You cannot change the multiplier on any retail Pentium III CPU. it is locked by intel.

Most Pentium II CPUs allow for a user defined multiplier. But only if it was produced before week 30-something in 1998. Intel locked the multiplier inside the CPUs after this date.

is there any way to get it to 133 fsb without changing the multiplier?
or maybe underclock the cpu?

i do have other pentium 2 and 3 cpus i could use ill see if theres any p2s before 1998
or some 400mhz p3s im thinking 133x3 would work with p3 400.
i think i read one of your old posts that said dos games benefit from 133 fsb.

what would you recommend i do,i have a decent amount of parts to choose from.

i mainly want the rig for 3dfx dos games, and some windows era stuff like resident evil, metal gear and a few others from that generation

ran half life goty for something to play around with until i get dos sound and everything working great just to test something with how the computer sits currently, and it ran alot better than i expected at 1024x768 got caught up playing for atleast an hour 🤣

okay so it finnaly just dawned on me that this cpu isnt going to go to 133fsb with this board. ive never had to mess with clock speeds like this sorry for the lack of understanding.

i dug up some slot 1 cpus for something else to possibly use and i have

p3 667/256/133 SL3XL http://www.cpu-world.com/sspec/SL/SL3XL.html (I'm thinking this one would be the best choice? if the multiplier is locked im betting its at x5 like the p3 500 i am currently using, so all i should have to do is change the sw 1 fsb setting to 133 right?)

P3/450/512/100 SL35D http://www.cpu-world.com/sspec/SL/SL35D.html (in the comments someone says this p3 was manufactured week 26 1999 and is capable of overclock and running at 133mhz)

p2 350mhz SL37F http://www.cpu-world.com/sspec/SL/SL37F.html

p2 300mhz SL2YK http://www.cpu-world.com/sspec/SL/SL2YK.html

i tried the p3 667 at 133 and 100 x 5 and no post, then i tried the p3 450 at 133x3 and its showing up as 600b Mhz computer works, ran half life again it works and everything seems stable.

maybe i got this wrong but if i set the sw 1 to 3-on 1,2,4-off and then set the switch 2 to 2-off 1,3,4-on shouldn't that be 133x3=399mhz cpu with a 133mhz fsb?

Arbys Guy wrote:

i tried the p3 667 at 133 and 100 x 5 and no post, then i tried the p3 450 at 133x3 and its showing up as 600b Mhz computer works, ran half life again it works.

That is a coppermine core. Like I wrote in the other topic: You need to check if your board supports Coppermine Core voltage requirements.

Arbys Guy wrote:

maybe i got this wrong but if i set the sw 1 to 3-on 1,2,4-off and then set the switch 2 to 2-off 1,3,4-on shouldn't that be 133x3=399mhz cpu with a 133mhz fsb?

Switch 2 does absolutely nothing with a retail Pentium III installed.

You seem to have these options:
1) Use an unlocked Pentium II if you have one, and use 133MHz FSB when desired. But net speed will be capped at around 400MHz.
2) Use a later locked Pentium II 450 or Pentium III katmai core up to around 600MHz. But your FSB will be limited to around 100MHz
3) If your board supports coppermine core you can try a native 133MHz FSB processors. But I doubt your board supports it.

We are all at the mercy of what intel supplied us with at the time.

I am not much of an overclocker myself (actually I often tend to downclock), so I may not entirely relate to your goals of this system. But 3D accelerated games are not FSB bound as VESA DOS games are. And for DOS VESA games it is better starting point to first use the MTRRLFBE tool, then trying to stress the FSB more.

gerwin wrote:

Arbys Guy wrote:

i tried the p3 667 at 133 and 100 x 5 and no post, then i tried the p3 450 at 133x3 and its showing up as 600b Mhz computer works, ran half life again it works.

That is a coppermine core. Like I wrote in the other topic: You need to check if your board supports Coppermine Core voltage requirements

Arbys Guy wrote:

maybe i got this wrong but if i set the sw 1 to 3-on 1,2,4-off and then set the switch 2 to 2-off 1,3,4-on shouldn't that be 133x3=399mhz cpu with a 133mhz fsb?

Switch 2 does absolutely nothing with a retail Pentium III installed.

So after some random forum searching i believe it does https://www.wimsbios.com/forum/cpu-isn-recogn … sion-t1704.html but the board still wont boot with that coppermine pIII 667 in it, so maybe he is wrong.

okay so the multiplier is locked at 4.5 , thus equaling 600mhz (598) cpu when when sw#1 is set to 133mhz fsb.. i get it now.

p2 350mhz SL37F came out in april of 1998 so week 20 or earlier

p2 300MHz SL2YK came out January 26th 1998 just before week 4

ill prob end up giving these cpus a try, im curious why you under clock though?, is it to prolong life or do certain dos games benefit from a lower clocked cpu.

and on a side note, were you able to get dma mode to work with your rev 2.0 board, under w9x.(Scratch that.. dma mode was enabled.. i could of swore it wasn't.)

ps watching phils computer lab video about MTRRLFBE tool and fastvid thanks for the tip (=

Arbys Guy wrote:

So after some random forum searching i believe it does https://www.wimsbios.com/forum/cpu-isn-recogn … sion-t1704.html but the board still wont boot with that coppermine pIII 667 in it, so maybe he is wrong.

Near Switch Block SW1 there is a VRM chip. When this chip is of the type "HIP6004BCB" it can support coppermine core voltages. When instead it is something like "HIP6004ACB", it cannot.

Arbys Guy wrote:

ill prob end up giving these cpus a try, im curious why you under clock though?, is it to prolong life or do certain dos games benefit from a lower clocked cpu.

Different reasons.
- Many DOS games are speed sensitive in some way, yes.
- I like cool running systems instead of hot and stressed systems. Just a personal preference.
- It allows you to cover more bases with a single flexible system. You can take a slower system like a 486, Pentium MMX or AMD K6 and try to squeeze out more speed, or you can go the opposite direction: Take a fast i440BX based motherboard and see how it can meet the speeds of the slower systems. This is also just interesting on itself, as an experiment, to find what is possible and what is not.

Arbys Guy wrote:

ps watching phils computer lab video about MTRRLFBE tool and fastvid thanks for the tip (=

You're welcome. I did not know he made a video about it.