• Please review our updated Terms and Rules here

IBM 5160 XT early BIOS 5000026 found with interesting date code

C

chris_nh

Experienced Member
Joined
Jan 28, 2019
Messages
372
Location
New Hampshire
Hi everyone, I want to share this interesting find with the vintage community. It is an early BIOS ROM for the 5160 XT that was recently rescued from ebay. The date code is 8242, which is a few weeks earlier than the other known 5000026 ROM ICs.

There are 3 examples documented at the excellent minuszerodegrees.net site here: https://minuszerodegrees.net/5160/bios/5000026_5000027_photos.htm
These known examples have date codes 8304 and 8251. I do believe the 8304 date code is the last week for these early BIOS ROMs because I have seen the 1501512 ROM with an 8305 date code.

Likewise, I found another example of a 5000026 ROM on the vogons site here: https://www.vogons.org/download/file.php?id=124484&mode=view with a date code of 8249.

This latest find is the 5th now that I know of, but I'm sure there are more out there yet to be found. The date code of 8242 is surprising and would be the earliest known date found so far.

Anyone possibly know the meaning of the "YH" code at the bottom right of the markings?


IBM_5000026_8242_ROM.jpg
 
modem7 said:
8242 is the week starting 18 Oct '82. That postdates the date, "08/16/82", in the code of the ROM.
Click to expand...
My surprise, i suppose, was that I had assumed these ROMs were produced for just a short window of time between Dec '82 and Jan '83, but the October date code changes my assumption, as they were obviously produced over a wider range of time, or possibly in more than one batch. I too would be very surprised if the rom pre-dated the 8/16/82 date.
 
chris_nh said:
My surprise, i suppose, was that I had assumed these ROMs were produced for just a short window of time between Dec '82 and Jan '83, but the October date code changes my assumption, as they were obviously produced over a wider range of time, or possibly in more than one batch.
Click to expand...
ROM production may have even started as early as August.
 
I made an interesting observation regarding the "YH" stamp and early IBM rom chips...

1) The 5000026 chip is dated 8242 with a "YH" stamp. (none of the known later date codes have it)
2) I found a picture of a 5700051 (early PC BIOS) chip which is dated 8124 and also has a "YH" stamp.
3) Then I noticed, on the minuszerodegrees.net site here, there is a 5700051 chip with date 8123 and "YG" stamp. (the week is 1 earlier, and letter G is 1 previous to H. Maybe a coincidence, but more data is needed. I believe 8123 is the earliest known date code for that rom as well.

On the second and third observations above, it's evident the G and H characters appear a bit spaced away from the Y character, and the H is smaller. Obviously they can change the letters on the stamp, so this looks like we may be able to loosely deduce that the first letter "Y" and the second letters (G and H) likely had different purposes or different significance. It would seem, it wasn't as important for them to appear together spaced properly, as it might with other portions of the stamp. Maybe the first letter then signifies something like a proof set or a first batch, and the second letter would perhaps signify something like production station or a position identifier for multiple positions on a single wafer?

I would assume there would be some delay between the 8/16/82 date and the time it would take to prepare the photo masks and such process, correct?

Also an interesting coincidence, the date 8/16/82 would have a date code of 8234 at the earliest. Just for kicks (going off the single data point above), if we were to correlate the week number and the second letter, H being the 8th letter of the alphabet, a theoretical "YA" value would correspond with that date code of 8234. I'm not sure this is a likely explanation, however, because it would seem to not have any purpose to be correlated like that. This too may be a coincidence, but I think observing more of these chips will yield the answers.

Thoughts anyone? Someone who knows about mask rom production may have insight here. Would a code corresponding to wafer position make sense for a first production run for QA or tracking purposes?
 
Last edited:
Wow!

I have 2 of IBM PC XT 5160 MB(motherboard) (P/N:1501490) with 5000026 / 5000027 in working condition.

One S/N of MB is 11xx, Another S/N of MB is 30xx, but each date code of BIOS (P/N.500026) are same as 8251

In my opinion, motherboards in the range of S/N. 1xxx to 3xxx were mass-produced in a relatively short period of time.

Yes, It seems to be very early.

I guess S/N of MB may be early number than 500.

I wonder Susanin79's S/N of MB.

Also early date code of BASIC ROM (P/N 5000027) is 8240.
(My date code of them.)

The date code for the it I have is 8247A abd 8248B.

*The last date code of P/N:5000026 is 8304, but the first date code of P/N:1501512 is 8304, not 8305.
IBM 1501512-1.png
 
Last edited:
A crude timeline:

'82 August: 08/16/82 BIOS. Date of code freeze. Testing end date unknown. Date of release-for-production unknown.

'82 November: 11/08/82 BIOS. Date of code freeze. Testing end date unknown. Date of release-for-production unknown, but could be April '83 (see below).

'83 January: Date of first edition of Guide to Operations.
'83 January: Date of first edition of Technical Reference. <------ Contains source listing for 08/16/82 or 11/08/82 BIOS ?

'83 March: The IBM 5160 was announced. This does not necessarily correspond to 'units ready to ship to customers' (the IBM 5150 being an example).
'83 March: Technical Reference is revised. <------ Contains source listing for 08/16/82 or 11/08/82 BIOS ?

'83 April: Week 4 of 1984. "The last date code of P/N:5000026 is 8304" <---- 08/16/82 BIOS
'83 April: Week 4 of 1984. "but the first date code of P/N:1501512 is 8304" <---- 11/08/82 BIOS
'83 April: Technical Reference is revised. Contains source listing for 11/08/82 BIOS
 
In my opinion, it was produced with the existing P/N: 5000026 until the 4th week, and then suddenly changed to P/N: 1501512 in the middle,
so when they were produced in the same 4th week, they overlapped each other in date code.
It seems that the part of P/N: 5000026 that has already been produced is mounted without any change.
Later, among those who purchased without knowing this fact, those who are interested in the BIOS version, even those who are not interested in BIOS, may feel as if they were influenced by luck.

P/N attached to the photo. I had a chance to get the BIOS ROM chip corresponding to 1501512, but I passed it because I was not interested in the situation where I already had two 5000026 chips.
I remember someone later winning the bid for less than $30, but I kept the photo for the record.
 
fs5500 said:
I wonder @Susanin79's S/N of MB.
Click to expand...
Where I can found the S/N? I'm afraid that the S/N from the PC case looks modern and may be glued later. There is no much information on the box, only this label can be readable, see photos.
 

Attachments

  • IMG_1857.jpg
    IMG_1857.jpg
    1.3 MB · Views: 11
  • IMG_1864.jpg
    IMG_1864.jpg
    2.4 MB · Views: 11
You must log in or register to reply here.
Back
Top