First post, by Sphere478
- Rank
- l33t++
I probably should have made a thread about this a while ago but better late than never. this will be my build log for my dual pentium 1 build
It hasn’t been a very successful build though, I originally wanted to put windows 7 on it which unless I can figure out how to make the bios acpi compliant I don’t think will ever happen.
I was pleased to find that I was able to install 768mb of ram in it though which is 256mb more than what it’s supposed to be able to use. But 256mb less than I ordered for it 🤣. unfortunately the slots that need to be used to make it work are 3-8 and the 8th one on the corner was broken in shipping so it’s kind of rigged in there. With it in the undammaged slots it doesn’t detect the full 768 and with all 8 installed it doesn’t post
(Edit 768 showed but doesn’t work, 768 is unstable, 512 works and is stable.)
Also the rtc was totally shot. I got it working enough, though not entirely by soldering a second battery to it and have a nice replacement on the way that incorporates a actual replaceable battery. Hopefully the new one will fully solve the issues with having to set the bios settings again at random intervals . until i temporarily rigged it it had basically made the board unusable with constant errors halting post and inability to change bios settings
The major disappointment is that I have tried to load a OS onto it probably about a hundred different times and ways and nothing seems to want to work the OS that I am trying are 2k, xp, vista, 7, freebsd, haiku, beos, openbsd, fedora 11, debian jessie. All continue to fail. For a variety of reasons, some relating to the mass storage card I was trying to use (one of those via combo cards, yeah, don’t get one the reports of them corrupting your drives are totally true) but I think the main thing is that the ide firmware on this board is just glitchy and not even jan’s patched bios was able to fix it but it did help and I think it’s somehow even affecting the add in cards I’ve tried onboard ide, a sil0680, the via card, all installations fail for a variety of reasons or no apparent reason at different times. Like this latest attempt with the sil0680 xp sp3 I load the sil0680 driver from the disk and make it to setup and it says no drives detected but clearly in bios it is shown as 120gb pny ssd
I may have more success with a smaller drive though a 60gb ssd didn’t work on the same settings also. But that’s currently the smallest drive I own except cf cards and those too have not worked (6gb) though, I haven’t tried it since the jan bios.
I’ve literally been trying to get this thing to work it seems for like a month with no real sucess my next thing to try is ordering a promise tx4 300 controller card which I’ve had very good success with on the k6 build. Ultimately it may turn out that this board is nothing more than a paperweight which would be unfortunate. And the more time goes on the more I wonder if maybe I need to re cap the whole board.
Figured out a funny way to make the at-atx adapter to work just ground both and use the power supply switch 🤣
It originally came with some ugly little yellow heatsinks with annoying fans that plugged into molex. Which were shot anyway and all rattely
I ordered these wakefield heatsinks and love them but I did have to clearance them for the levers and the rtc
here is some information I was able to figure out while building it
Re: Dual CPU Socket 5/7 Boards
btw incase anyone is wondering […]
btw incase anyone is wondering
undocumented voltage settings for tyan tomcat s1564 are as follows
vd0 vd1 vd2 vd3
no no no no 1.2v
yes no no no 2.15v
no yes no no 2.25v
yes yes no no 2.35
no no yes no 2.45v
no yes yes no 2.65v
yes yes yes no 2.75vin addition to documented settings of:
no no no yes 2.8v
yes no no yes 2.9v
no no yes yes 3.2v
yes no yes yes3.3v
no yes yes yes 3.4v
yes yes yes yes 3.5vas you can see the voltage of 2.15 should be good for k6+ cpus and probably tillamook
it appears there is also a VIO jumper that should be installed for dual voltage chips if i am reading the manual correct.
voltage can be checked between the computer case for ground and the forward most fet where the flat large metal part on the back of the fet is soldered to the board
I think there is a undocumented setting for (yes yes no yes) but didn’t test it. It’s probably one of the mid-high range settings.
fsb setings
j13
50mhz fsb 1-2,3-4
60mhz fsb 3-4
66mhz fsb 1-2
?? mhz fsb no jumper (no post)
?? mhz fsb 2-3 (didn't try)multiplier settings
j19 j20 j21
1.5x/3.5x no,no,no
2.0x yes,no,no
2.5x yes,yes,no
3.0x no,yes,notried dual k62 (one 366 and one 500) both were 2.2v cpus one boots and bios sees as k6, two won't post. performed test on 2.15v and 2x multi(6x setting)
tried a k62+ 570 boots works fine on 2.15v and 2x multi(6x setting) bios detects at 66mhz 486 but it's running at 66x6 (4oo) in reality.
tried a tillamook 266 bios detects as tillamook runs fine at 233 had to jump the pins on the cpu package to get cache working, jumping the pins on the cpu to make 4x work doesn't seem to have worked seems to be at 3.5x
tried a tillamook in primary socket and a 233mmx in the second socket, it boots but no two processor detection in bios768mb is achievable and stable by installing 128mb edo 60ns simms in slots 3-8 but 1gb will not post using all slots.
Attachments
D4F0B6BC-8B42-49D9-A3C6-9F2A8103A7C4.jpeg
FilenameD4F0B6BC-8B42-49D9-A3C6-9F2A8103A7C4.jpeg File size1.72 MiBViews297 viewsFile license
Public domain