VOGONS


First post, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Drop first hand data on what multi’s work on your pentium cpus.

!!!Include s-spec number!!!

Example: my pentium mmx 200 black plastic chip can do 2.5 and 3x (sl27j)

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 1 of 28, by Tetrium

User metadata
Rank l33t++
Rank
l33t++

Sorry for not providing the data you requested, but I would like to advice you to include production dates.

I at one time tested a fairly large batch of MMX CPUs and later lost my notes with the results ( -_- ) but I do remember that 200MHz and 166MHz would have a bigger chance at having at least some of its multipliers disabled (the CPU would ignore 0 or 1 on the motherboard) with chips with a more recent production date having a higher chance at having disabled multipliers. There was some cut-off date after which CPUs started getting 'locked' and another date after which unlocked chips would be something of a rarity if not non-existent.

So please do include the date code 😀

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 3 of 28, by clueless1

User metadata
Rank l33t
Rank
l33t

I have a P120 that I believe is multiplier locked. My POD200MMX seems to be completely unlocked. I've run it anywhere from 75Mhz to 250Mhz. I've heard that if you disable the integrated fan, it will run at 60Mhz?

The more I learn, the more I realize how much I don't know.
OPL3 FM vs. Roland MT-32 vs. General MIDI DOS Game Comparison
Let's benchmark our systems with cache disabled
DOS PCI Graphics Card Benchmarks

Reply 4 of 28, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie
clueless1 wrote on 2022-01-02, 13:30:

I have a P120 that I believe is multiplier locked. My POD200MMX seems to be completely unlocked. I've run it anywhere from 75Mhz to 250Mhz. I've heard that if you disable the integrated fan, it will run at 60Mhz?

Early and late p120/p133 appear multiplier locked, in some cases because they are missing one of the multiplier pins

The p133mmx oddity however would be interesting to observe.

Reply 5 of 28, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

Need production dates and stepping for context, for example early P75, P90, P100, 1.5 was the multiplier period, 2x didn't exist, and when it did for the first 120 and 133, (Gold cap maybe) it was a weird abuse of another pin to signal it. I do not know if cores produced at that time and demoted to P100 or below retained that ability. Then by the late B steppings and CC0 stepping, they were using the 1.5 and 2x and the BF pins we know and love, and all CPU from P75 through P133 seemed to have both 1.5 and 2x implemented... meaning you can do ridiculous things with a (lucky) CC0 P75, like run it at 2x75 with a big heatsink on. Some of the less lucky ones will still do 120 or 133. Then we had P150 and P166 with the first implementations of 2.5 multiplier, and initially, I think that's all they did. I don't think 3x were seen on those until they were producing P200.

Be aware that when looking up the intel datasheets, they don't do history, it's a snapshot, this is what we're shipping now, current product, so you will see significant differences between the oft cited 1997 one, when P54C was "final form" vs the initial ones in 94/95 .. We could use finding some 96 ones to fill the gaps.

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 6 of 28, by rmay635703

User metadata
Rank Oldbie
Rank
Oldbie

Something worth noting is the p75 didn’t end up a “volume “ product until after the p90 was basically discontinued (p100 was the mainstream offering)
and the p120 was already released.

Because of this it is actually somewhat rare to find one that doesn’t have 2x, most of mine so far have had 2x but as always YMMV

Reply 8 of 28, by BitWrangler

User metadata
Rank l33t++
Rank
l33t++

The 1.5 is remapped to 3.5

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 9 of 28, by Riikcakirds

User metadata
Rank Member
Rank
Member
BitWrangler wrote on 2022-01-02, 16:39:

Need production dates and stepping for context, for example early P75, P90, P100, 1.5 was the multiplier period, 2x didn't exist, and when it did for the first 120 and 133, (Gold cap maybe) it was a weird abuse of another pin to signal it. I do not know if cores produced at that time and demoted to P100 or below retained that ability. Then by the late B steppings and CC0 stepping, they were using the 1.5 and 2x and the BF pins we know and love, and all CPU from P75 through P133 seemed to have both 1.5 and 2x implemented... meaning you can do ridiculous things with a (lucky) CC0 P75, like run it at 2x75 with a big heatsink on. Some of the less lucky ones will still do 120 or 133. Then we had P150 and P166 with the first implementations of 2.5 multiplier, and initially, I think that's all they did. I don't think 3x were seen on those until they were producing P200.

Be aware that when looking up the intel datasheets, they don't do history, it's a snapshot, this is what we're shipping now, current product, so you will see significant differences between the oft cited 1997 one, when P54C was "final form" vs the initial ones in 94/95 .. We could use finding some 96 ones to fill the gaps.

I remember reading in February 1995 that P75 could already be overclocked to 133mhz with x2 multi (posts on usenet confirm this, some posts in January 1995). It came out in December 1994 but P75's were nearly always better overclockers than P90 and p100s. Even though these P75 processors didn't have 2x multi, the first 430fx triton motherboards could override and still set x2 and get 133mhz (months before this chip came out). I haven't used CCO stepping but the E0 stepping could do 150mhz no problem just using 1 inch tall passive heatsink.
P 150 were good sweet spot, you could even in socket 5 boards using BF pins overclock this x3 to 200(and also under clock the entire range from P75.90,100,120....P200.

Reply 10 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
AlessandroB wrote on 2022-01-02, 17:59:

I Have (if i remember correctly) one P233MMX that accept every multiplier, can go from 75 to 290Mhz.

4x?

Can you verify? Post pic of the back?

Something I’d like to figure out is
A:if there were any non tillamooks that ever got the 4x multiplier even if you had to do a pin mod like on the tillamook.

