V2M10 partial deck found

[Jaume Lopez]

Hello,

I have recently acquired a deck of about 90 cards which relate themselves with the 1130. They identify themselves as Disk Monitor V2M10.

The case is, I would like to archive that deck, but due to its lack of integrity I don't know if it would be worth the effort. What do you think?

In any case, I would like to know more about it, and also about your archiving formats, please.

Thank you in advance!

Jaume

[Carl Claunch]

Hi Jaume

The control software for almost all IBM 1130 systems, Disk Monitor System, is organized into components

• Disk Utility Program
• Fortran compiler
• Cobol compiler
• Supervisor
• Core Load Builder
• Core Image Loader
• System IO Device Routines
• RPG compiler
• Assembler
• Resident Supervisor
• Cold Start program
• Shared data area COMMA

The components are assigned a range of phase numbers which are two hex digits long. For example, the DUP phases are from 01 to 12 hex. The Supervisor runs from 6E to 74. The Assembler starts at CF and runs to F8. 

Loading a disk cartridge with the control software is done by running a card deck that begins with the loader software and is mostly the phases in numeric sequence. At the end, it has many small library routine object decks to the stored into a library on the disk cartridge - DUP is started automatically after the last phase is loaded and then reads in the library decks. 

If this is the deck that someone else mentioned to me, with a first card of 1 SLS 0F 07/01/70 ASM 1130 DISKMONITOR 2 V2M10 PTM10001

which is a phase card. The number 1 is the marker for a new phase, the lead author responsible for the code in this phase has the initials SLS, the phase number should be CF but is 0F on this specific phase for some random reason. It is 0F on the decks to load the disk monitor V2 M11 and V2 M12 as well - some historical anomaly. This updated version was committed on July 1, 1079. The sequence number on the card, PTM10001, is the code for the first phase of the Assembler, which matches the text identification ASM 1130 DISKMONITOR 2 V2M10 on the card. 

In the decks we have archived, for V2 M11 and V2 M12, that first phase was updated so the card for the phase is 1 SLS 0F  07/01/70  ASM     1130 DISK MONITOR 2                V2M11 PTM11001

The sequence number begins with PTM11001 to reflect that it is a V11 version of the code. 

Loading the system involves thousands of cards and just the phases for the Assembler was 576 cards in V2 M11. Based on your estimate of about 90 cards, I believe this would be phases CF, D0, D1, D2, D3 and D4 only. Only the first phase, CF, is different in V2M11 from your deck, so only those 40-ish cards are different from what is already archived. 

In order to recreate the V2 M10 version of the Assembler we could merge your deck's first phase with the other phases from V2 M11 since only the first phase was modified to create M11. However, we don't know what other DMS changes it depends upon in V2 M10 since you only have this section of DMS. I am not sure of the utility of archiving the Assembler from V2 M10 but nothing else, but happy to do so if someone wants it.

The phases are object decks produced by assembling the source code for each phase. As such they use a condensed format on punched cards that requires all 12 rows of each column to be recorded when reading the deck in. We use the object card format from the SimH based 1130 simulator, which records each card column in a 16 bit word holding a bit for each row and padded with four 0 bits on the right. The rows are stored in the sequence 12, 11, 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 just as they exist from top to bottom on the physical card. Thus each card is 160 bytes in a file. 

Carl

[Carlos Abramo]

Excellent explanation Carl

All the best 

Carlos 

--
IBM1130 discord channel: https://discord.gg/nrFEMt48mE
---
You received this message because you are subscribed to the Google Groups "IBM1130" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibm1130+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ibm1130/f6044f3c-74c8-4376-86c8-5e3cff223a8dn%40googlegroups.com.

[Jaume Lopez]

Hello Carl,

Yes, I'm the owner of that deck.

So, even if incomplete you haven't it in your repositories? Then I think it would be worth the effort. We would have some starting point for this version for when other incomplete decks appear.

If I understood correctly the SIMH format would only use the lower nibble on the second byte of each column, right? I find that raw representation very simple, but I also see a lack of description for a more future-proof archival.

My only handicap right now would be reading them, as I have no infrastucture for doing it. The deck is in Andorra, so sending it to the US might be costly. The user nicknamed Shadoko (in France) offered to read them, I will ask if I could send them to him later on. I would need to do it manually otherwise.

Thank you very much for all that information!

Jaume

El dia dimarts, 11 de novembre del 2025 a les 18:38:25 UTC+1, Carl Claunch va escriure:

[Bob Flanders]

With a casual search, I found two thing that read card images:

https://github.com/digitaltrails/punchedcardreader

and 

https://www.masswerk.at/cardreader/?home=1

Maybe you can try one of these. But note that the binary columns may not translate into a useful value. For the purpose of the 1130, 

I would translate them into pure 16 bit binary where the 12 punch goes to the high order bit 0 and the 9 punch goes to bit 12. 13-15 would always be zero.

