VOGONS


ALR Evolution V ST (Socket 5) build

Topic actions

Reply 60 of 69, by JamesF36

User metadata
Rank Newbie
Rank
Newbie
Sphere478 wrote on 2022-06-27, 01:39:
Can you clarify your readings? 0 means short. Did you mean to say that or do you mean open? […]
Show full quote

Can you clarify your readings? 0 means short. Did you mean to say that or do you mean open?

Also 1.553 what? 1.553 ohms? K ohms, m ohms?

Without knowing the range of the readings it doesn’t tell anything.

If I make some assumptions about what I think you “meant” it sounds like you may be missing both pull ups and not have a jumper at all for bf1.

But I can’t really make that assumption without clarification on your readings.

You could just order a tweaker and some smd resistors/switch. If your soldering skills are up to it.

You will then have switches for all of this and figuring it out will be easy.

Sorry forgot to mention it's K ohms, that's after I added the 1K resistor though if that makes a difference, and yeah i'm not sure I believe it just measured 0, it's so interesting how this board is set up though, the fact that there is 8 jumper pins to change CPU speed but of course this being a OEM board it's just silkscreened on the board what speed it changes to rather than the multiplier..

Maybe I am missing both pull ups, in that case assuming that what should I do? I might order a tweaker eventually but I wouldn't mind experimenting, I think I remember one post was talking about soldering a wire from BF1 to ground not sure if that applies to my scenario though

Reply 61 of 69, by Sphere478

User metadata
Rank l33t
Rank
l33t

I stuck a mmx and a non mmx on a volt meter.

it's interesting they say internal pull up/down but the readings I get out of them are in the high k ohms (like hundreds) or low megaohms (so basically almost no connection).

when they say "internal pull up/down" I wonder if they actually just mean the logic defaults to it. because there isn't much there in reality.

your 200 may support 3.5x btw, but you won't know until you try it. Most don't, but I have a few that do. (they weren't all locked)

you know, looking at the pic of your board really close, I'm not even convinced it has ANY BFx jumpers.

yeah, you probably need a Tweaker man.

obviously you also need a voltage interposer. but it's still beta until I get the prototype in the mail and confirm that it works. ( am expecting it to) you could snag say a tweaker shim in the meantime. those are pretty painless, but you need good soldering skills. and you have to perminantly solder it to the processor unfortunately. The alternatives go on the back of the mobo.

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 62 of 69, by JamesF36

User metadata
Rank Newbie
Rank
Newbie
Sphere478 wrote on 2022-06-27, 07:26:
I stuck a mmx and a non mmx on a volt meter. […]
Show full quote

I stuck a mmx and a non mmx on a volt meter.

it's interesting they say internal pull up/down but the readings I get out of them are in the high k ohms (like hundreds) or low megaohms (so basically almost no connection).

when they say "internal pull up/down" I wonder if they actually just mean the logic defaults to it. because there isn't much there in reality.

your 200 may support 3.5x btw, but you won't know until you try it. Most don't, but I have a few that do. (they weren't all locked)

you know, looking at the pic of your board really close, I'm not even convinced it has ANY BFx jumpers.

yeah, you probably need a Tweaker man.

obviously you also need a voltage interposer. but it's still beta until I get the prototype in the mail and confirm that it works. ( am expecting it to) you could snag say a tweaker shim in the meantime. those are pretty painless, but you need good soldering skills. and you have to perminantly solder it to the processor unfortunately. The alternatives go on the back of the mobo.

I did it - I finally did it! It's running at 200mhz, I'm gonna get the tweaker just for the sake of having it clean but having the resistor between BF0 and VCC3 and I soldered a small wire between VSS and BF1 which finally got it to run at 200mhz, I'm happy.. next is trying to get a K6-233 to run which should be the same concept right?

