VOGONS


Reply 140 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

No longer preventing boot and VGA card working...
So ..I assume the codes are correct...though they are a bit difficult to see in the photo/
Finishes at FF 63
So unless the POST card was damaged ....this should be correct ?

Reply 142 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie
weedeewee wrote on 2021-10-03, 18:03:

FYI. FF is the last code, 63 is the previous code

Thanks for confirming what I was thinking Weedeewee...
When it had booted and completed at FF ..I was fairly sure all was good..

Reply 143 of 165, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

Well that all looks pretty good, 2 working motherboards and a post card. Dunno why it gives the wrong CPU type, the only jumpers I can see missing are J2,8,9, but they're to do with the VLB slots. Do you know which version of the BIOS you're running? Maybe it's an earlier one that doesn't know about AMD DX4 (I think the Intel DX4 was a bit earlier).

I notice in the manual I'm looking at that a couple of things are only officially supported on the Rev 8 boards, like Write Back cache on the missing JP30, so that SV8B you have might not work. OTOH, the pads for JP30 are there on the board, so there's a small chance that fitting a couple of jumper pins and installing the most recent BIOS might make it work.

Reply 144 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

Hmmm..
it would be easy enough to fit JP30 but the side furthest from the cpu doesnt appear to go anywhere...
no traces top or bottom of the pcb to it...
...having said that ..it ocurred that it might be connected to ground internal plane...but it wasn't
Checked again with the meter to +5v ...which also might have a sandwiched layer in the board..
and yes it does..
So one side goes to +5v inside the pcb.
Since it is connecting +5v to ...something...
I might try and find out _what_ before I fit it

Reply 145 of 165, by Robin4

User metadata
Rank l33t
Rank
l33t
Vipersan wrote on 2021-10-03, 09:27:
I just Love old Tech in general Horun .. VCRs from the 80s ...PCs and Retro Computers in general ..Reel To Reel tape recorders e […]
Show full quote

I just Love old Tech in general Horun ..
VCRs from the 80s ...PCs and Retro Computers in general ..Reel To Reel tape recorders etc etc..
but then I am a child of the 60s and 70s..
My motto..if it can be fixed..it should be fixed.
..and to prove that love...
Here is the first snapped image of this motherboard booting again.
still reporting the wrong cpu.(must check all the jumpers)
maybe the new MX chip (when it arrives) will help but I doubt it...🤣
rgds
VS

What was the issue with it? I didnt following this thread, but still iam very inquisitive what the problem was.

~ At least it can do black and white~

Reply 146 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

So Snufkin..
from what I've traced out with the meter..
one side of JP30 is tied to +5v
The other side goes to a resistor/pull-up ? which in turn goes to Pin Ref B15 ..reserved on socket 3 ..
or WB/WT on the 486
Write Back Write Through.
So definately of interest and I shall fit this jumper tomorrow.

Reply 147 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie
Robin4 wrote on 2021-10-03, 22:26:
Vipersan wrote on 2021-10-03, 09:27:
I just Love old Tech in general Horun .. VCRs from the 80s ...PCs and Retro Computers in general ..Reel To Reel tape recorders e […]
Show full quote

I just Love old Tech in general Horun ..
VCRs from the 80s ...PCs and Retro Computers in general ..Reel To Reel tape recorders etc etc..
but then I am a child of the 60s and 70s..
My motto..if it can be fixed..it should be fixed.
..and to prove that love...
Here is the first snapped image of this motherboard booting again.
still reporting the wrong cpu.(must check all the jumpers)
maybe the new MX chip (when it arrives) will help but I doubt it...🤣
rgds
VS

What was the issue with it? I didnt following this thread, but still iam very inquisitive what the problem was.

Basically the fault (in short) was traced to an open circuit or very high resistance ferrite chip surface mount inductor.
Other ascociated components have already or will be changed in the near future.
This inductor is responsible for the 5v supplies to the clock processing/generation IC for the CPU ...so no clock ..no startup ...and the cpu remains in reset.

Reply 148 of 165, by Robin4

User metadata
Rank l33t
Rank
l33t

thx for your response.

~ At least it can do black and white~

Reply 150 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie
Bruno128 wrote on 2021-10-05, 09:30:

Can you please post the BIOS string here

The entire bios was dumped and uploaded on page 1 of this thread Bruno..

Reply 151 of 165, by Nexxen

User metadata
Rank l33t
Rank
l33t
Vipersan wrote on 2021-10-05, 11:49:
Bruno128 wrote on 2021-10-05, 09:30:

Can you please post the BIOS string here

The entire bios was dumped and uploaded on page 1 of this thread Bruno..