B:if said cpu ever had smp pins because tillamook doesn’t

C:what was the earliest cpu to support 3.5x

I have several dual boards and it’s so sad to be stuck at 233mhz becauee the boards do not overclock at all

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 11 of 28, by Tetrium

User metadata
Rank l33t++
Rank
l33t++
AlessandroB wrote on 2022-01-02, 17:59:

I Have (if i remember correctly) one P233MMX that accept every multiplier, can go from 75 to 290Mhz.

So which multipliers did this chip accept?

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 12 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
Tetrium wrote on 2022-01-03, 09:11:
AlessandroB wrote on 2022-01-02, 17:59:

I Have (if i remember correctly) one P233MMX that accept every multiplier, can go from 75 to 290Mhz.

So which multipliers did this chip accept?

I believe my 233s do this: (will update with s-spec number when I can)

1.5x=3.5x
2x
2.5x
3x

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 13 of 28, by Tetrium

User metadata
Rank l33t++
Rank
l33t++
Sphere478 wrote on 2022-01-03, 09:45:
I believe my 233s do this: (will update with s-spec number when I can) […]
Show full quote
Tetrium wrote on 2022-01-03, 09:11:
AlessandroB wrote on 2022-01-02, 17:59:

I Have (if i remember correctly) one P233MMX that accept every multiplier, can go from 75 to 290Mhz.

So which multipliers did this chip accept?

I believe my 233s do this: (will update with s-spec number when I can)

1.5x=3.5x
2x
2.5x
3x

All of mine behaved like this.

Whats missing in your collections?
My retro rigs (old topic)
Interesting Vogons threads (links to Vogonswiki)
Report spammers here!

Reply 14 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Noticed something interesting when testing cpus.

I have board set to 2x100mhz

But one chip posts at 250mhz but the other posts at 200mhz

But have the same spec number…. (The plot thickens…)

It’s the one with the rounded edges that posts at 200

Anyone able to make sense of this?

Edit: tried another test: 66mhz doesn’t work on this board so I tried setting 83.3x3.5

One chip did 250 and the other did 290mhz!! (The one with rounded edges) so it appears that this processor, despite being a 200 actually has the 3.5x multiplier intact! Where as the other one does not.

Attachments

  • image.jpg
    Filename
    image.jpg
    File size
    1.85 MiB
    Views
    709 views
    File license
    Public domain

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 15 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Okay another test:

Lowered the the board down to 75 and ran all the processors I have of this s-spec through the board.

All but one posts at 266 now

The outlier is this one by it’s self it only posts at 225mhz

Can’t confirm better than that that there is definitely some kind of multiplier manipulation going on here. What’s odd is that 2x became 2.5x on this chip…🤔

Attachments

  • image.jpg
    Filename
    image.jpg
    File size
    1.97 MiB
    Views
    700 views
    File license
    Public domain

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 16 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++

Another test 3.5x 75

with this new s-spec number results are

One 266
Rest are 190

so inverse, most are locked!

As per the last one the one by it’s self is the outlier but this time the outlier is a unlocked one whereas before the outlier was locked!

Is anyone seeing a pattern here? Because I am not… odd that this time 3.5x equals 2.5x instead of 3x on the locked ones this time around.

Attachments

  • image.jpg
    Filename
    image.jpg
    File size
    1.57 MiB
    Views
    688 views
    File license
    Public domain

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 17 of 28, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie

Very weird, cause AFAIR, it was 1.5 that was remapped to 3.5, no other remaps.

On the image with 2 200's, I notice that the "round edged" one is manufactured wk 15 of '99, the "sharp edged" is wk 51 of '97. The difference in manufacturing time might be the difference in locked vs unlocked. IIRC, the Pentium 2s started being locked sometime in '98 ...

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

Reply 18 of 28, by Sphere478

User metadata
Rank l33t++
Rank
l33t++
H3nrik V! wrote on 2022-01-18, 08:58:

Very weird, cause AFAIR, it was 1.5 that was remapped to 3.5, no other remaps.

On the image with 2 200's, I notice that the "round edged" one is manufactured wk 15 of '99, the "sharp edged" is wk 51 of '97. The difference in manufacturing time might be the difference in locked vs unlocked. IIRC, the Pentium 2s started being locked sometime in '98 ...

How do I read the mfg dates?

Of course the board is set to 1.5/3.5 but on one s-spec that is locked I get 3x, on the other I get 2.5x 🤔

Last edited by Sphere478 on 2022-01-18, 09:11. Edited 1 time in total.

Sphere's PCB projects.
-
Sphere’s socket 5/7 cpu collection.
-
SUCCESSFUL K6-2+ to K6-3+ Full Cache Enable Mod
-
Tyan S1564S to S1564D single to dual processor conversion (also s1563 and s1562)

Reply 19 of 28, by H3nrik V!

User metadata
Rank Oldbie
Rank
Oldbie
Sphere478 wrote on 2022-01-18, 09:03:
H3nrik V! wrote on 2022-01-18, 08:58:

Very weird, cause AFAIR, it was 1.5 that was remapped to 3.5, no other remaps.

On the image with 2 200's, I notice that the "round edged" one is manufactured wk 15 of '99, the "sharp edged" is wk 51 of '97. The difference in manufacturing time might be the difference in locked vs unlocked. IIRC, the Pentium 2s started being locked sometime in '98 ...

How do I read the mfg dates?

L9150229-1858: L is manufacturing plant IIRC, "9" is last digit of mfg. year, i.e. 99 (or 2009 😉 ) "15" is week of manufacturing (or assembly)

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