VOGONS


First post, by jasa1063

User metadata
Rank Oldbie
Rank
Oldbie

I just put together an AMD K5 system using an Amptron PM-9800 motherboard with a 430TX chipset and 64MB of EDO memory. I read that most AMD K5 PR166 will only overclock to 3-4MHz extra. I decided to set the motherboard multiplier to 3x and just see if it would even boot up. To my amazement, the system has been totally stable with the voltage set to 3.52v. I have run then entire suite from Phils Computer Lab DOSBench package, memtest86 v4.3 and Windows 95 OSR2.5. I have not had any system crashes or freeze ups at all. My question at this point is should I keep the system running at 133MHz or put it back to 116.5MHz on the basis I could damage the CPU running it a the higher clock speed for an extended period of time.

Reply 1 of 16, by Horun

User metadata
Rank l33t++
Rank
l33t++

Yes you could damage it over time if it's internally running hot. It would be interesting to know what the DIE cap temperature is, if not above about 30C at rest and 50C under super heavy load then it might work at that speed for years without issue.
Maybe you have a very late PR166 when AMD was trying to come out with the PR200 so it can work well at that speed. There is also the issue of electron degradation or whatever they call it where doped silicon looses it ability to function as intended, that is typically 30-50 years but less when heat and speed stressed AFAIK....

Hate posting a reply and then have to edit it because it made no sense 😁 First computer was an IBM 3270 workstation with CGA monitor. https://archive.org/details/@horun

Reply 2 of 16, by jasa1063

User metadata
Rank Oldbie
Rank
Oldbie
Horun wrote on 2021-07-04, 00:41:

Yes you could damage it over time if it's internally running hot. It would be interesting to know what the DIE cap temperature is, if not above about 30C at rest and 50C under super heavy load then it might work at that speed for years without issue.
Maybe you have a very late PR166 when AMD was trying to come out with the PR200 so it can work well at that speed. There is also the issue of electron degradation or whatever they call it where doped silicon looses it ability to function as intended, that is typically 30-50 years but less when heat and speed stressed AFAIK....

The CPU does not even get that warm to the touch under load, so I am going to assume it should be Ok. I also thought this may be a late PR166 that would be a candidate for a PR200. Either way I am just really happy that it is stable at 133MHz.

Reply 3 of 16, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

That's pretty awesome! What's the stock voltage of the pr166?

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 4 of 16, by Garrett W

User metadata
Rank Oldbie
Rank
Oldbie

Hey that's pretty cool!

I would honestly not really care to much about potential danger to the CPU, it's not like you'll be running it as a daily driver anyway and the voltage bump is minimal anyway.

Reply 5 of 16, by Imperious

User metadata
Rank Oldbie
Rank
Oldbie

I had no problems getting a K5 Pr133 running at 125mhz (83.3x1.5) on my Epox Super 7 motherboard.
I have yet to try on another board as the max voltage the Epox could manage was 3.3v, so on another board that can
provide 3.52v 133mhz is possible. It wouldn't do 133mhz on the Epox board.

Atari 2600, TI994a, Vic20, c64, ZX Spectrum 128, Amstrad CPC464, Atari 65XE, Commodore Plus/4, Amiga 500
PC's from XT 8088, 486, Pentium MMX, K6, Athlon, P3, P4, 775, to current Ryzen 5600x.

Reply 7 of 16, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie
jasa1063 wrote on 2021-07-04, 13:17:
H3nrik V! wrote on 2021-07-04, 07:47:

That's pretty awesome! What's the stock voltage of the pr166?

The stock voltage is 3.52v

So you're running 133 at stock voltage? I wouldn't be too worried, then .. After all, a PR166 is not that rare .. And I would think the risk is low. Electromigration is mostly an effect of higher voltage rather than higher clock AFAIK.

Please use the "quote" option if asking questions to what I write - it will really up the chances of me noticing 😀

Reply 8 of 16, by jasa1063

User metadata
Rank Oldbie
Rank
Oldbie
H3nrik V! wrote on 2021-07-04, 19:08:
jasa1063 wrote on 2021-07-04, 13:17:
H3nrik V! wrote on 2021-07-04, 07:47:

That's pretty awesome! What's the stock voltage of the pr166?

The stock voltage is 3.52v

So you're running 133 at stock voltage? I wouldn't be too worried, then .. After all, a PR166 is not that rare .. And I would think the risk is low. Electromigration is mostly an effect of higher voltage rather than higher clock AFAIK.

Thanks for info. I feel alot better running @133MHz now:)

Reply 9 of 16, by NostalgicAslinger

User metadata
Rank Member
Rank
Member

I also have an AMD K5 PR166 which runs completely stable with PR200 speed, Prime stable with 133MHz and also with a reduced VCore 3.52>3.30V! I have used this CPU for my 3dfx Voodoo Rush Video Project in 2019.