Maybe you could rewrite the first part of post n.1 and add all the files to have all the relevant data/files at the beginning, avoiding to go read all the posts.
After this bad sentence, just to make it a little more user friendly after all the pages.
Also add a pic of the screen just before booting (the one with cpu, ram and bios string at the bottom).

Troubleshooting is long and mainly has a lot of waiting. Glad you could do something with yours.

Last edited by Nexxen on 2021-10-05, 18:52. Edited 1 time in total.

PC#1 Pentium 233 MMX - 98SE
PC#2 PIII-1Ghz - 98SE/W2K

Reply 152 of 165, by Jasin Natael

User metadata
Rank Oldbie
Rank
Oldbie

Just wanted to say that this thread has been awesome. Very informative and interesting, and it's great to see a mostly positive outcome.

Reply 153 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

Finally got round to erasing a 27C512 ...in order to try the 486-GA-486SVF.BIN ....
No change as expected..
System boots ok ...CPU still misidentified as a DX2
I think we decided it was the same bios as mine earlier in the thread...but it doesn't hurt to try.
If anyone knows of the last bios for the GA-486VF please let me know ..
I do believe there was a REV 8c of this mobo so just possible I guess ...since mine is a rev 7
rgds
VS

Reply 154 of 165, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

The manual ( http://www.motherboards.org/files/manuals/47/486vs8a.pdf ) says that Write Back is support by Rev 8A PCB with the BIOS July 12th 1994 or later. It looks like the BIOS you dumped from your board is the same as on UH19 (04/27/94-SIS-85C471-2C4I8G01-00, http://www.win3x.org/uh19/motherboards/2716 ).

I found a link on the wayback machine to an old GigaByte support page that included the following for the GA-486VF:

Vsfbi.ZIP GA-486VS/VF BIOS File (06/27/95) […]
Show full quote

Vsfbi.ZIP GA-486VS/VF BIOS File (06/27/95)

Support LBA.
Support AMD DX4-100.
486VS/VF does not have a flash BIOS, if you need to upgrade this BIOS you have to have a EPROM burner.

So I had a look for the VSFBI.ZIP and found a Taiwanese site that had it. Looks like it has the BIOS string 11/21/94-SIS-85C471B/E/G-2C4I9G01-00

My guess is that whether it will work will depend on the version of the 85C471 (original, or B/E/G). Possible that the original version either has bugs or doesn't support everything needed for write back. Might still be worth giving it a try, it's unlikely to break anything.

Reply 155 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

Thanks buddy...
I've grabbed the dump..
Most likely try it tomorrow ...once I've erased the 512 again..
cheers
I'll let you know if anything changes.

Reply 156 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

@ Snufkin..
I wont get to test further 'till later...but you can see for yourself from the photo..
..that bios update (which was a ROM file not in BIN form) nonetheless ...allows boot and recognises the DX4

Still not convinced however that the V8T hasn't got an internal cache problem.
Currently have a replacement AMD Am486 DX4-100 NV8T on order but will take a while to get here
rgds
VS

Reply 157 of 165, by snufkin

User metadata
Rank Oldbie
Rank
Oldbie

Oh, I think that V8T has a problem (same symptom on two different boards, fixed in both cases by disabling internal cache), I just suggested you use it in case the changes to the clock circuit broke the CPU. Better to properly break an already faulty part than a good CPU. It's handy to have something around with a known fault but that still works well enough to use for testing. When you're happy that the clock circuit is ok and isn't going to break anything then it's probably worth trying the SV8B to see if that mod for J30 makes the write-back cache work on this board, or if other jumper mods (e.g. Jp24) might be needed. Good to see that the new BIOS at least gets past POST.

Reply 158 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

The MX clock chip arrived today ...fitted in place of the original...just in case.
No difference at all ..so I can reasonably conclude the one originally on the mobo is ok.
That said I've tested it and now will go in my spares box.
I've also now got a replacement keyboard controller also tested good...
so apart from the cpu which both myself and Snufkin are fairly sure is defective I would say this GA-486VF motherboard is healthy.
..and everthing tested so far boot wise with the internal cache turned off runs.
and the new bios identifies the CPU correctly.
I am just waiting for the replacement DX4-100 V8T to arrive so I can conclude the testing.
I've learned a lot with this...and am no longer scared to do such repairs..
In actual fact I'm looking out for my next faulty motherboard from this era to tackle..

Reply 159 of 165, by Vipersan

User metadata
Rank Oldbie
Rank
Oldbie

Gigabyte board update...
The replacement 100NV8T DX4 I ordered arrived today ...so of course I lost no time in trying it..
Conclusion..Internal and External cache enabled in bios ..
Successful boot from Floppy and Hard drive.
So it looks like my original V8T is at fault as we suspected..
I'll be labelling it accordingly and will now become a test chip going forward.
rgds
VS