Attachments

  • IMG_5142.jpg
    Filename
    IMG_5142.jpg
    File size
    1.08 MiB
    Views
    155 views
    File license
    Public domain

Reply 63 of 69, by Sphere478

User metadata
Rank l33t
Rank
l33t
JamesF36 wrote on 2022-06-27, 15:11:
Sphere478 wrote on 2022-06-27, 07:26:
I stuck a mmx and a non mmx on a volt meter. […]
Show full quote

I stuck a mmx and a non mmx on a volt meter.

it's interesting they say internal pull up/down but the readings I get out of them are in the high k ohms (like hundreds) or low megaohms (so basically almost no connection).

when they say "internal pull up/down" I wonder if they actually just mean the logic defaults to it. because there isn't much there in reality.

your 200 may support 3.5x btw, but you won't know until you try it. Most don't, but I have a few that do. (they weren't all locked)

you know, looking at the pic of your board really close, I'm not even convinced it has ANY BFx jumpers.

yeah, you probably need a Tweaker man.

obviously you also need a voltage interposer. but it's still beta until I get the prototype in the mail and confirm that it works. ( am expecting it to) you could snag say a tweaker shim in the meantime. those are pretty painless, but you need good soldering skills. and you have to perminantly solder it to the processor unfortunately. The alternatives go on the back of the mobo.

I did it - I finally did it! It's running at 200mhz, I'm gonna get the tweaker just for the sake of having it clean but having the resistor between BF0 and VCC3 and I soldered a small wire between VSS and BF1 which finally got it to run at 200mhz, I'm happy.. next is trying to get a K6-233 to run which should be the same concept right?

233 should be the 1.5x setting. Google the datasheet for the k6.

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 65 of 69, by Sphere478

User metadata
Rank l33t
Rank
l33t
evasive wrote on 2022-06-29, 12:20:

UR page:
https://www.ultimateretro.net/en/motherboards/11986

Still looking for a manual

Awesome!

It would be cool to drop links to threads like this in UR pages

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 66 of 69, by Riikcakirds

User metadata
Rank Member
Rank
Member
JamesF36 wrote on 2022-06-27, 15:11:

I did it - I finally did it! It's running at 200mhz, I'm gonna get the tweaker just for the sake of having it clean but having the resistor between BF0 and VCC3 and I soldered a small wire between VSS and BF1 which finally got it to run at 200mhz, I'm happy.. next is trying to get a K6-233 to run which should be the same concept right?

I skipped the soldering and used a thin thread from an ATA cable and wrapped it around bf1 and the closest VSS ground pin. This was to get a P200 (non mmx) working on an Intel Zappa socket 5 board. Without that it would only go up to 133mhz.
Didn't need any other mod or to add a resistor.

Reply 67 of 69, by Sphere478

User metadata
Rank l33t
Rank
l33t
Riikcakirds wrote on 2022-06-29, 15:11:
JamesF36 wrote on 2022-06-27, 15:11:

I did it - I finally did it! It's running at 200mhz, I'm gonna get the tweaker just for the sake of having it clean but having the resistor between BF0 and VCC3 and I soldered a small wire between VSS and BF1 which finally got it to run at 200mhz, I'm happy.. next is trying to get a K6-233 to run which should be the same concept right?

I skipped the soldering and used a thin thread from an ATA cable and wrapped it around bf1 and the closest VSS ground pin. This was to get a P200 (non mmx) working on an Intel Zappa socket 5 board. Without that it would only go up to 133mhz.
Didn't need any other mod or to add a resistor.

I think a better way to do this is to solder the resistors to the top of the cpu. It’s kinda bad practice to pull these with a 0 ohm link.

For pull down a 25-100 ohm is a pretty good bet.

I did this in the tillamook thread

Re: Tillamook 266MHz and working L2 cache?