Each row would be two bytes (maybe 4 hex digits), hence Carl's 160 byte notation.

Regards,

Bob

--
IBM1130 discord channel: https://discord.gg/nrFEMt48mE
---
You received this message because you are subscribed to the Google Groups "IBM1130" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibm1130+u...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/ibm1130/fee7400f-9d80-45f9-9b40-069e535e7b1fn%40googlegroups.com.

[John Pierce]

If I remember correctly, a typical V2 m11 or M12 deck, with Fortran and assembler, but no Cobol was just about an entire box of cards. That's like 2,000 cards.

To view this discussion visit https://groups.google.com/d/msgid/ibm1130/CA%2BcnuoZrrPCASNYKDsVDmSk63JWyw5Q8pqidAuynm17NMcN%3Dvw%40mail.gmail.com.

[Jaume Lopez]

Yeah, it is very incomplete... but if I document it and other fragments appear, we could make the full deck in the future.

El dia dimarts, 11 de novembre del 2025 a les 23:03:48 UTC+1, John Pierce va escriure:

[Carl Claunch]

Hi Jaume

So, even if incomplete you haven't it in your repositories? Then I think it would be worth the effort. We would have some starting point for this version for when other incomplete decks appear.

I do not have it, but would certainly keep a copy if you archive it. 

 

If I understood correctly the SIMH format would only use the lower nibble on the second byte of each column, right? I find that raw representation very simple, but I also see a lack of description for a more future-proof archival.

The SIMH format uses the first byte and the top nibble of the second byte to encode all 12 rows, then has a padding nibble of  b0000 for the lower nibble of the second byte. All mention of first and second byte or lower vs upper must be interpreted in light of big endian/little endian realities of the system where the data is stored. 

My only handicap right now would be reading them, as I have no infrastucture for doing it. The deck is in Andorra, so sending it to the US might be costly. The user nicknamed Shadoko (in France) offered to read them, I will ask if I could send them to him later on. I would need to do it manually otherwise.

If the user Shadoko will read them in the lossless SIMH format I mentioned, that might be a way to get your deck digitized. It is important that the user not read it in a format that attempts to convert and store it as ASCII, since that lossy conversion will fail to produce a useful file. 

I would digitize them for you although I am also in the United States thus the postage and shipping time would be a problem. It would be better if you find someone more local to digitize the cards. 

It is not impossible to do this manually with a reasonably small deck. Error checking will be essential to record around 86,400 bits without getting any one of them wrong. 

Regards,

    Carl 

[Al Kossow]

On 11/12/25 4:44 AM, Carl Claunch wrote:

> I do not have it, but would certainly keep a copy if you archive it.

Is it time to push the 1130 archive(s) to bitsavers to have it archived internationally?
The side-effect is the ever-popular Internet Archive scrapes bitsavers, so it gets pushed
there (semi) anonymously.

[Jaume Lopez]

Hello!

For the time being I will make an expedient with the scans of the deck. I have seen, thanks to Mr. Flanders, that python has really good libraries of image processing, so the data could be recovered using software means over the scans rather than by using an actual card reader. The scans could be batch-processed at a later date always that there such images, so my priority will be to take the pictures of the cards as they are and document as much as possible.

Regards,

Jaume

El dia dimecres, 12 de novembre del 2025 a les 14:16:31 UTC+1, Al Kossow va escriure:

[Carl Claunch]

Hi Al

Great idea, lets email to work out the details.

Carl

[Jaume Lopez]

Good morning!

I have talked with the developer of a card reader program and he told me the program is free as in liberty, so I could modify it for my needs.

I will be at home for some days, so I expect to make some good scans. I will have to use some colored paper to generate enough contrast for the holes to be properly identified by the program, but after I modify the program to work within this parameter, it could be a good tool to save the decks.

Have a nice day you all!

Jaume

El dia dijous, 13 de novembre del 2025 a les 14:17:47 UTC+1, Carl Claunch va escriure:

[mwright]

I've been working on web-based UIs for interacting with 1130 devices, including punch-card related ones (e.g., that save/load card decks to files on Linux/Mac) and soon load to 1130 disk emulator.  I was inspired by Bob's 1130 emulator.  I want to be able to read and punch card images on an emulated 1130.  So, I'd be interested in using any data from punch cards.  I use quite a diverse set of AI tools, including for OCR, and for cleaning up noise using diffusion models, so I would be interested in developing a tool that takes images of cards (or printouts) and converting that to data (native, virtual card decks, virtual disk image).  

