Anyone have an SBC-254S bubble memory board they'd like to part with?

[scott baker]

This is not the iSBX-251 multimodule, but a full-size multibus board with between one and four 7110 bubble memories on it, providing between 128KB and 512KB of memory.

I actually got excited when I found one at one of the PLC stores. But they decided they had to test it before shipping it to me, and their technician concluded it was nonfunctional and they refunded the sale. I've reached out to see if they would consider selling it as parts/repair rather than as tested and working, for a discount. Hopefully they didn't toss it in the trash. I like to fix things.

But anyway, if that doesn't pan out, I'm curious whether anyone here has an SBC-254. They're fairly uncommon, but not unobtainable. It would have a happy home in my System 310.

Also, if anyone has the technical manual for the SBC-254 or SBC-264 that would be great. Marc's site has a few pages out of the data book, basically a brochure, but I was unable to find the full manual anywhere.

On a related note, I did manage today to get an iSBX-251 multimodule to work with RMX-86, but as the multimodules are only 128KB, they're not all that exciting. A fully populated SBC-254 would be enough to boot and operate a small RMX86 / RMX286 system.

Scott

[craig andrews]

I suspect Patrick Wong in brookings Oregon has a shelf of them.

Craig

On Aug 2, 2025, at 7:57 PM, scott baker <smb...@gmail.com> wrote:

This is not the iSBX-251 multimodule, but a full-size multibus board with between one and four 7110 bubble memories on it, providing between 128KB and 512KB of memory.

--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/aa66c204-c6d2-4fc8-b154-13e70c0e1a6an%40googlegroups.com.

[scott baker]

Interesting. I don't know Patrick, but a quick google search came up with the link to Northwest Technical (whom I've also never dealt with). Do you have any more background on Patrick or his company? Generally I've stayed away from most of the PLC equipment vendors I've spotted online due to high prices.

Scott

[William Beech]

Scott,

I have dealt with Patrick in the past with excellent results.  He lives by the NDA they must have signed with Intel and believes the hardware descriptions and schematics of the boards he handles can't be shared with us. I have gotten other parts of manuals xeroxed for me.  I have traded boards and bought an Intel 320 power supply from him.

Bill

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/ecf9b491-cbe4-4efb-9ad5-0b69daf91964n%40googlegroups.com.

[scott baker]

Thanks Bill. I saw a phone number at Northwest Technical, but I didn't see an email address. Is telephone the best way to contact him, or does he have an email?

Scott

[Craig Andrews]

As I understand, Patrick worked at Intel years ago. At some point he made an agreement to continue servicing Intel boards and was given much of the proprietary technical support documentation.  I suspect he is the last one in the US (and one of the few in the world) that still services and stocks Intel boards.   

He has shelves and shelves of boards, reference boards, and test fixtures, I would guess a few thousand just in  multibus I

I am going over to the coast next week or two and will probably drop down to visit him. He is just north of the border in a beautiful part of Oregon. 

On Aug 3, 2025, at 11:58 AM, scott baker <smb...@gmail.com> wrote:

Thanks Bill. I saw a phone number at Northwest Technical, but I didn't see an email address. Is telephone the best way to contact him, or does he have an email?

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/3d45b148-59d5-42e8-8280-2fa8f5c56007n%40googlegroups.com.

[Al Kossow]

On 8/3/25 1:32 PM, Craig Andrews wrote:
> As I understand, Patrick worked at Intel years ago. At some point he made an agreement to continue servicing Intel boards and was given much
> of the proprietary technical support documentation.  I suspect he is the last one in the US (and one of the few in the world) that still
> services and stocks Intel boards.

Doesn't sound like there is a sunset clause on his NDA.
Eventually, it won't be a profitable enough business.
Something should be arranged for when that day happens
so it doesn't just get scooped up by bottom feeders like
the scalpers dealing in used VME

[Al Kossow]

On 8/3/25 2:05 PM, Al Kossow wrote:

> Doesn't sound like there is a sunset clause on his NDA.

s/is/isn't/

[Al Kossow]

On 8/2/25 7:57 PM, scott baker wrote:
> This is not the iSBX-251 multimodule, but a full-size multibus board with between one and four 7110 bubble memories on it, providing between
> 128KB and 512KB of memory.