Of course a different configuration is required on p55c/p54c

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 68 of 69, by Riikcakirds

User metadata
Rank Member
Rank
Member
Sphere478 wrote on 2022-06-29, 15:18:
I think a better way to do this is to solder the resistors to the top of the cpu. It’s kinda bad practice to pull these with a 0 […]
Show full quote
Riikcakirds wrote on 2022-06-29, 15:11:
JamesF36 wrote on 2022-06-27, 15:11:

I did it - I finally did it! It's running at 200mhz, I'm gonna get the tweaker just for the sake of having it clean but having the resistor between BF0 and VCC3 and I soldered a small wire between VSS and BF1 which finally got it to run at 200mhz, I'm happy.. next is trying to get a K6-233 to run which should be the same concept right?

I skipped the soldering and used a thin thread from an ATA cable and wrapped it around bf1 and the closest VSS ground pin. This was to get a P200 (non mmx) working on an Intel Zappa socket 5 board. Without that it would only go up to 133mhz.
Didn't need any other mod or to add a resistor.

I think a better way to do this is to solder the resistors to the top of the cpu. It’s kinda bad practice to pull these with a 0 ohm link.

For pull down a 25-100 ohm is a pretty good bet.

I did this in the tillamook thread

Re: Tillamook 266MHz and working L2 cache?

Of course a different configuration is required on p55c/p54c

I did it back in school during 1997, so didn't have a soldering iron etc and was reading on Usenet how people were getting a classic P200 to work on socket 5 boards. The board I used is very, very early socket 5, Asus P54TP4, bought in Feb'1995. Some of the chip dates on this board are week December 1994, very early for a 430fx. It still to this day runs with the P200.
I remember reading on Usenet those years ago the internal pull up resistor on pin BF1 will protect it from floating or any short after connecting it to ground. It's worked for 20+ years but again I'm not certain how correct that info is. I guess it wouldn't be possible to overclock it to a 233 using 3.5x because of the missing BF2.

Reply 69 of 69, by Sphere478

User metadata
Rank l33t
Rank
l33t
Riikcakirds wrote on 2022-07-01, 11:02:
Sphere478 wrote on 2022-06-29, 15:18:
I think a better way to do this is to solder the resistors to the top of the cpu. It’s kinda bad practice to pull these with a 0 […]
Show full quote
Riikcakirds wrote on 2022-06-29, 15:11:

I skipped the soldering and used a thin thread from an ATA cable and wrapped it around bf1 and the closest VSS ground pin. This was to get a P200 (non mmx) working on an Intel Zappa socket 5 board. Without that it would only go up to 133mhz.
Didn't need any other mod or to add a resistor.

I think a better way to do this is to solder the resistors to the top of the cpu. It’s kinda bad practice to pull these with a 0 ohm link.

For pull down a 25-100 ohm is a pretty good bet.

I did this in the tillamook thread

Re: Tillamook 266MHz and working L2 cache?

Of course a different configuration is required on p55c/p54c

I did it back in school during 1997, so didn't have a soldering iron etc and was reading on Usenet how people were getting a classic P200 to work on socket 5 boards. The board I used is very, very early socket 5, Asus P54TP4, bought in Feb'1995. Some of the chip dates on this board are week December 1994, very early for a 430fx. It still to this day runs with the P200.
I remember reading on Usenet those years ago the internal pull up resistor on pin BF1 will protect it from floating or any short after connecting it to ground. It's worked for 20+ years but again I'm not certain how correct that info is. I guess it wouldn't be possible to overclock it to a 233 using 3.5x because of the missing BF2.

I don’t believe that the 3.5x multi exists on the p54c core. Only on p55 and tillamook. Sometimes it’s locked on them though. (And obviously other mfgs have it also)

Don’t misread me, a direct short seems to work in instances that I have seen of people trying it, doing it for pull up is more dangerous though. (They both are) I’m saying that it’s just not how you are supposed to do it in electrical design. I can think of one instance in the tillamook thread where not using pull up/down resistors actually resulted in a blown resistor on the (my) board. I shouldn’t have listened to them haha. I know better now.

I recently did some tests on a p54 (75) and a p55 (233)

I actually couldn’t find any pull ups or downs in the cpu. Only very high ohm connections. I think when they say internal pull up or down what they really mean is that the logic defaults to said setting when left floating.

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)