• Please review our updated Terms and Rules here

IBM PC XT 5160 BIOS versions

modem7

10k Member
Joined
May 29, 2006
Messages
10,424
Location
Melbourne, Australia
According to:
1. Various editions of Upgrading & Repairing PC's (Scott Mueller)
2. BIOS listing for 01/10/86 version
3. ROMs in possession
4. IBM's Technical Reference for the 5160
5. Internet sources
6. Contributions from forum members


-------------------------
08/16/82
-------------------------
Not much information on this BIOS.
Mueller's 'Upgrading & Repairing PCs' book doesn't list it, but some other sources do.
There is a dump of part of BIOS on the Internet. The date string within it is "08/16/82" and the copyright string within it is "5000026 COPR. IBM 1982"

ROM SIZE = 40K
ROM U18 = 32K chip labeled with IBM part number of 5000026
ROM U19 = 8K chip labeled with IBM part number of 5000027 (reused in the next BIOS - dated 11/08/82)

U18 contains the actual BIOS plus part of BASIC. U19 contains the remainder of BASIC.

It is possible that this BIOS did not make it into the 5160s released to the public.
If it did, there can't be too many about because this BIOS was replaced three months later (11/08/82).

Question: Does the source code for it appear in the first edition of IBM's Technical Reference for the 5160?
Question: Does anyone actually have a "5000026" stamped chip?


-------------------------
11/08/82
-------------------------
ROM U18 = 32K chip
ROM U19 = 8K chip
KNOWN ROM SETS = U18=1501512/U19=500027 or U18=1501512/U19=6359116
ID/SUBMODEL/VERSION = FE/--/--
* Used on 64/256K motherboards.
* "The original XT BIOS had all of the features of the latest 10/27/82 PC BIOS."

Memory map:
. = unused
b = ROM BASIC
R = ROM BIOS
------: 0---1---2---3---4---5---6---7---8---9---A---B---C---D---E---F---
0F0000: ........................bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbRRRRRRRR
------: ........................|< U19 >|<----------- U18 ------------->|


-------------------------
01/10/86
-------------------------
ROM SIZE = 64K
ROM U18 = 32K chip
ROM U19 = 32K chip
KNOWN ROM SETS = U18=62X0851/U19=62X0854 (PROM) or U18=62X0852/U19=62X0853 (EPROM)
ID/SUBMODEL/VERSION = FB/00/01
* Used on 256/640K motherboards. Up to 640K RAM.
* Enhanced POST (one known enhancement is a faster test of RAM).
* Added support for the 101-key Enhanced keyboard. (Note: Some later released 101-key enhanced keyboards dropped support for the IBM 5160.)
* Support for 720K floppy drives. Support for 1.2M floppy drives. (Note: 1.2M drives also require a 1.2M compatible floppy controller.)

Memory map:
. = unused
b = ROM BASIC
R = ROM BIOS
------: 0---1---2---3---4---5---6---7---8---9---A---B---C---D---E---F---
0F0000: RRRRRRRRRRRRRRRRRRRRRRRRbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbRRRRRRRR
------: |<----------- U19 ------------->|<----------- U18 ------------->|


-------------------------
05/09/86
-------------------------
ROM SIZE = 64K
ROM U18 = 32K chip
ROM U19 = 32K chip
KNOWN ROM SETS = U18=59x7268/U19=62x0819 (EPROM) or U18=62x0890/U19=68x4370 (PROM)
ID/SUBMODEL/VERSION = FB/00/02
* Contained some fixes for minor keyboard bugs related mainly to the enhanced keyboard.

Memory map:
. = unused
b = ROM BASIC
R = ROM BIOS
------: 0---1---2---3---4---5---6---7---8---9---A---B---C---D---E---F---
0F0000: RRRRRRRRRRRRRRRRRRRRRRRRbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbRRRRRRRR
------: |<----------- U19 ------------->|<----------- U18 ------------->|

 
Last edited by a moderator:
The XT BIOS doesn't have the cassette I/O support that the PC BIOS has.
It's actually 8K I think...

Although technically correct I always thought each ROM was 4k and since IBM had a HI & LOW rom you ended up with 8k total between the 2.
 
The XT BIOS doesn't have the cassette I/O support that the PC BIOS has.

"The original XT BIOS had all of the features of the latest 10/27/82 PC BIOS" is a quote from the 8th edition (and others) of Mueller's 'Upgrading & Repairing PCs'.
One mustn't read too much into that. So I wouldn't expect cassette support to be in an 5160 BIOS because the 5160 lacks cassette hardware.
Likewise, I wouldn't expect to see any code that attempts to read two switch blocks.

It's actually 8K I think
40-32=8 How did I end up with 4 ? I've just replaced the batteries in my calculator, so it doesn't happen again.
Thanks for picking up my mistake. I've corrected the post.
 
Although technically correct I always thought each ROM was 4k and since IBM had a HI & LOW rom you ended up with 8k total between the 2.
You're thinking of the AT. Two ROMs each supplying 8 data bits for a 16 bit data bus.
 
According to:
-------------------------
01/10/86
-------------------------
ROM SIZE = 64K
ROM U18 = 32K chip labeled with IBM part number of 62X0851
ROM U19 = 32K chip labeled with IBM part number of 62X0854

In case you're interested in another data point, I have a motherboard with
62X0852.U18 and 62X0853.U19. But their contents are the same as the ones
listed in your post for 10JAN86.