Bob sent me a link to some scans/photos of listings.  Where can I find a link to scans of cards/decks?  I produce software under MIT license and share the source publicly if anyone is interested.  

(BTW, I'm a retired software engineer that started my career as an IBM CE working mostly on 1130s and 5100s--late 70s early 80s.)

[Jaume Lopez]

Hello mwright,

The tool that obtains the data from pictures already exists. I talked to David, maintainer of the ICL-CES page and he gave me authorization to employ his code.

https://iclces.uk/articles/reading_a_punched_card.html

There are a few things that need to be changed in order for it to become useful for my needs. The first two modifications will be to make it use different background colors and disable the encoding so it works with the raw values directly. From there it is just a matter of improving it.

In any case, I think it's great that you have interfaced those peripherals. I'm in awe!

I will provide a deck of cards soon. I am at home until Tuesday, so I might be able to scan many of them. However I warn you that my deck is incomplete.

Personally, I am more worried about defining my own generic format for storing the cards, as I think the current ones have a serious lack of metadata.

Thank you!

El dia dissabte, 15 de novembre del 2025 a les 16:25:41 UTC+1, mwright va escriure:

[Bob Flanders]

I started a program in https://github.com/semuhphor/CardReader. It reads a card from the phone card by card and the you can have the cards sent to you as JSON. One iteration almost got the card read working, but the card analysis is not quite there. 

IDK anything about using transforms to retrieve data from images... I let AI build it. Maybe this can serve as the start of program?

I don't have time right now to do any work on it.

Regards,

Bob

To view this discussion visit https://groups.google.com/d/msgid/ibm1130/23465c76-f270-4cf8-8ed0-068159a104ban%40googlegroups.com.

[Jaume Lopez]

Hello Bob,

Your program sounds very interesting. I will take a look at it too.

Still, I think we should go with XML rather than JSON due to the ability to define schemas for verification. I know that JSON is much more lightweight, but I think in certain occasions XML has better prestations.

I defined some metadata both for cards and decks. You can take a look at them here. If you think something is missing or wrong, feel free to comment here.

I know this is the 1130 user group, but I think we should try to abstract a little and be more generic in order to gain some format adoption.

In any case, thank you very much!

Jaume

El dia dissabte, 15 de novembre del 2025 a les 19:39:44 UTC+1, Bob Flanders va escriure:

[Bob Flanders]

Cool... XML is good. I've had to deal with XML a lot in my job, and it's cumbersome. I just like json notation but teach his own. Do it dude. 

If you can get it working, please send me a poll request and I'll and I'll apply it back to the original code. Thank you so much.

 Have a great weekend.

Regards, Bob

To view this discussion visit https://groups.google.com/d/msgid/ibm1130/7ef6f84e-bed2-4486-bb5d-179a74942640n%40googlegroups.com.

[Jaume Lopez]

Sorry Bob, I didn't have intention to anger you.

To be fair, I also think that for most applications JSON beats XML by far. However, in my opinion (yes, my subjective thought process) is that for long term storage of the data, that rigidity that is characteristic of XML may be an asset rather than an issue. This is the reason I named it.

I am in process of defining formats both for cards and for decks. Once I figure them out, if I still remember (I have memory issues), I wouldn't mind to modify your apps to give compatibility.

In any case, your programs are great and I have observed your metadata has some similarities with mine.

Sorry for the trouble. I wish you a great weekend too.

Regards,

Jaume

El dia dissabte, 15 de novembre del 2025 a les 20:45:54 UTC+1, Bob Flanders va escriure:

[Bob Flanders]

Hey man,

No problem. I wasn't upset at all. I just was expressing my preference for json. That's all. Please enjoy the code if it's useful and Chuck it if it ain't it was a very quick project just to see if I could begin to get something running. I just don't have time right now to work on it. Thanks for your message. Take care, Bob

To view this discussion visit https://groups.google.com/d/msgid/ibm1130/9eb3f46d-d0b9-4a14-9fd8-0851b4559e63n%40googlegroups.com.

[Jaume Lopez]

I have been given an alternative for the format which I am seriously considering. They told me that it would be far better to employ PDF with embedded data. This way the deck would be joined and there won't be a custom format being defined. I am studying now how to modify the XMP data file - it can be serialized both as a XML and also as a JSON, so it would be great! I am also analyzing what tools could I use to do so programmatically.

By the way, if any of you have some time please take a look at my metadata definition I posted a bit earlier. I am a newcomer when it comes to punched cards while you are veterans. For this reason I would like to ask you to give it a read and comment if you don't understand, would add, remove or modify any of the traits.

I will keep you all informed!

Thank you very much for everything!

El dia diumenge, 16 de novembre del 2025 a les 18:32:39 UTC+1, Bob Flanders va escriure: