A mystery Intel kit

[Jon Hales]

Hi All

I'm attaching two photos of an Intel product with the code 146007-003. It's a card with:

- a set of 10 EPROMs (146396-001 to 146396-010), a couple of D8226 ICs, four 7438N ICs, a 74LS03 on a small PCB and 8 resistor packs.

- four cables.

These items are sealed with a plastic layer.

I'm reluctant to break the seal to read the EPROMs in the search for clues. The ten EPROMs are 2764s.

IF the code 146007-003 relates to Intel's numbering of publications (which is unknown, perhaps unlikely) then the date would be around 1983. The labels of the EPROMs also refer to 1983.

Any suggestions about where this set might have been used would be welcome.

Best regards

Jon

[Mark Gent]

Hi,

For what it's worth the 50 pin IDC header reminds me of SCSI-1/-2, and there is clearly an 8 bit bus running from it to a 25pin connector, which was also used for SCSI-1.  However, the pins on the 50 pin connector are off by 1 as in the SCSI standard the data line are the first 8 on a single-ended bus.

However, it could fit a SCSI differential bus as per https://www.scribd.com/document/236828446/SCSI-Internal-Pinout-Diagram-Pinouts.  Maybe an adapter kit for SCSI differential to SCSI 25pin single ended D-type with associated software update?

Best regards,

Mark

---

From: intel-...@googlegroups.com <intel-...@googlegroups.com> on behalf of Jon Hales <jonh...@gmail.com>
Sent: Tuesday, 09 June 2026 10:45:22
To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: intel-devsys A mystery Intel kit

 

--
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/CANPXPJkXuDw%3DSaLKXBEdEuceULdKozgnL01ZuiW1SMC2jb7JMg%40mail.gmail.com.

[Mark Gent]

Which all seemed fairly plausible until I noticed that in my enthusiasm I failed to notice that pin1 is clearly on the other side of the connector, helpfully indicated with a pointer, so back to the drawing board...

---

From: Mark Gent <mg11...@gmail.com>
Sent: Tuesday, 09 June 2026 11:42:01
To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: Re: intel-devsys A mystery Intel kit

 

[roger arrick]

It looks like the cables go like this:

• 26 pin card edge from a PCB to a cable with a DB25
• DB25 to DB25 external cable
• DB25 to 50 pin male header
• 50 pin male header to 50 pin female card edge.

Then EPROMS and terminating resistor packs and a daughter board to update an existing board.

I vote this is some sort of external disk drive upgrade, possibly even a SASI HD, but I wouldn't rule out some sort of unsupported 8" floppy format.

I also vote NOT to take it out of the packaging.  It's a pristine piece of history, whatever it is.  It's likely rare and the only one in existence.

--  Roger Arrick -- Tyler, Texas, USA -- Ro...@Arrick.com --

---

From: intel-...@googlegroups.com <intel-...@googlegroups.com> on behalf of Jon Hales <jonh...@gmail.com>

Sent: Tuesday, June 9, 2026 4:45 AM

To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: intel-devsys A mystery Intel kit

--

[William Beech]

Guys,

In my dealings with Intel (1985-1990) I never saw them field a SCSI product in the Intel 310/320 family.  The presence of 7438s and the 50-pin cable make me think 8-inch disk drive.  Ten each 2764s and 1983 looks like some mod kit for a system like a 310/320.  Might be for the 8-inch SMB drives that were used in these systems?

Bill 

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/DM6PR14MB3449B008F01702FEBDD806B0D81D2%40DM6PR14MB3449.namprd14.prod.outlook.com.

[Mark Gent]

Thing is that the iSBC 220 uses a 60 pin cable for the SMD drive, and a 2^nd cable for the data channel, so that's not it either. 

Another reason it made me think of SCSI is because the 186/03 board has the option to turn the parallel port into a SCSI port, all detailed in the tech ref manual, which has always intrigued me.

Not exactly a good match as it doesn't have sockets for all those EPROMs, but that parallel port does have a 50 pin connector.

Maybe there are more boards with a similar option.

Only  other 50 pin option is indeed an 8" fdd, but it's not clear to me how that makes to a DB25 connector, while SCSI peripherals did use to use that.  And the adapter board in the kit does seem to show a parallel bus.