AP-187

fyi
https://web.archive.org/web/20210911002216/https://doc-04-38-docs.googleusercontent.com/docs/securesc/g1g552rb4d6010u7mbnstmtm9jbosh4v/htsg9go71s2l02hdsuqrakc5mieb2g15/1631331750000/03146173954874291632/04405842910372606365Z/0B9rh9tVI0J5mYzVjMmZhOTQtMTE3OS00MTdlLTgwN2MtM2Q4MDIzOWE5ZWY0?nonce=1t7n058145fo8&user=04405842910372606365Z&hash=7dr8ugn6lokvn5lg49p2jn9e7lidleoo

I have some third-party bubble boards I'm trying to find.
Sold a couple on eBay but can't find the listings

[Royce Taft]

I’m not sure if anyone else here has struggled to get a full PDF from one captured in the wayback machine, but I figured I would mention it.

PDFs in the wayback machine seem to just show up as a single page consisting of the first page of the actual PDF. If you click “About this capture” at the upper right, then within the new section that pops up, the URL below “Timestamps” will take you to the full PDF.

If this was already known, feel free to ignore this message. If there’s a more obvious or intuitive way to access the full PDF from a wayback machine page, please let me know.

Royce

Sent from my iPhone

> On Aug 3, 2025, at 17:11, Al Kossow <a...@bitsavers.org> wrote:

> --
> You received this message because you are subscribed to the Google Groups "intel-devsys" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.

> To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/c50ccfa3-9e2d-87cd-d02b-2297810d5790%40bitsavers.org.

[craig andrews]

Royce, thanks for the tip.

Craig

> On Aug 3, 2025, at 7:07 PM, Royce Taft <royc...@gmail.com> wrote:
>
> I’m not sure if anyone else here has struggled to get a full PDF from one captured in the wayback machine, but I figured I would mention it.

> To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/390E9C4C-C738-49B9-996A-A61CFF5573E8%40gmail.com.

[Al Kossow]

On 8/3/25 7:07 PM, Royce Taft wrote:
> I’m not sure if anyone else here has struggled to get a full PDF from one captured in the wayback machine, but I figured I would mention it.

I cleaned up the pdf and it is now on bitsavers under http://bitsavers.org/components/intel/bubbleMemory

[Vale, Martyn]

Hi Scott,

 

Patrick’s email is patric...@gmail.com

 

Tell him you’re with intel-devsys.

 

He usually very helpful, but he does deal with these sort of emails out of hours, so you may need to wait a little while for a reply.

 

Bye

Martyn.

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/3d45b148-59d5-42e8-8280-2fa8f5c56007n%40googlegroups.com.

[Al Kossow]

On 8/2/25 7:57 PM, scott baker wrote:
> This is not the iSBX-251 multimodule, but a full-size multibus board with between one and four 7110 bubble memories on it, providing between
> 128KB and 512KB of memory.

Found them, I have five boards.
http://bitsavers.org/components/intel/bubbleMemory/iSBC254S_f.jpg

[Al Kossow]

programming and jumpering information would be good to try to get

[scott baker]

Nice! :D are you interested in selling one?

I have reached out to Patrick. He does have some, but we haven't talked cost yet, and I don't really have anything I can "trade".

The closest thing I've found to documentation, other than the couple of pages on Marc's website, is this Master's Thesis: https://upload.wikimedia.org/wikipedia/commons/6/6e/Utilization_of_a_Bubble_Memory_System_as_a_microcomputer_disk_resource._%28IA_utilizationofbub00thei%29.pdf. 

Scott

[Al Kossow]

On 8/4/25 11:25 AM, scott baker wrote:
> Nice! :D are you interested in selling one?
>

sure, I have no need for them. they will be cheaper than Patrick

[andre...@gmail.com]

Nice, I may be interested in getting in on this deal too

craig

-----Original Message-----
From: intel-...@googlegroups.com <intel-...@googlegroups.com> On Behalf Of Al Kossow
Sent: Monday, August 4, 2025 11:45 AM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys Anyone have an SBC-254S bubble memory board they'd like to part with?

--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/7c868fd8-b12c-3a0e-5ce9-f0633ae212bf%40bitsavers.org.

[William Beech]

Scott,

I don't have the email address handy. 

Sorry.

Bill

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/3d45b148-59d5-42e8-8280-2fa8f5c56007n%40googlegroups.com.

[William Beech]

Craig,

You might try to find out what we need to get from Intel to allow him to sell us complete old Intel manuals and schematics.  As I said, I have gotten partial manuals but not the hardware sections nor the schematics.  Also if we could get copies of the old software.  That would be more difficult but I would be willing to pay for some of the things I need!

Bill

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/AA9A5BB4-12EB-4417-8A82-394EFCACAB0E%40gmail.com.

[William Beech]

True!

[andre...@gmail.com]

Bill

 

I will bring this up when I go back down to see him next week.  But we have talked about this extensively and the answer has always been the same.  Unfortunately, I no longer know anybody that works at intel and, quite frankly, they are going through their own problems at the moment.

 

Craig

 

From: intel-...@googlegroups.com <intel-...@googlegroups.com> On Behalf Of William Beech
Sent: Monday, August 4, 2025 3:30 PM
To: intel-...@googlegroups.com
Subject: Re: intel-devsys Anyone have an SBC-254S bubble memory board they'd like to part with?

 

Craig,

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/1ae7d110-47ea-4c8b-8c7c-ac1f4a51084e%40nj7p.info.

[William Beech]

Al,

Again, thanks for all your work archiving all this old material.  I know
what a task it really is and I, for one, appreciate you doing it!

Bill

[William Beech]

Craig,

All true!  But we can only continue to try.  Does the NDA have an expire date?  What is really being protected here?

Bill

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/217f01dc0590%2402cab1e0%24086015a0%24%40gmail.com.

[scott baker]

Thanks to Al, I have a solution for my iSBC-254S boards! Thanks Al!

Technical documentaiton would be really nice -- I wonder if Patrick would be willing to sell a copy of the hardware reference manuals for some of these more obscure boards? Presumably the documentaiton would have been for sale at some point, even having been supplied with the boards. If someone owned a board that became separated from its manual, there must have been a process for obtaining a replacement manual for a nominal cost. I can see how Patrick might not want to sell the documentation if he had reason to suspect our goal was to immediately scan it in and distribute it in violation of copyright. I wonder if there's a library-like solution where we could "check in" the document and then one person at a time could "borrow" it? For example, archive.org offers borrowable texts. Archive.org unfortunately seems to have some rules around this requiring borrowable texts to be sponsored by a library. (anyone wanna start a new library?)

The problem with large corporations is that they're large. A casual engineer at Intel wouldn't know where to start to find a decider who could decide on releasing decades-old documentation from organizational hierarchies that may not exist anymore. Then even if you somehow found a decider, then the process would start, which might involve legal review. Do they know the documentation is even theirs to release? Maybe they sold the rights? Would partners be impacted? Especially given the current economic reality, I can't see anyone undertaking that without a clear business advantage in doing so.

Scott

[Jon Hales]

Hi Scott

I don't have an iSBC-254/264. Hopefully you will be able to acquire the one from PLC Stores.

Given your focus on iRMX-86, you may be interested in the Intel 80130, where iRMX system calls were incorporated in ROM in a chip with a range of other functions. The 188/48 'Advanced Communicating Computer', is a Multibus 1 board (PBA 175213-001) with an 80130-6. I'm not aware of other Intel boards to which the 80130 was fitted.

There doesn't appear to be any documentation of the 188/48 beyond 230890-002, which is a 10-page outline (dated September 1984) probably from an Intel Catalog/Databook. The references to the 80130 in that outline refer to interrupt and timer features.

The 80130 was developed during enthusiasm for 'Software in Silicon' at Intel. One source of information about it is AP-130 (also 210295-001, March 1982) 'Using Operating System Firmware Components to Simplify Hardware and Software Design" (in Mark Ogden's collection - More detailed documentation is in 210811, mentioned below).

In June 2022 I posted a comment on Intel-Devsys about the 80130 (iRMX in 16Kb of ROM) and 80150 (the same 'host' chip with CP/M-86 in ROM). A response from Eric Smith stated that the 80130 was used for the three timers, interrupt controller, baud rate generator it contained and may have been issued without the iRMX-86 ROM code. When Intel no longer had stock, a corresponding chip was issued by NEC based on a gate array (product ID 80130-A, also uPD65044C-143 - photo on CPUShack museum). Eric Smith thought it unlikely this included the iRMX system calls.

There's a Wikipedia page on the 80130:

The Intel 80130, referred to as an "Operating System Processor," was developed as a support chip for the 8086/8088 processors and the Intel iRMX86 operating system. Intel referred to the chip as "software in silicon".[1]: 8–1 

Overview

It contained 16-KB of ROM containing the code for 35 of the iRMX 86 system calls, an interrupt controller similar to the 8259A, timing circuits, a baud generator circuit, and all the necessary circuitry for bus buffering and control.[2]

It was not used in the IBM/PC, and as such, is a less prominent chip.

Architecture

The 80130 uses an object-oriented architecture, with objects representing tasks, jobs, mailboxes, regions, and segments.[1]: 8–2  These objects are acted upon by primitives, which are invokable as PL/M-86 procedures.[1]: 8–3 

Notes

 IAPX 86, 88, 186, and 188 user's manual : programmer's reference. Intel Corporation. Santa Clara, CA. 1983. ISBN 0835930351. OCLC 11091251. (* Note)

 Microcomputer Systems: The 8086/8088 Family by Yu-cheng Liu and Glenn A. Gibson

(* Note: this manual is in Mark Ogden's collection as 210811-001. Page i: "Chapter 8 describes the 80130 Operating System Firmware (OSF) Component" (109 pages).

Best regards

Jon

--

[Jon Hales]

Footnotes to previous message:

Intel document 210912-001 iAPX 86/88, 186/188 User's Manual Hardware Reference includes Chapter 5 on the 80130 (15 pages) including details of interrupts and timers.

The Intel 80150 (CP/M-86)

Intel document 210814-001 "Silicon Software Components" (Also Tech Paper TP-1, October 1982, 8 pages. Presented at Mini-Micro-82) outlines the 80150 - with CP/M-86 v1.1 in 16k of ROM, of which 8k was the BIOS by Intel, with 8k containing CCP and BDOS based on code by DRI. "The standard CP/M-86 CCP has been enhanced to allow the user to add more built-in commands to further customise his CP/M-86 system"). I'm attaching a copy of the PDF (6Mb). It appears generally accepted that the 80150 was produced in very limited quantities, if it was ever released as a product. It wasn't given the same attention as iRMX in Intel's documentation.

Intel document 210705-003 (12 pages) also details the 80150. This was published in 230843-002 "Microsystem Components Handbook - Microprocessors and Peripherals Volume 1" (1985) available from Mark Ogden's collection (pages 3-220 to 231). The 80130 is also documented in the same volume.

Jon

[Jon Hales]

Hi Scott

I purchased a few items from Patrick - he was straightforward to deal with. He offers discounts for ex-Intel employees and for 'enthusiasts'. I don't know whether he has a fixed formula. It may help to mention your dissemination of information about Intel's systems. I don't know whether he is aware of intel-devsys, but it seems a good time to inform him - I'm not sure what the current count of those signed-up would be - and I believe the discussions allow anyone to find a posting by searching for a part or manual.

The attachment clarifies the situation.

His business is primarily servicing Intel parts from 'legacy' systems and presumably replacing broken components with new and tested equivalents. There are legacy systems in surprising places (in 2025). The sites presumably installed Intel equipment some decades ago because of Intel's reputation.

It seems plausible that if he has a quantity of an item and no-one else has asked about it, he may be wiling to offer a larger discount. But that could also mean he has to search among boxes, which he may or may not be willing to do.

Best regards

Jon

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/3d45b148-59d5-42e8-8280-2fa8f5c56007n%40googlegroups.com.

[Royce Taft]

To what level of “working at Intel” do you think is necessary to gain traction on this effort?

I have a buddy who works for them keeping their tooling running at their fab in Hillsboro. 

Royce

Sent from my iPhone

On Aug 4, 2025, at 15:35, andre...@gmail.com wrote:



To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/217f01dc0590%2402cab1e0%24086015a0%24%40gmail.com.

[craig andrews]

As I perceive the situation, since he has thousands of multibus boards on the shelf, Patrick wants to make sure he is not undercut by someone buying boards at a good deal as a “hobbyist “  and then re-selling them for cheap.  I would do the same if I were in his shoes.

Craig

On Aug 5, 2025, at 3:30 AM, Royce Taft <royc...@gmail.com> wrote:

To what level of “working at Intel” do you think is necessary to gain traction on this effort?

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/E365D38A-9AE9-4E11-A618-7D335A02F4ED%40gmail.com.

[William Beech]

Craig,

This is a problem for any business that sells a product.  This gets down to his relationship with us, a small user community. We are trying to enjoy machines that we could not have afforded when new. 

I believe if we could get full copies of some of the missing manuals, we could do more with what we have.  But there is either a reluctance on Patrick's part or it is part of an NDA. I do not know which.  Intel routinely sold these complete manuals to anyone who would pay for them in the past. What changed in the transition from Intel to Patrick?

