PDA

View Full Version : Ibm xt 5160 5000026 bios



MicroCoreLabs
January 30th, 2017, 08:50 PM
I created a zipfile that contains the 5000026 and 5000027 BIOS PROMS for the first version of the IBM XT 5160.

I included the binary as well as a few hex versions such as Intel, Motorola and plain ASCII.

I may need help uploading the 1.7MB zip file, so in the meantime here is a picture of the PROMS...

35813

Enjoy!

-Edward

modem7
February 2nd, 2017, 07:37 PM
I have received the file from Edward, and over the weekend, will use it to update the IBM 5160 information and downloads at minuszerodegrees.net

modem7
February 4th, 2017, 04:57 PM
I have received the file from Edward, and over the weekend, will use it to update the IBM 5160 information and downloads at minuszerodegrees.net
I have done that (using the photo and the BIN files).

The photo of your 5000026/5000027 ROM set and that of Tyler Durden is at [here (http://minuszerodegrees.net/5160/bios/5000026_5000027_photos.htm)]. I couldn't help but notice the similarities and have come to the conclusion that you now have his chips (in a different motherboard).
A 5000027 comparison is at [here (http://minuszerodegrees.net/temp/3/temp_78hge09374h9g28.jpg)].
As for the 5000026, the comparison is weaker. The same corner is nicked. Note the line running from the top-left of "1" in "1982" up into the gap between the "0" and "2" in "5000026".

fs5500
February 4th, 2017, 05:59 PM
@modem7
Thanks very much.

By the way, can you show me checksum of 5000026 BIOS?
I also have the file of 5000026 ROM file ago.
Also you seem to receive hex dump of 5000027 file.


P.S. Thanks. I've checked it.

MicroCoreLabs
February 4th, 2017, 06:27 PM
I agree, those look like the same PROMs in a different motherboard. I still have this motherboard as I swapped it in an XT I just sold with one that had the rework to bring memory to 640KB.

fs5500
February 5th, 2017, 01:19 AM
The only problem is that IBM PC BASIC C1.10 doesn't launch if dumped BIOS (08/16/1982) is correct.
I think BASIC rom is not dumped perfectly.

If your IBM PC launches IBM PC BASIC C1.10, 5000026's ROM Binary is redump needed.

(If not, 5000026 is good dumped.)


I've added screenshot of PC-Tools System Information.


1) BIOS Revision (08/16/1982)
http://i.imgur.com/6Yb3JRJ.png

2) BIOS Revision (11/08/1982)
http://i.imgur.com/RC4mjSz.png


On PC-Tools, it is recognized as IBM PC on system information, (Not IBM PC XT)

MicroCoreLabs
February 5th, 2017, 09:32 AM
Hmm... I used an EPROM programmer to read each of the PROMs.. 32K for one and 8K for the other. I thought that would be the best dump method.

But if these are Mostek chips I now wonder if the address lines are not being accesses properly with the programmer. See link below.

http://www.minuszerodegrees.net/5160/motherboard/5160_u18_u19_sockets.htm

modem7
February 6th, 2017, 01:50 AM
The only problem is that IBM PC BASIC C1.10 doesn't launch if dumped BIOS (08/16/1982) is correct.
Cassette BASIC does not launch for me either. So, with no floppy controller fitted:
1. I see 601 error, then the 'Press F1' message.
2. I press F1
3. Screen clears, then cursor appears in top-left corner.
4. About 25 seconds later, the 5160 restarts.

If I swap out my 5000027 EPROM for the 5000027 masked ROM from my 11/08/82 BIOS ROM set, I see the same. Expected because they contain the same code.


On PC-Tools, it is recognized as IBM PC on system information, (Not IBM PC XT)
I am pretty sure that is because PC-Tools is examining the ID byte (model byte) in the BIOS.
See [here (http://minuszerodegrees.net/temp/3/temp_mb_uhgw8752h3gq98y8.jpg)] for the list from the 'Upgrading & Repairing PCs' book.
In the 08/16/82 BIOS, the byte is FF, and so PC-Tools assumes 'PC'.

Malc
February 6th, 2017, 08:29 AM
Didn't work for me either so I had a look at this 16th Aug 1982 U18 bios in a hex editor and it's nigh on a match for the 8th Nov 82 U18 bios except for the fact it contains no Basic code in the top half of the Rom.

The 16th Aug 1982 U18 bios code Starting from Hex: 4000 is pretty much the same as from the 8th Nov 82 U18 bios except for part No and date. And has been doubled up to fill the 32k rom.

fs5500
February 6th, 2017, 08:49 AM
@modem7

Thanks for informe me.

@Malc

Yes, you seem to be right.

Another way to check BIOS checksum is to use program on DOS's GWBASIC or IBM DOS BASIC.


