voice chat, yet some people overusing online communications talk and hear voices.

Reply 2 of 5, by Crinale0

User metadata
Rank Newbie
Rank
Newbie

Yes this does seem to be the case as it does have a 55mhz clock setting. I've tried to set it to 66mhz but it just won't move it's completely stuck so i guess this will have to do.

It now plays Jazz Jackrabbit 2 at 640x480 smoothly which was my aim so this will suffice. If anything i aim to make the machine slower now for DOS games so Jazz2 was the peak of what i needed performance wise

Anything above these requirements are why i have my second windows 98 rig.

Athlon XP 2400+ @2Ghz|512MB DDR RAM|QDI 7X/400|PNY Geforce 4 Ti 4200 128MB w/ AGP 8x|Sound Blaster Live

Reply 3 of 5, by analog_programmer

User metadata
Rank Oldbie
Rank
Oldbie

Just replace the broken DIP switch and set correct CPU frequency by 3 multi x 66 MHz FSB.

The word Idiot refers to a person with many ideas, especially stupid and harmful ideas.
This world goes south since everything's run by financiers and economists.
This isn't voice chat, yet some people overusing online communications talk and hear voices.

Reply 4 of 5, by Chkcpu

User metadata
Rank Oldbie
Rank
Oldbie

Hi Crinaleo,

What happened to this motherboard? It looks like it is caked in some sticky dust. That’s probably why the dip-switches got stuck. Perhaps cleaning the dip-switched with a non-aggressive solvent, like isopropyl alcohol, will get them going.

Anyway, here are some details about the FSB and Multiplier settings on this SY-5VD board.
Looking at the manual from https://theretroweb.com/motherboards/s/soyo-sy-5vd is see that SW1/1 and SW1/2 set the Multiplier and SW1/3, SW1/4, and SW1/5 select the FSB.
Here are the tables:

Multiplier:
x1.5: SW1/1-Off; SW1/2-Off
x2.0: SW1/1-On; SW1/2-Off
x2.5: SW1/1-On; SW1/2-On
x3.0: SW1/1-Off; SW1/2-On
x3.5: SW1/1-Off; SW1/2-Off
Note that the x1.5 and x3.5 settings are the same. You get x1.5 only on the Pentium and K5. All other CPUs, like Pentium MMX and K6 work as x3.5 with this setting.

Bus Clock (FSB):
50MHz: SW1/3-On; SW1/4-On; SW1/5-Off
55MHz: SW1/3-On; SW1/4-Off; SW1/5-On
60MHz: SW1/3-On; SW1/4-Off; SW1/5-Off
66MHz: SW1/3-Off; SW1/4-On; SW1/5-Off
75MHz: SW1/3-Off; SW1/4-On; SW1/5-On

Your Pentium 166 and Pentium 200 MMX both use a 66MHz FSB, so changing SW1/3, SW1/4, and SW1/5 was not needed and you only had to set DIP-switch SW1/1 from On to Off to change the Multiplier from x2.5 to x3 to get the CPU Speed to 200MHz. 😉

The SW1 setting that gave you 100MHz on the Pentium and 233MHz on the MMX is easy to explain with the above tables. You had an 1.5 x 66MHz setting there, that worked as 3.5 x 66 on the MMX.

I agree with analog_programmer that the later 192.5MHz setting you got on the MMX can only be explained by 3.5x 55MHz. However, this low FSB will hurt performance and 3.0 x 66MHz is better. 😀
I hope you can get the DIP-switches going again.

Now about the indications from my CPU Identification utility, the “Time Stamp Counter” text is to inform you it is using the most accurate method to measure the Internal CPU Speed. This program has two ways to measure the CPU Speed and when you don’t see te TSC message, you know that the less accurate instruction timing method was used.
The 3 dashes behind the Clock Multiplier mean that this CPU does not report its Multiplier through software. Therefore the program can’t tell you what it is.
This CHKCPU tool comes with a CHKCPU.TXT file that explains these indications in detail. 😉

Cheers, Jan

CPU Identification utility
The Unofficial K6-2+ / K6-III+ page

Reply 5 of 5, by Crinale0

User metadata
Rank Newbie
Rank
Newbie
analog_programmer wrote on 2024-08-24, 18:22:

Maybe you've set multiplier of 1.5 which is interpreted by MMX CPUs as 3.5 multi and the FSB is set for the "weird" frequency of 55 MHz (supported by some s.7 chipsets), resulting in 3.5 multi x 55 MHz FSB = 193 MHz CPU.