I have found Patrick much more helpful than any of the other competitors in this business.  I greatly appreciate all the help he has given me!

Bill

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/5F58677B-41A0-42C0-BF53-BAD5CC13B17B%40gmail.com.

[scott baker]

Thanks to Al, I have my isbc-254S boards!

It took me a while to determine the jumpers for address and interrupt, but once I did and adjusted the address (boards came set to 0x820, iRMX was configured for 0x880) the board came to life. A short demo is below. I am experiencing some corruption issues copying larger files to the bubble -- I'll have to take some time to troubleshoot that and see what's going on. 

- attachdevice b0 as :bub:
:bub:, volume is not a NAMED volume

- diskverify :bub:
iRMX 86 Disk Verify Utility, V3.0
Copyright 1981, 1982, 1984 Intel Corporation
*disk
Device name = b0            
   Physical disk
      Device gran = 0100
       Block size = 0100
     No of blocks = 00000800
      Volume size = 00080000

- format :bub:scottb
volume (scottb) will be formatted as a NAMED volume
    granularity   =   256        map start =        992
    interleave    =     5
    files         =   200
    extensionsize =     3
    volume size   =   512 K


volume formatted

- dir :bub: e
25 JUL 84  23:33:23
DIRECTORY OF  :bub:  ON VOLUME scottb
                                                 GRAN
  NAME         AT   ACC      BLKS     LENGTH   VOL FIL OWNER           LAST MOD

         0  FILES           0  BLKS             0  BYTES

       200  FILES       1,958  BLKS       501,248  BYTES FREE

- copy hello.c to :bub:
hello.c copied TO :bub:hello.c

- copy :bub:hello.c to :co:
#include <stdio.h>
main() {
    printf("Hello, World!\n");
    return 0;
}
:bub:hello.c copied TO :co:

[scott baker]

I've started documenting the jumper settings here: https://github.com/sbelectronics/multibus/blob/master/isbc-254s/NOTES.md . I have most of them figured out.

I also resolved my "corruption" problem. Boards in the 310 use parallel priority resolution, where the topmost slot in the case has highest priority and the bottommost slot has least priority. The iSBC-215 winchester controller was in Slot 7 and the iSBC-254S was in Slot 4. Both of these boards can bus master. Swapping the two boards resolved the issue. I suspect what was happening was that while doing a COPY, the disk and the bubble are simultaneously active, and the bubble was occasionally getting starved for bus access during the writes.

Anyhow, I'm now 100% up and running with the 254S boards!

I also snagged a reasonably priced SBC-264 from eBay one late night. I didn't pay close attention to the pictures, and I thought it was just a newer variant of the 254. However, upon receiving the 264 today, I can see that it has 7114 bubbles (4 megabit) compared to the 7110 (1 megabit) on the 754. The 264 is a cool find, but unfortunately no drivers in the iRMX releases we have available to us.

Scott

[scott baker]

Note: renamed the file from NOTES.md to README.md -- https://github.com/sbelectronics/multibus/blob/master/isbc-254s/README.md

Scott

[scott baker]

I started to list my bubble memory projects in one place at https://www.smbaker.com/magnetic-bubble-memory-mega-post

Recent acquisions include an iSBX-264 (which uses the 4-megabit 7114 bubble memorys) and the Memtech Bubble Memory Cassette and its associated iSBX-258 multimodule. The memory cassette is pretty cool as it works just like the iSBX-251 multimodule.

Scott

[Al Kossow]

I have on 1Mb bubble memory board left if anyone is still looking for one.
$175 including shipping

[Vale, Martyn]

Hi Al,

I'd be interested in the board if it's still available.

I'm in the UK though so obviously I'd pay for the extra shipping costs assuming you were happy to ship to the UK of course.

Thanks
Martyn.

-----Original Message-----
From: intel-...@googlegroups.com <intel-...@googlegroups.com> On Behalf Of Al Kossow
Sent: 07 September 2025 14:57
To: intel-...@googlegroups.com
Subject: Re: intel-devsys Anyone have an SBC-254S bubble memory board they'd like to part with?

I have on 1Mb bubble memory board left if anyone is still looking for one.
$175 including shipping

--
You received this message because you are subscribed to the Google Groups "intel-devsys" group.
To unsubscribe from this group and stop receiving emails from it, send an email to intel-devsys...@googlegroups.com.

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/8e28d6ae-800d-f8e4-0136-2e514820356d%40bitsavers.org.