10 CLS
20 DEF SEG=&HF000
30 FOR B=0 TO 7
40 GOSUB 60
50 NEXT:END
60 A=0:FOR I=&H2000*B TO &H2000*(B+1)-1
70 A=A+PEEK(I):NEXT
80 PRINT "SUM";B;"=";A
90 RETURN


SUM 0 = 2088960 (default, but variable, 0000-1FFF)
SUM 1 = 2088960 (default, but variable, 2000-3FFF)
SUM 2 = 2088960 (default, but variable, 4000-5FFF)
SUM 3 = 929280 (BASIC C1.10, 6000-7FFF) -- 5000027 ( U19 ) -- 8KB (BASIC ROM Area 1) -- BA1
SUM 4 = 994560 (BASIC C1.10, 8000-9FFF) -- 5000026 ( U18 ) -- 1st 8KB of 32KB (BASIC ROM Area 2) -- BA2
SUM 5 = 956160 (BASIC C1.10, A000-BFFF) -- 5000026 ( U18 ) -- 2nd 8KB of 32KB (BASIC ROM Area 3) -- BA3
SUM 6 = 979712 (BASIC C1.10, C000-DFFF) -- 5000026 ( U18 ) -- 3rd 8KB of 32KB (BASIC ROM Area 4) -- BA4
SUM 7 = 956928 (BIOS version 08/16/82, E000-FFFF) -- 5000026 ( U18 ) -- 4th 8KB of 32KB (BIOS ROM Area) -- BI


http://i.imgur.com/Rv8V6hi.png

http://i.imgur.com/CsKjvyU.png

modem7
February 11th, 2017, 04:36 PM
Edward made another copy of the 5000026 ROM, this time using DEBUG (which is valid for the IBM 5160 motherboard ROM sockets).

I noted that is has the same 24-bit checksum (3B5100) as U18 in the later 11/08/82 BIOS.

Available via [here (http://minuszerodegrees.net/rom/rom.htm)].

Thank you Edward.

fs5500
February 11th, 2017, 08:08 PM
@modem7

Thanks very much.

Now, all revision of IBM PC 5160 BIOS are verified.


P.S. The next is to verify all revision of IBM PS/2 8530 BIOS.

MicroCoreLabs
February 11th, 2017, 08:44 PM
I am glad to have helped!

Now I have a question: I swapped this motherboard into my "beater" 5160 to run Modem7's DEBUG program. This machine has a 20MB hard drive and with an AST sixpak has 640K. Hard disk formatted with DOS 2.1

The first time I powered up the machine it booted as usual from the hard drive, but the VGA card's BIOS banner which is usually blue was now black and white. It eventually went back to blue after a few reboots but I dont remember how many.

The second time I powered up the machine it hung at "Invalid boot disk" or something like that. Basically it would not boot from the hard disk or drop to BASIC. I tried power cycles a few times to no avail.

I tried running "sys c:" from DOS but that did not fix it. What fixed it was using Norton Utilities 4.5, I think it was Norton Disk Doctor, to make drive C: bootable. Thankfully that worked and now I am able to boot from C: again.

Normally I would just attribute this to being an old machine and something got jarred when I swapped motherboards. (Although it booted fine the first time)

Im just wondering if because the machine was running a later version of the 5160 motherboard with the updated BIOS ROMs and was then run with the 5000026 BIOS version that somehow this older BIOS affected the boot sector and made C: unbootable.

That or Modem7's DEBUG memory dumping routine is the reason which I find very unlikely.

Just something to think about for those who want to try the 5000026 BIOS ROMS in their machines with working hard disks. :)

-Edward

modem7
February 12th, 2017, 10:49 PM
Just something to think about for those who want to try the 5000026 BIOS ROMS in their machines with working hard disks.
Prior to publishing the second ROM image that you created, I put a copy of that image into my 5160, a 5160 fitted with an AST SixPakPlus, IBM hard drive controller, and ST-412.
I was able to boot to floppy (IBM DOS 3.3), hard drive (IBM DOS 3.3), and BASIC.
In the about 10 minutes of 'playing' that I did, I did not observe anything unexpected.

fs5500
April 12th, 2018, 10:08 PM
I've got original IBM PC XT 5160 motherboard with 5000026 / 5000027

https://i.imgur.com/gbgqZYg.jpg

https://i.imgur.com/eTxmdTo.jpg

fs5500
April 12th, 2018, 10:12 PM
https://i.imgur.com/rGbCNXS.jpg

https://i.imgur.com/vYmzyv0.jpg

modem7
April 13th, 2018, 05:58 PM
I've got original IBM PC XT 5160 motherboard with 5000026 / 5000027
So, a second 5000026 has turned up.
I have added the photo to [here (http://www.minuszerodegrees.net/5160/bios/5000026_5000027_photos.htm)], and altered the fourth bullet point at [here (http://www.minuszerodegrees.net/5160/bios/5160_bios_16AUG82_status.htm)].

fs5500
April 13th, 2018, 07:18 PM
@modem7

Awesome, thanks for update the information.

This is motherboard that is out of work.

So I put the bios on another 5160 board.

(BIOS has no problem to dump)

http://i.imgur.com/jvDLDgE.jpg

MicroCoreLabs
April 14th, 2018, 03:48 PM
So I'm not the only one in posession of rare ROMs worth thousands of dollars? :)

modem7
April 14th, 2018, 05:08 PM
So I'm not the only one in posession of rare ROMs worth thousands of dollars? :)
It's only a matter of time now before fake 5000026 ROMs appear on eBay. :-)

fs5500
April 14th, 2018, 10:18 PM
First time, I found dumped BIOS of 5000026 on MESS project a few years ago.
I thought it seems to be fake or beta BIOS.

MicroCoreLabs confirmed the chip and BIOS last year.
I found and bought the motherboard with same BIOS on eBay.

The motherboard seems not to work, but BIOS is safe, it is enable to work on another 5160 motherboard.
Not it has been dumped with DUMPPC.EXE and verified.