Anyway, here are some details about the FSB and Multiplier settings on this SY-5VD board. Looking at the manual from https://the […]
Show full quote

Anyway, here are some details about the FSB and Multiplier settings on this SY-5VD board.
Looking at the manual from https://theretroweb.com/motherboards/s/soyo-sy-5vd is see that SW1/1 and SW1/2 set the Multiplier and SW1/3, SW1/4, and SW1/5 select the FSB.
Here are the tables:

Multiplier:
x1.5: SW1/1-Off; SW1/2-Off
x2.0: SW1/1-On; SW1/2-Off
x2.5: SW1/1-On; SW1/2-On
x3.0: SW1/1-Off; SW1/2-On
x3.5: SW1/1-Off; SW1/2-Off
Note that the x1.5 and x3.5 settings are the same. You get x1.5 only on the Pentium and K5. All other CPUs, like Pentium MMX and K6 work as x3.5 with this setting.

Bus Clock (FSB):
50MHz: SW1/3-On; SW1/4-On; SW1/5-Off
55MHz: SW1/3-On; SW1/4-Off; SW1/5-On
60MHz: SW1/3-On; SW1/4-Off; SW1/5-Off
66MHz: SW1/3-Off; SW1/4-On; SW1/5-Off
75MHz: SW1/3-Off; SW1/4-On; SW1/5-On

Your Pentium 166 and Pentium 200 MMX both use a 66MHz FSB, so changing SW1/3, SW1/4, and SW1/5 was not needed and you only had to set DIP-switch SW1/1 from On to Off to change the Multiplier from x2.5 to x3 to get the CPU Speed to 200MHz. 😉

The SW1 setting that gave you 100MHz on the Pentium and 233MHz on the MMX is easy to explain with the above tables. You had an 1.5 x 66MHz setting there, that worked as 3.5 x 66 on the MMX.

I agree with analog_programmer that the later 192.5MHz setting you got on the MMX can only be explained by 3.5x 55MHz. However, this low FSB will hurt performance and 3.0 x 66MHz is better. 😀
I hope you can get the DIP-switches going again.

Chkcpu & analogue programmer, THANK YOU!

This got me thinking i wondered what the 55Mhz settings were and turns out it was the inverse of what setting i have made as per the manual.

gfgs.jpg

Turns out, these are printed in error and i tried inverting these settings for the sake of it and...

Screenshot-20240824-220320-Gallery.jpg

Amazing! So yeah, turns out i had managed to set them correctly but the manual is actually incorrect (or hell maybe the printed "ON" location on my drip switch is incorrect) who knows but thankfully now its all setup correctly as it should be.

20240824-213826.jpg

Even all the multiplier info is showing up now and im getting 66Mhz FSB which has actually had a pretty big effect on performance. PlayerBench score went from 18.1 to 20.4 & 143 to 149 on processor benchmark.

If it wasn't for that post i wouldn't of bothered to check - thank you again!

What happened to this motherboard? It looks like it is caked in some sticky dust. That’s probably why the dip-switches got stuck. Perhaps cleaning the dip-switched with a non-aggressive solvent, like isopropyl alcohol, will get them going.

Its an ex warehouse machine so yeah... it's pretty filthy it could use a good soak to be honest but it works and i've cleaned it significantly since i got it haha these machines are so expensive over here and this was cheap so a new PSU later and some cleaning and it's going to be a great DOS machine 😁

Athlon XP 2400+ @2Ghz|512MB DDR RAM|QDI 7X/400|PNY Geforce 4 Ti 4200 128MB w/ AGP 8x|Sound Blaster Live