In any event I totally agree with leaving the kit intact in its plastic wrapping, would be a shame to open that.

-Mark

---

From: intel-...@googlegroups.com <intel-...@googlegroups.com> on behalf of William Beech <nj...@nj7p.org>
Sent: Tuesday, 09 June 2026 18:29:47
To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: Re: intel-devsys A mystery Intel kit

 

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/6c187316-105a-408f-95cb-9f2c75db6b2b%40nj7p.org.

[Herbert Johnson]

What Intel product of the early 80's would hold *ten* ROMS? and small
2764's at that? Answer: a dedicated ROM Multibus board.

https://www.ebay.com/itm/256921852642

Intel SBC 428 PBA 146021-001 MULTIBUS I SBC428 PB 146022-001
instrumentl (1027)
eBay item number:256921852642

10 X 2764 = 80K bytes of code. Only an 8088 or later would address that
much code contiguously.

Note the two ROMS present on the eBay board: "146009-001". Numbers in
the ballpark!

Distracting clues:

Something about this pile of ROMS at this date, suggests to me
iRMX-86/88. What 1980-ish product of Intel included a SBC428 ROM board
and ran iRMX-86? Intel docs say iRMX-86 was ROMable. Intel's list of
manuals include 146XXX numbered manuals for ... iRMX86.

The cables include a 50-pin flat cable with IDC *edge connectors*. What
uses a 50-pin edge connector? An eight-inch floppy drive! But why a PC
board 50-pin IDC to DB25 connector?

There's another cable, with some 26/25 pin connectors. More likely
serial or parallel.

My guess? This is an upgrade kit for some kind of 8086/88 software
debugger, or some kind of iRMX-86/88 development system. Maybe it turned
a one-floppy-drive system into a two-floppy system?

That's all I've got. - regards Herb

On 6/9/2026 5:45 AM, Jon Hales wrote:
> Hi All
>
> I'm attaching two photos of an Intel product with the code 146007-003.
> It's a card with:

> - a set of 10 [2764] EPROMs (146396-001 to 146396-010), a couple of D8226 ICs,

> four 7438N ICs, a 74LS03 on a small PCB and 8 resistor packs.
> - four cables.
>
> These items are sealed with a plastic layer.
>
> I'm reluctant to break the seal to read the EPROMs in the search for
> clues. The ten EPROMs are 2764s.
>
> IF the code 146007-003 relates to Intel's numbering of publications
> (which is unknown, perhaps unlikely) then the date would be around 1983.
> The labels of the EPROMs also refer to 1983.
>
> Any suggestions about where this set might have been used would be welcome.
>
> Best regards
>
> Jon
>

Herb Johnson, New Jersey USA
http://www.retrotechnology.com or .net
preserve and restore 1970's personal computing
email: hjohnson @ retrotechnology dot com
or try later at herbjohnson @ comcast dot net

[scott baker]

The only way you're going to know for sure is to cut the shrinkwrap and read the ROMs... :)

I understand the allure of a new unopened package that has been unopened for decades (I own a couple unbuilt heathkits), but if you read them an re-shrinkwrap them, then there's no practical difference. This probably isn't a high value collectable that someone would pay a large premium for it to remain unopened. I say do it. It's easy for me to say though because I don't own it.

Scott

[Herbert Johnson]

I'm inclined to agree with Scott, but there's no hurry, except as follows.

One consideration is that 1983 antistat foam is normally likely to be
either powder or goo. But it's possible that the wrap reduced the
oxidation or breakdown of whatever they made that stuff from in 1983.
Most ancient antistat has had decades of air (thus moisture) exposure.

YOu might feel the foam as wrapped to get some idea of its consistency.
A very close inspection of the pins of the ROMS and chips is in order.
If the foam has broken down, the pins may have a sticky goo on them. In
that case, un-wrapping plus cleaning plus repacking in modern antistat
would seem to be a good idea.

But if it were really "goo", you'd know that from handling. It's
probably just dry.

Information research I did years ago on antistate, suggested that the
filler material was burnt sugar (to make conductive carbon black), and
so it was hydrophilic sugar that contributed to the decay. Plastics
become depolymerized into that power or that goo. It may also be,
silver or tin on the pins is catalytic. I'm not an organic chemis, just
a schooled engineer.

