• Please review our updated Terms and Rules here

what kind of memory does a P75 require?

Dave Farquhar

Experienced Member
Joined
May 23, 2010
Messages
461
Location
the midwest
I ran 60ns FPM on my Pentium-75 with the original Triton chipset, but I do think 70ns FPM from my 486 worked in it. Seems to me when I first got my Pentium board, I swiped some of my 486's memory for it for a couple of weeks until I could afford some new memory.

My board had 72-pin SIMMs. You had to use them in matched pairs.
 

Dave Farquhar

Experienced Member
Joined
May 23, 2010
Messages
461
Location
the midwest
I'm not sure if you're asking if same memory will work in a P100, or if you can run a P100 at 75 MHz. But the answer is yes on both counts. Intel didn't lock FSBs and multipliers in those days.
 

tipc

Banned
Joined
Jan 16, 2005
Messages
2,760
Location
Principality of Xeon W-2140B the Great State of Ce
ok, good in any event.

So therefore a 100mhz chip in a 75mhz box/mobo will run @ 75mhz? I always thought that was a function of the crystal in olden days. I thought the mechanism that clocked the chip was onboard (but what do I know). At least it seems that way w/later P3s and whatnot. But again WDIK...
 

Neon_WA

Experienced Member
Joined
Jan 15, 2010
Messages
270
Location
Perth, Western Australia
the P100 had a minimum bus speed of 25MHz or maximum bus speed of 66MHz
so you could run it easily at any speed from 37.5 to 100MHz

newer Pentiums.. mainly MMX were multiplier limited. Locked multipliers didnt really start until part way thru the P2 range
 

Jorg

Veteran Member
Joined
Aug 31, 2003
Messages
1,323
Location
Switzerland
Just for the record then.. ofcourse you would have to set the jumpers for bus speed and multiplier.
But then the cpu will run on that, wether it's specced for that or not. Then the question is for how long, if you overclock :)

If you go look for 72 pin SIMMS, be aware that many of them might be EDO, not FPM. Depending on if your board supports EDO, it might run or not.
 

Dave Farquhar

Experienced Member
Joined
May 23, 2010
Messages
461
Location
the midwest
Yes, on those early Pentiums, you just set your multiplier and bus speed, and the chip would try to run at that speed. It was very common for people to buy Pentium 75s and then see how fast they would run. In the later days, a lot of people reported getting P75s to run as P133s without issue. I tried overclocking P75s twice and both times they acted goofy, even if I only went to 90 MHz. But a lot of P75s, especially the early ones, were just P90s that couldn't run reliably at 90 MHz.

It was common practice back then to mess with bus speeds and multipliers. Once boards with a 75 MHz bus arrived (to support Cyrix chips that ran at 75 MHz x 2), people started trying to run Pentiums at 75 MHz x 2 and found they ran faster than P166s did. Then people discovered an undocumented 83 MHz bus on a couple of boards, and found a P166 running at 83 x 2 was faster than a regular P200. And a Pentium running at 208 MHz, 83 x 2.5, was the bomb.

The trouble was that when you did this with an Intel chipset, you ended up with the PCI bus running at odd speeds, so the systems weren't always stable. It was easier to get away with a 37.5 MHz PCI bus than a 41.5 MHz one. And that was on top of overclocking.

That had a lot to do with why Intel started locking the multipliers with the P2 generation. People would buy P120s and P133s and then clock them as high as they could get them to run. I remember going to Tom's Hardware the first time, and right there on the front page was a series of articles telling how to overclock Pentiums on Asus motherboards. And most of the ads on the site were advertising bundle pricing on low-end Intel CPUs and those same Asus motherboards. That was how most of the long-running hardware enthusiast sites got started.
 
Top