PDA

View Full Version : Anonymous i430FX Clone



Ole Juul
March 16th, 2012, 11:30 PM
My headless secondary DOS box lost it's BIOS - making it troublesome. Upon much inspection it dawned on me that it has one of those &!@# Dallas non-batteries. So, since it has a nice gold top in it, I'll take it out so that can be seen, and replace the board with something appropriate.

Digging around I found a socket 7 board that had 4 ISA and 3 PCI slots ... good enough. More digging through CPUs, and I decide on a Pentium 100. It's the right year, a little less favoured among other people, and is a number that none of the other Pentium machines around here has. Perfect.

Now to figure out the jumpers. It wasn't all obvious on the board, but I did finally find a very grainy pdf of a Triton manual for a board which looks identical. I wasn't comfortable until I could find the voltage jumpers. Now I'm ready to plug it in, but would still like to know some more about this board if anybody has any information. It is a very interesting one, which is why I picked it. First of all it has a large white socket labelled "VRM", presumably for the VR-Module. Can someone explain what that would be, and specifically if I should have those four jumpers sitting in it which are there now?

From my searches, it would appear that the board is a PC Chips M506 or thereabouts. This is a copy of the Intel i430FX based boards, if I understand correctly. Other interesting information is that there is only solder pad where there should be a header for a mouse connection. Apparently that is because the UMC 8663 cannot support a mouse and keyboard at the same time. Hehe, this is looking more and more like a classic second rate board. The crowning glory however is the two "write back" chips. The first thing I noticed is that they were a shiny black with very clear marking of the function rather than a brand name and a model number. If you look closely, it is obvious why they don't have a brand marking - they don't have connections! They're completely fake. I've read about those, but now I have some. Very cool. :) This also explains why the MB manufacturer was not too keen on putting any identifying marks on it. Check it out:

http://i793.photobucket.com/albums/yy217/Eyonymus/Computers/bits/MB-X.jpg?t=1331966004

Caluser2000
March 16th, 2012, 11:46 PM
Just put the rodent on a serial port. Seemed to work aright on my 286 :)

Ole Juul
March 17th, 2012, 12:05 AM
Just put the rodent on a serial port. Seemed to work aright on my 286 :)

What rodent?

Mention something and I've probably got a box of them - serial mice is no exception, but I don't use them. I recently decided to give Arachne a try after all these years (for posting here), and had to load a mouse driver, plug in an adaptor to the serial header, and plug in a mouse. If you were here, you would be able to see me blush as I say those words. :p

Anyway, this is a headless DOS box so it wouldn't do much good.

Caluser2000
March 17th, 2012, 12:09 AM
I did gather that. It was just a cheezy comment.

Ole Juul
March 17th, 2012, 12:53 AM
I did gather that. It was just a cheezy comment.
Hehe, and most appropriate too. This is a pretty cheezy board. :)

It's certainly no smorgås board! I think it will work well for this. I connect with parallel interlnk, and read extra HDDs and Floppy Drives. I don't think there'll be any bottle necks or glitches. It's probably a pretty good board for basic stuff - as well as being a classic of marketing sleaze.

Chuck(G)
March 17th, 2012, 02:04 AM
Some of the old P1 AT boards would leave a header for a PS/2 mouse--you bought a cable with a mini-DIN on one end and figured out where to mount it.

I think I still have a couple of the Amptron 8600 boards (made by PC Chips), but they're not nearly as stripped down as your board. And both still work just fine today. They have integrated VR onboard, however. At least one is running an IBM-branded PR300 MMX CPU. Yours obviously takes a plug in module of some sort--I don't believe that there were any standards on those.

Ole Juul
March 17th, 2012, 03:38 AM
Yours obviously takes a plug in module of some sort--I don't believe that there were any standards on those.
Yes that module plug in is a new one on me, but then there's lots I haven't seen. It might be a couple of days before I get time to stick this in the box, but my worry is about those red jumpers in the module plug. The Triton manual makes no mention of it.

Hmm, as I write this, I just noticed the same (and red too!) jumpers in the same relative position on a hazy picture of a very close board. I think I'll take that to mean that they're original and placed right.

I wonder if I need a fan on that. My P133 has a taller heat sink and runs 24/7 with no fan and no problems. This being a P100 (3.3v) I'm going to assume it will be fine. I've got more of those chips, but chose this one because of the already attached heat sink.

keropi
March 17th, 2012, 04:36 AM
nice find.... IMHO if the cache is fake on this mobo (typical pc-chips) then don't bother at all with it unless you want a slugish p1 system for some reason...
Also I remember reading somewhere that even if you plugged a cache module on the special port it would not work either... yep so many wonderfull things those pc-chips boards have to offer :p

RJBJR
March 17th, 2012, 04:10 PM
Looks like M507 (which is close to M506 & M505)
http://motherboards.mbarron.net/models/pcchips/m507.htm

Ole Juul
March 17th, 2012, 04:53 PM
Looks like M507 (which is close to M506 & M505)
http://motherboards.mbarron.net/models/pcchips/m507.htm

Thanks RJBJR! I think you nailed it. That's the one. The two fake chips are there, and the 4 red jumpers are there too. This information is the same as I found, except specific to this board and clearer. It's what I need for my collection. I'd been to that site, but was looking at the 506 and didn't realize they had other related ones since the search didn't work for me.


keropi: nice find.... IMHO if the cache is fake on this mobo (typical pc-chips) then don't bother at all with it unless you want a slugish p1 system for some reason...

