I got this motherboard working (missing inductor taken from dead mobo) and I am confused what update file should I get? I only know I have Award BIOS from 1997 and I'm waiting for Intel Pentium 2 266MHz SL2HE as my other SLOT1 processors show weird numbers and MMX processor.
haker120 wrote on 2023-01-08, 12:44:I got this motherboard working (missing inductor taken from dead mobo) and I am confused what update file should I get? I only k […]
I got this motherboard working (missing inductor taken from dead mobo) and I am confused what update file should I get? I only know I have Award BIOS from 1997 and I'm waiting for Intel Pentium 2 266MHz SL2HE as my other SLOT1 processors show weird numbers and MMX processor.
Assuming you simply want the latest available BIOS for your board installed.
Looks like that board shipped with both 1Mbit and 2Mbit BIOS EEPROMs. You'll have to find out which one you have. Easiest and surest way would be to just look up the EEPROM part number printed on the top of the chip, but that would probably involve removing a sticker covering it, which you might not want to do. You could also dump the currently installed BIOS - use Awdflash bundled with any of those BIOS updates - and check its size. If the dump is 262144 bytes you got the 2Mbit chip and want the first download in your screenshot (3.2), if it's 131072 bytes you got the 1Mbit one and want the second one (2.5).
If you got the 1Mbit version, you might also want look into getting a suitable 2Mbit EEPROM and hotflash the 2Mbit BIOS to it. Might be a little bit scary if you haven't done it before, but it's easy enough. Just boot the computer to DOS, carefully remove the 1Mbit chip and install the 2Mbit one without shorting anything and then simply flash the 2Mbit BIOS with Awdflash as any other BIOS update. You can prepare by removing the chip before, and reseat it loosely, just enough for all pins to get connection in the socket. Makes the hot removal part a little bit easier.
If installing the latest official BIOS doesn't sort the weird numbers etc, the rom.by BIOS Patcher probably would be helpful. I've used it a couple of times to add support for newer CPUs in 440LX boards with Award BIOS.
The thing is I want to update it to make better s370 processors to work, if not I'll update with official one and stick with one that actually works. Thanks for detailed response @kaputnik. 😀
haker120 wrote on 2023-01-09, 15:24:The thing is I want to update it to make better s370 processors to work, if not I'll update with official one and stick with one that actually works. Thanks for detailed response @kaputnik. 😀
Hang on, you just mentioned "other Slot 1 processors", but here you're talking about "better s370 processors".
We could help you better if you were a bit clearer:
- which CPUs are you exactly trying to get to work?
- what happens with current BIOS when you try to run them?
- what makes you think a different BIOS, specifically one of the three listed, will change that?
MS-6117 is a plain retail board with BIOS that might complain with unknown CPUs, but should allow you to boot with them.
But it's an old board with VRM 8.1 spec. That means:
a) min 1.8V
b) max 11.3A at 2.0V, 10.2A at 1.8V
If you try to use a CPU that requests a lower voltage than 1.8V via VID0-3 pins, the VRM wil -by design- not output anything. So Coppermine CPUs will not work unless you have a slocket that lets you set VID pins with jumpers/switches to request 1.8V.
Then the second limitation comes into play: if you draw more than the specified current, voltage levels might drop (so the system might get unstable) or the VRM might get damaged. Celeron Mendocino runs at 2.0V, but anything above 366MHz will draw more than 11.3A (C400 uses 23.7W = 11.85A which you might get away with, C533 uses 28.3W = 14.15A which is definitely out). Coppermine CPUs draw less power at nominal voltage, but because the voltage is lower, that's still a lot of current which you will also draw at 1.8V. The Celeron 533A already draws up to 9.3A; if you look for a Coppermine Celerons with 1.7V nominal voltage you could get away with a Celeron 667 max.
None of the above regarding VRM capabilities and behaviour is in any way affected by BIOS...
To be clear, I ordered Intel Pentium 2 266MHz SL2HE to even make this mobo to work as my slockets1 do work but hang at weird processors' speed detected as '-MMX'.
I'm trying to make work one of my P3s (slot1 in modified slockets) or 500MHz mendocino in it's original(?) slocket that work with another slot1 mobo and detected as MMX 250MHz and hangs in POST.
Well, I assume it'll help as it did with my ASUS P2B but there it was detected as wrong (older) processor and didn't hang at POST.
And I know newer slot1 mobos give minimum 2.0V what I actually set with my P3 900MHz and it works in mentioned Asus.
This board has an i440LX chipset, no P3 will work at default speed on it due to FSB spec limited to 66MHz and in practice 83MHz being the upper limit of what's possible (mainly due to PCI being run way out of spec with 1/2 divider).
As I said, if you run a Mendocino >366MHz, you're overloading the VRM on this board. The P2B has a VRM 8.2 which offers the same voltages, but specifies 50% higher currents (up to 16.9A at 2.0V). So what works safely on the P2B will not necessarily work safely here. A Celeron 500 is a bad idea, as its max 13.5A current draw is higher than the max 11.3V current allowed at 2.0V according to VRM 8.1.
Have you measured VCore with one of these CPUs present? It's entirely possible your system is hanging after POST due to sagging voltage, not just because BIOS doesn't understand the CPU type.