Bill
 
Last edited:
In case you're interested in another data point, I have a motherboard with
62X0852.U18 and 62X0853.U19. But their contents are the same as the ones
listed in your post fr 10JAN86.

Bill
Are the checksums of the contents (and not just the bios date) the same as the other ones?
CRC32 Checksums should be:
62x0851 = 1054f7bd
62x0854 = b5fb0e83

If the checksums are not the same, I'd be interested in seeing what the differences are.

LN
 
Are the checksums of the contents (and not just the bios date) the same as the other ones?
CRC32 Checksums should be:
62x0851 = 1054f7bd
62x0854 = b5fb0e83

If the checksums are not the same, I'd be interested in seeing what the differences are.

LN

The entire contents of the ROMs are identical, byte for byte.

Bill
 
All models of the XT use the same basic board. You can upgrade from a 11/08/82 to a 05/09/86 BIOS merely by replacing the ROMS/EPROMS. This would suggest that the U19 ROM/EPROM is actually a 32K chip with only 8K used. Otherwise, how does the board deal with the two extra address lines when an 8K chip is installed?
 
All models of the XT use the same basic board. You can upgrade from a 11/08/82 to a 05/09/86 BIOS merely by replacing the ROMS/EPROMS. This would suggest that the U19 ROM/EPROM is actually a 32K chip with only 8K used. Otherwise, how does the board deal with the two extra address lines when an 8K chip is installed?

Mine, 62X0853.U19, has about a 16K hole in it. Starts about 1D38 and ends at 5FFF.

It also contains the string 62X0854 at the beginning.

Bill
 
Otherwise, how does the board deal with the two extra address lines when an 8K chip is installed?

The 8Kb of the 8K EPROM will repeat 4 times through the 32Kb a 32k EPROM would appear in. Half of the BIOS (early revision) does this, and you can easily see it in Debug; Just type:

Code:
D F000:0000
D F000:2000
D F000:4000
D F000:6000
 
hello I have an original IBM XT board and when it boots it gives an 601 floppy error because I have not put an original floppy controller. I have read that on other versions of bios this does not happen so I would like to reprogram the bios and basic chips.
are there any bin/hex/rom files for these so I can reprogram my motherboard?
 
hello I have an original IBM XT board and when it boots it gives an 601 floppy error because I have not put an original floppy controller. I have read that on other versions of bios this does not happen so I would like to reprogram the bios and basic chips.
are there any bin/hex/rom files for these so I can reprogram my motherboard?

Gosh! I tried to read the bios and basic chips using my willem eprom reader and I kept a backup file.
Now I re-insert the original chips on the motherboard and it does not boot! what should I do???
 
hello I have an original IBM XT board and when it boots it gives an 601 floppy error because I have not put an original floppy controller. I have read that on other versions of bios this does not happen so I would like to reprogram the bios and basic chips.
That is a known issue for the IBM AT (IBM 5170). Where did you read that it happens on the IBM XT (IBM 5160) ?
 
Gosh! I tried to read the bios and basic chips using my willem eprom reader and I kept a backup file.
Now I re-insert the original chips on the motherboard and it does not boot! what should I do???
Some possibilities:
* You've inserted the ROMs in the wrong orientation/direction.
* You've put the U19 ROM into the U18 socket, and the U18 ROM into the U19 socket.
* Whilst inserting the ROMs, one or more pins have become bent.
 
Gosh! I tried to read the bios and basic chips using my willem eprom reader and I kept a backup file.
Now I re-insert the original chips on the motherboard and it does not boot! what should I do???

I read the bios chip of one vga card and my willem corrupted the EPROM! The contents of the backup file and the eprom after reading it were different. I flashed the backup file to a 27C512 eprom and the vga card worked again.

Don't trust the willem readers, they are not the most reliable!
 
I read the bios chip of one vga card and my willem corrupted the EPROM! The contents of the backup file and the eprom after reading it were different. I flashed the backup file to a 27C512 eprom and the vga card worked again.

Don't trust the willem readers, they are not the most reliable!

When it comes to BIOS files, you can just dump them using simple software routines.

When I dump other ROM chips, I allways use a card with a BIOS-extension ROM socket on it. For general 27xxx chips, I use some ethernet card, and for the more uncommon 8KB chips used in the early PC's, I use the ROM socket on the IBM HDD controller.
 
Hi.

I think "5000026" (08/16/82) ROM version.
Before this, I've checked dumped of other IBM PC 5150 (Revision 3, 10/27/1982) and IBM PS/2 Model 25 (Revision 1, 06/26/87)

I found that ROM of IBM PC 5150 Revison 3 exists as 2 kinds. (One is copyrighted 1981, another is 1982)
Also ROM of IBM PS/2 Model 25 Revision 1 exists at least 2 kinds. (One is 00F2092, another is 00F2122)

So I'm sure "5000026" ROM is not fake. (At least this must be prototype or beta, not for retail)
But it is not only verified if this ROM is exist or not.
I've checked checksum of 8KB.
Checksum is correct as 00h.

IBM should replace ROM chip of "5000026" with "1501522", but the reason is unknown.

The more interesting thing is that BIOS data of IBM XEBEC HDC (5000059) is also same "08/16/1982"
 
Last edited:
If I can get a PC/XT 256-640K system board that on the backside says "YB 620-6181655-01 94V-0 Rev 1 date on bio chips is 1981,85", would that board have the latest 4/86 BIOS that supports the 101 keyboards, and if not, can the BIOS be upgraded/flashed using a diskette?

Regards,
Mike
 
Back
Top