B220 paper tapes

[Al Kossow]

Posting this here, since I'm pretty sure all of the same people are working on the B220 simulator

The five B220 paper tapes that I got from eBay are read.
I've included pictures from all of the auctions. I may try to go after TR1101

http://bitsavers.org/bits/Burroughs/B220/B220_Paper_Tapes.zip

The tapes are narrower than 1" since they are only 7 channel
They appear to be odd parity

[Paul Kimpel]

Thanks, Al. I have the tape images decoded and am trying to run them in my emulator now. I'll reply privately with the results.

For anyone who's curious about this, Al acquired some paper tapes that are reputed to be diagnostics for the Burroughs 220. I'll eventually post our experience with these on Tom Sawyer's 205/220 blog, https://datatron.blogspot.com.

On Saturday, August 1, 2020 at 9:58:47 AM UTC-7, aek wrote:

Posting this here, since I'm pretty sure all of the same people are working on the B220 simulator

<snip>

[Al Kossow]

The seller has relisted the remaining tapes.
There is only one that jumps out at me, T1101 (?) but it doesn't have a label on the box.
He also hasn't dropped the price at all.

Are there any still worth going after, or anyone willing to contribute to getting any of
them?

[Loren Wilton]

Al, if I have found the listings you are referring to, I make out the tapes
as being:

Lot #2 $25.00 TR1305 Magnetic Tape Margin Test
Lot #3 $29.00 TR1301 Mag Tape Standard Performance Test
Lot #7 $50.00 TR1501 Paper Tape Reader Standard Performance Test
(Data Tape)
Lot #9 $60.00 TR1101 (Systems Maintenance Test)
Lot #10 $39.00 TR2621 Part II unknown (sounds like a PPT reader
calibration tape)
Lot #11 $39.00 TR2071 unknown

I agree that his prices are a little dear, but I'd be willing to kick in 30
bucks to help defray costs for TR1101.

I can't see much use for the first three lots without actual original
hardware to calibrate, and it is hard to justify the remaining two without
knowing what they are really for. Does anyone have a listing of Electrodata
tape numbers so they could be looked up?

Loren

[Al Kossow]

On 8/29/20 11:44 AM, Loren Wilton wrote:
>
> I agree that his prices are a little dear, but I'd be willing to kick in 30 bucks to help defray costs for TR1101.

OK, will go after that one tape then

[Paul Kimpel]

I think Lot #3 TR1301 might also be useful, even without actual hardware. I'll kick in $59.00 for that and the balance of TR1101. Send the check to you Al? I have your CHM address.

[Al Kossow]

On 8/29/20 10:24 AM, Al Kossow wrote:

> The seller has relisted the remaining tapes.
> There is only one that jumps out at me, T1101 (?) but it doesn't have a label on the box.
> He also hasn't dropped the price at all.

The two tapes arrived and have been read and uploaded to bitsavers along with a picture
of the instructions in the cover of T1101

The .bin files have had the leader/trailer removed but still have bit 8 set

The .ptp files have the bit removed.

T1101 appears to be two tapes. I read the two tapes twice but the two reads
differ in part 1. I suspect the second read is correct. Part 2 matches between
the two reads. If neither of part1 is correct I will re-read but it is difficult
dealing with the tape because it is so long.

[Paul Kimpel]

Al,

Thanks for scanning these tapes and making the images available. I have been busy with an other projects and am only now turning to these tapes. I have attached a zip archive with the current conversion of the tape images to retro-220 emulator decimal images, disassembly listings of those images, and a transcription of my reading of the instructions inside the box lid for TR1101. This archive can also be downloaded from

www.phkimpel.us/Burroughs-220/software/Diagnostics/TR1101_TR1301/B220_TR1101_TR1301_20200923.zip.

The only problem I can see between the two versions of TR1101_part1.* is in the last (3000th) word. For read #1 the word is 0490001 (SLA 1, a left-rotate of the A register for one digit). For read #2 the word is missing entirely. Curiously, the word is present in both of the combined images, tr1101.bin and tr1101rd2.bin. I found quite a few other differences between tr1101.bin and tr1101rd2.bin, but they were all hex 80 bytes, which represent blank tape. Those get discarded by my conversion utility.

Looking at the code, I think that SLA word needs to be there, which would make read #1 the correct one.

My emulator currently halts in TR1101 at location 0445 in the first overlay after failing an IBB (Increment B and Branch) instruction. I can't see any problem with the way that instruction is executing. Its operand is modified earlier in the program, however, so I suspect the problem will take a while to run to ground.

TR1301 is going to require some study to understand how it needs to be set up.

Getting this far has required corrections to my tape image converter, my disassembler, and the retro-220 emulator, mostly to handle the full 64-character set supported by the 220 paper tape subsystem -- I had not been supporting the IBM 046 punch codes (END CARD 1, PI 1, etc.) that were not assigned printable glyphs. TR1101 requires those additional codes, however. These changes have been committed to the project repo (https://github.com/pkimpel/retro-220/) and the hosting site (www.phkimpel.us/Burroughs-220/) as version 1.03a.

I am not going to have much time in the next month to work further on these tests, so if anyone else is interested, go for it.