I did not, then 2.8V Pentium 2 will be ok and 1.7 overclocked to 2.0 (or even not because this mobo gives minimum of 1.8V) won't? There is something I don't understand. 😁
haker120 wrote on 2023-01-09, 19:43:I did not, then 2.8V Pentium 2 will be ok and 1.7 overclocked to 2.0 (or even not because this mobo gives minimum of 1.8V) won't? There is something I don't understand. 😁
A switching power supply can deliver more current at a higher voltage. So a VRM that complies with Intel's VRM 8.1 spec is happy to deliver 16A at 2.8V to a Klamath P300. However at 2.0V it can't do more than 11.3A and at 1.8V it's limited to 10.2A. That means that it can supply 45W at 2.8V, but only 18W at 1.8V.
To make matters worse, if you overvolt a CPU, the current draw stays the same. So if your CPU would use 15.8W at 1.5V, it's drawing 10.5A. At 1.8V it would still draw 10.5A (and 19W) which is more than the 10.2A and 18W that VRM 8.1 can deliver at that voltage.
But this is all theory. It's possible the specific VRM on your board is specced to VRM 8.1 but is capable of delivering more. Or less, after 25 years. You have a system that won't boot, so something's wrong. That's why I recommend measuring VCore to see if this is the problem. If you're getting a solid 1.8V, it should be OK. If not, you know what your problem is.
Ok but I thought it's normal behavior for slot1 PCs for processors that mobo doesn't have microcodes for. As that was the issue with my other mobo as well and processor stopped freezing POST after update that included microcodes for P3. 😀
haker120 wrote on 2023-01-10, 11:26:Ok but I thought it's normal behavior for slot1 PCs for processors that mobo doesn't have microcodes for. As that was the issue with my other mobo as well and processor stopped freezing POST after update that included microcodes for P3. 😀
It can happen. Intel OEM boards always exhibited that 'feature', but retail boards generally just gave an error but allowed you to continue. I have an MS-6168 rev 2.0 with a BIOS that supports up to early stepping Coppermine CPUs, but happily lets me run my Tualatin P3-1400S on a Slot-T adapter for which it most definitely does not have correct microcodes. It tells me that it's an unknown CPU running at 1150MHz, but it's definitely doing 1400MHz at 133MHz FSB. That's what I'd consider expected behaviour, at least with an i440BX board with decent VRM and 1/4 PCI divider. Your MS-6117 has an i440LX with old VRM and only offers a 1/2 PCI divider, so you're not going to run any P3 at stock speed, even the 100MHz FSB versions.
Well, honestly, I don't plan to use full speed out of any processors I have, my plan is to make Celeron 766MHz 66FSB to work at 2.0V (slocket minimum voltage). 😀 But for this I need to get (hopefully tomorrow) compatible P2, do update, try Celeron and if works leave as is and if not then try to modify BIOS update with these microcodes and try again but hopefully with the first option.
There is no such thing as a Mendocino 766MHz 😮
The fastest Mendocino was the Celeron 533MHz and already draws too much current for this board's VRM. The Celeron 766 is a Coppermine Celeron that nominally needs 1.75V. 1.8V is no problem, 2.0V might fry it, and could take the VRM with it.
Also, unless that slocket has a VRM on it, it has no minimum voltage other than what you can set with its jumpers. It might not have documentation for 1.8V, but the only difference between 2.0V and 1.8V is VID2, which is 0 for 2.0V and 1 for 1.8V. It also needs to be 1 for 2.1V ,2.2V, 2.3V, 2.4V and 2.8V. So check for a jumper setting those all have in common, and set it in addition to what you set for 2.0V. Hey presto, you've set 1.8V. That will probably stop you frying the Coppermine CPU, but it won't stop you overloading the VRM.
See here for documentation, particularly page 7:
https://cdn.preterhuman.net/texts/computing/i … 0Guidelines.PDF
Ok, my mistake, I meant Celeron 766MHz. 🤣
Also I have P3 900MHz that is 1.7V at slocket set to take 2.0V in Asus and all is good and working.
And I heard my good acquaintance overvolted a processor 0.5V or more to the edge and everything was working. 😁
Guys help, I took BIOS file from mobo but I cannot flash new one because doesn't match something. BIOS in attachement, what to flash?
haker120 wrote on 2023-01-13, 13:22:Guys help, I took BIOS file from mobo but I cannot flash new one because doesn't match something. BIOS in attachement, what to f […]
Guys help, I took BIOS file from mobo but I cannot flash new one because doesn't match something. BIOS in attachement, what to flash?
Well, that's a 1Mbit EEPROM, so the 2.5 BIOS from your OP should work if you've identified your board correctly. Sometimes there are different HW revisions of the same board, needing different BIOS versions. Have you checked that?
It would probably help if you told us exactly what it was that didn't match too 😀
This to be exact.
haker120 wrote on 2023-01-13, 15:48:This to be exact.
Try the bios you can download here. Compared the ID from the POST string (440LX-W977-2A69JM49C-00) to the one from your dump, it's the same. That image should work with your board.
If it doesn't , we'll try a stopover on some version between the one installed on your board (1.4) and the latest (2.5). Have had to do that on one or two occasions 😀
Updated successfully and Celeron at 500MHz works!
haker120 wrote on 2023-01-13, 16:02:Updated successfully and Celeron at 500MHz works!
Yay, good job 😀
I wonder if Celeron 766MHz will work and do I have to modify slocket1 for it?