Maybe this CPU could also run with 75MHz x2=150MHz and the stock 3.52V Voltage? Never tried it out, because this is one of my creme de la creme S7 CPUs. My other K5, a PR150 does not like any overclocking, PR166 clocks (116.6MHz) gave a black screen. Only 75MHz FSB x 1.5 was stable under DOS. Maybe the stepping is also important? The PR150 has the B and the PR166 the C Stepping.

Here, a picture of the PR166 with the used mainboard, a Gigabyte GA-586T2 Rev. 1.2 with the Intel 430TX (latest 2.2 Bios) and a AIDA CPUID screenshot with 133 MHz and 3,30V VCore.

Attachments

  • DSC01522.JPG
    Filename
    DSC01522.JPG
    File size
    1.53 MiB
    Views
    811 views
    File license
    Fair use/fair dealing exception
  • AMD K5.png
    Filename
    AMD K5.png
    File size
    376.73 KiB
    Views
    811 views
    File license
    Fair use/fair dealing exception
Last edited by NostalgicAslinger on 2021-07-30, 11:23. Edited 2 times in total.

Reply 10 of 16, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie

My m571 allowed an undocumented higher than 3.5 volt setting (3.8volts)
This allowed me to get a pr133 up to an auctual 133mhz (Pr200)

If I had to guess a similar board could probably break the 133mhz barrier

Reply 11 of 16, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

3.6V wasn't unheard of either as a cpu voltage setting. Seen it on a few. Might be worth looking at manuals of other boards in the series and if their volt reg jumper layout is similar and they have 3.6 then try that setting on the board that "doesn't support" it.

I don't think I ever went over 3.52 "in the day" to get my Cyrixes to run at PR rating. Well most of my Cyrixes, the MII-366 didn't wanna... though that was a kinda high bar, because by then they were setting PR at 40-50% over clock rather than 25%.... but I didn't manage 25% either, 290 was the best it did from 250 stock.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 12 of 16, by NostalgicAslinger

User metadata
Rank Member
Rank
Member
rmay635703 wrote on 2021-07-30, 00:02:

My m571 allowed an undocumented higher than 3.5 volt setting (3.8volts)
This allowed me to get a pr133 up to an auctual 133mhz (Pr200)

If I had to guess a similar board could probably break the 133mhz barrier

Wow, because also the VIO voltage will go up to 3,8V. So RAM, cache modules and chipset will be also fed with this voltage. Would be a bit too risky for me.

Here is a 150Mhz record with an PR166: https://hwbot.org/submission/929856_orange_cp … r166_150.46_mhz
Don't know how much voltage, maybe 3,7V?...but Watercooled!

Reply 13 of 16, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
NostalgicAslinger wrote on 2021-07-30, 10:59:
Wow, because also the VIO voltage will go up to 3,8V. So RAM, cache modules and chipset will be also fed with this voltage. Woul […]
Show full quote
rmay635703 wrote on 2021-07-30, 00:02:

My m571 allowed an undocumented higher than 3.5 volt setting (3.8volts)
This allowed me to get a pr133 up to an auctual 133mhz (Pr200)

If I had to guess a similar board could probably break the 133mhz barrier

Wow, because also the VIO voltage will go up to 3,8V. So RAM, cache modules and chipset will be also fed with this voltage. Would be a bit too risky for me.

Here is a 150Mhz record with an PR166: https://hwbot.org/submission/929856_orange_cp … r166_150.46_mhz
Don't know how much voltage, maybe 3,7V?...but Watercooled!

Don’t use a m590 then VIO is fixed up by the limit though it never caused any issues for me

Re: Can this Pentium 100 motherboard accept Pentium MMX CPU?

A95BBADC-7B5E-402F-AC6F-3BD52A0A2E40.jpeg
Filename
A95BBADC-7B5E-402F-AC6F-3BD52A0A2E40.jpeg
File size
174.44 KiB
Views
756 views
File license
Public domain
Last edited by Stiletto on 2021-07-30, 19:42. Edited 2 times in total.

Reply 14 of 16, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Don't screenshot, it's lame, link the post.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.

Reply 15 of 16, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
BitWrangler wrote on 2021-07-30, 14:53:

Don't screenshot, it's lame, link the post.

Then you have to go read said post and when you don’t understand
What happun
We go over the same subject again

Reply 16 of 16, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Thanks for putting the link in, IMO understanding is more likely when the post is in context.

Unicorn herding operations are proceeding, but all the totes of hens teeth and barrels of rocking horse poop give them plenty of hiding spots.