Support and general discussion.
Post Reply
Posts: 36
Joined: Mon 15 Feb, 2016 8:27 pm


Post by BigAlUK » Sat 29 Sep, 2018 2:23 am

I was quite stunned that I appear to be the only one to notice this, (or at least mention it on the web), but...

First read the comment on MinusZeroDegrees about this BIOS.
[Update: Turns out this particular part was another red herring - see next posted reply below. But read on, all the same.]

Notice the assumption that it was for the IBM PC XT? The 5160?
[Update: Correct. In this case, but...]

Well now have a look at page A-108 in the April 1983 Technical Reference for the 5150 PC.
Notice the BIOS date in that listing?

Everybody seems to assume that there were only three BIOS issues for the IBM PC 5150.

In fact this listing in the Tech Ref manual is not one of those three!

How has this ever been missed?

Is it because it has exactly the same part number as the 10/27/82 BIOS? Take a look at page A-5 of the April 1983 document, and then compare that with page 5-33 of the April 1984 version of the document. Notice the difference? No? That's because there isn't any - they have the same part number and copyright date - but look at the date of the one in April 1984 document (page 5-114). Yep, it's our beloved "third" BIOS - and it actually appears to be the fourth. It would be quite useful here if anyone who has a 5150 that was actually manufactured and sold before April 1983 to check whether the motherboard is carrying this mysterious 08/16/82 BIOS! Then it might be possible to do a ROM-compare (a bit easier than comparing every line of the two listings in the two Tech Refs).

While we're on this subject - what the heck is going on with this "MAME" BIOS ROM that has a 1982 copyright date but is otherwise identical to the 10/27/82 BIOS? Was this someone objecting to the 1981 date for some reason? I'm kind of assuming this isn't a real part - which is potentially ironic, since the MAME team appear to be promoting it as such over the actual real part: it has a different checksum thanks to this 1981-1982 copyright date switcheroo. Tut tut. Of course, it might just be that it really did come from a real ROM taken from a real 5150 machine - but it strikes me as a pretty bizarre thing for a manufacturer to so randomly modify something without altering the part number. Certainly when it's as big as IBM. Very strange. And yet... remember what prompted this post?

I'd really love for someone to take a look at the part number written into this mysterious 08/16/82 BIOS ROM (after carefully confirming that it isn't the XT one with that same date - so confusing!) to see if it also matches that in the April 1983 listing, since it already matches the date. Anyone got a verifiable dump of it?
Last edited by BigAlUK on Sun 30 Sep, 2018 2:39 pm, edited 5 times in total.

Posts: 36
Joined: Mon 15 Feb, 2016 8:27 pm

Re: THE MYSTERIOUS 5150 08/16/82 BIOS

Post by BigAlUK » Sun 30 Sep, 2018 2:24 pm

Ok - I found the dump on the MinusZeroDegrees site, and the particular ROM he refers to is definitely an XT image - that matching of the date and the identity byte to a listing of a 5150 BIOS was just pure "coincidence".

Also, the larger ROM size should've given the game away.

I've slightly updated the original post accordingly, so as to avoid any further confusion with the XT part.

So that red herring aside, I resummarize: it's very strange the the actual listed BIOS in the 5150 April 1983 Technical Reference should have this same date - especially as that obviously is not the date that appears in the earlier 10/27/82 BIOS ROM, which should have been current when the manual was published six months later. Was this just poor configuration control by the writers of the manual?

So that's the real mystery here: Where is this 08/16/82 5150 BIOS that appears listed in the 5150 April 1983 Technical Reference instead of the expected 10/27/82 BIOS ROM?

Was it ever released? Has anyone got one? Is the manual listing different to the actual ROM everyone assumes it lists? For safety's sake, when looking for disassembly of the 10/27/82 BIOS, one should be careful to use the April 1984 version of the Technical Reference manual, and not entirely trust the listing of that same part number in the 1983 one. It is entirely possible - even likely - that it is not just the date in that BIOS listing that differs. It may even be that the date on the BIOS could further differentiate between 16-64K (Rev A) motherboards and 64-256K (Rev B) motherboards. Ha. If only life were that simple, eh?

Post Reply