I'm not sure it will be sluggish. When I get it in, I'll compare it to it's mate which is a P133 on a good board. I'll write a floppy on each one and report back with the speed difference. Anyway, if it works properly, then it's really the collector's value thats important to me. I could put in the fastest chip it will take, or get a faster board with a faster chip, or put in and even faster board with an even faster chip than that. :p

keropi
March 18th, 2012, 01:22 AM
Ole Juul:

try and compare speeds with something else than writing a floppy, maybe some SpeedSYS benchmark or a DOS 3D game (quake?)... the best way to compare would be to use the same components (cpu/vga/ram) on both motherboards, so you can really tell the difference

Ole Juul
March 18th, 2012, 03:32 AM
Ole Juul:

try and compare speeds with something else than writing a floppy, maybe some SpeedSYS benchmark or a DOS 3D game (quake?)... the best way to compare would be to use the same components (cpu/vga/ram) on both motherboards, so you can really tell the difference

Hmm, I just got the thing together half an hour ago and connected to a time server so I could set the clock, and it all works fine except floppy reads. Odd. I'll check that out tomorrow.

I don't think it makes much difference what speed it is. The machine will be (as before) run headless so the I/O using a parallel port network connection will be 15200 AFAIK. Any other bench marks will probably not be applicable, except like I mentioned, the floppy read/write and that, as we know, won't change anyway. This machine is mostly for the remote use of a 1.2 meg floppy drive since I have the 360 in my main DOS box - and both of them need a 1.44. It also contains the HDD of a computer called Marasmius which has followed me for years and I refer to for the oldest of my files. In short, it is really a part of my main DOS machine, just an extension on the end of a wire.

Quake? hehe I'm half a century older than you probably think. :) There's no mouse possible on this thing and I didn't add an external serial port since I had no particular use for it. The headers are there if I change my mind.

You can see the arrangement here (http://i793.photobucket.com/albums/yy217/Eyonymus/c4-a.jpg?t=1250041017). The top one is ANA, my main DOS box (it's rocket fast) but before it got a CF drive and the 360 floppy. The lower one is the old version of MAR which I just rebuilt. Now it just has a video card (for setup and special occasions) and a network card. I put the cards right at the top so I can use the bottom for storage.

keropi
March 18th, 2012, 06:25 AM
ah, now I understand the purpose of this machine... it doesn't matter if it has cache or not , it's all the same :)

Ole Juul
March 18th, 2012, 04:21 PM
I don't do stuff in a normal fashion. hehe

Talking about cache, I found the solution to the broken floppy read problem. When looking through my configs, I noticed SMARTDRV which is standard for me. It occurred to me to rem it out and try again. Floppies now read and write! I'm going to put that in another thread about caching, as I think there are some interesting issues there.

keropi
March 18th, 2012, 04:37 PM
IIRC I had some weird issues with smartdrv and SHSUCDX (the FreeDOS MSCDEX replacement) , I eventually dropped SHSUCDX due to incompatibilities with some of my old games but I kept smartdrv from caching any removable media just to be sure...

Ole Juul
March 18th, 2012, 05:57 PM
IIRC I had some weird issues with smartdrv and SHSUCDX (the FreeDOS MSCDEX replacement) , I eventually dropped SHSUCDX due to incompatibilities with some of my old games but I kept smartdrv from caching any removable media just to be sure...

Very interesting. I run SHSUCDX too - on both DOS machines. It sounds like the functionality of the MB that I refer to in this thread may not be related to these problems then. I'd be curious how you determined the connection between smartdrv and SHSUCDX, but I'll post another thread about smartdrv that right now.

Edit: The smartdrv thread is here (http://www.vintage-computer.com/vcforum/showthread.php?29819-SMARTDRV-caching-and-floppy-read).

Ole Juul
March 18th, 2012, 11:26 PM
Just for interest sake I ran DriveSpeed (v3.10) on this machine (P100) with the fake cache, as well as the better DOS box with the ASUS (P133) mother board. It's about what I would have expected, regardless of cache. Smartdrv is not used on either drive.


P100 using disk drive:
ramdisk write 18.3 MB/Sec read 18.3 MB/Sec
hdd write 0.747 MB/Sec read 1.66 MB/Sec

P133 using CF drive:
ramdisk write 41.5 MB/Sec read 30.7 MB/Sec
hdd write 0.853 MB/Sec read 5.49 MB/Sec