Regards Herb

> http://www.retrotechnology.com <http://www.retrotechnology.com> or .net

> preserve and restore 1970's personal computing
> email: hjohnson @ retrotechnology dot com
> or try later at herbjohnson @ comcast dot net
>

> --
> 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

> <mailto:intel-devsys...@googlegroups.com>.

> To view this discussion visit

> https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com <https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--

[Mark Gent]

Having looked through the various databooks (I just *had* to know what it was) I'm now certain that this is part of one of the iSBC-957 series of debug packages.

The best indicator that I could find so far is a photo of the iSBC-957A package on page 4-31 of the 1981 Intel Systems Data Catalog (also attached).  It only has 4 roms there but the rest matches.

So it's likely a 957B or later.

The description also details how the package can transfer data over a parallel connection using the parallel port on Multibus boards, which normally have 50 connections either as an edge connector or an IDC-compatible connector.

And of course the number of lines on the PCB, 8 bi-directional data lines and 3 control lines, is all you need for that (so no, nothing close to SCSI).

What it's missing is the accompanying manual and floppy disks.

Sadly detailed information on the 957 packages seems to be scarce.

---

From: intel-...@googlegroups.com <intel-...@googlegroups.com> on behalf of Herbert Johnson <hjoh...@retrotechnology.com>
Sent: Tuesday, 09 June 2026 22:35:29

To: intel-...@googlegroups.com <intel-...@googlegroups.com>
Subject: Re: intel-devsys A mystery Intel kit

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/6b94d885-5c59-46be-9d0d-72341c838f7c%40retrotechnology.com.

[William Beech]

Herb,

Yes, the Intel order numbers for software releases were in that ballpark.

These would fit on the iSBC-464 board,  as well.

Bill

[Herbert Johnson]

143979-002 1 20.00 iSBC 957B- iAPX 86/88 User's Guide
Intel list of documents - good luck finding this.

http://bitsavers.informatik.uni-stuttgart.de/components/intel/80186/210911-001_iAPX86_88_186_188_Programmers_Reference_1983.pdf

document page 2-13 (PDF page 30) says:

> 2.9 THE 80130 OPERATING SYSTEM
> FIRMWARE (OSF)
> The 80130 firmware (software in silicon) is, in con-
> junction with the 8086,88 or 80186,188 CPUs, the
> nucleus of a real-time, high-performance multitask-
> ing operating system. The 80130 adds task
> management, interrupt management, message
> passing, synchronization and memory allocation
> capabilities to the CPU. A block diagram of the OSF
> is shown in Figure 2-19.

Chapter 8 of the document is
THE 80130 OPERATING SYSTEM
FIRMWARE COMPONENT

document page 8-109 (PDF page 548) says:

> THE 80130 OPERATING SYSTEM FIRMWARE COMPONENT
> 8.11 ADDING iRMX 86 FEATURES
> TO THE OSP
> There are two ways to add functionality to the OSP.
> One way, discussed in Section 8.9, is to create OS
> extensions. The other way is adding iRMX 86
> system calls (and possibly object types as well) to the
> OSP.
> The iRMX 86 Operating System is a functional su-
> perset of the OSP and consists of the following parts:

> Human Interface: The Human Interface is an inter-
> active interface and an application system. It gives
> the operator the ability to invoke an application pro-
> gram from a terminal.
> Supplied with the Human Interface are commands
> that perform the following operations:
> • Creating, copying, renaming, and deleting
> files
> • Loading and starting application programs
> • Formatting and verifying device volumes
> • Backing up and restoring files on devices
> • Reading commands from a file, rather than
> from a terminal
> • Communicating with the iSBC 957B package <----
> to debug programs and to copy files to and
> from an Intel development system

> The Human Interface also provides a number of
> system calls that applications programs can invoke to
> utilize Human Interface services.

I have an impression, that Intel developed all kinds of real-time OS's
in rapid succession. The 80130 looks to me some kind of masked ROM plus
real-time support signaling and timing in an 8086-type 40 pin package.

Maybe, this set of ROMs is an earlier Intel effort that put that OS
support and code in UVROMS? so it could be added-to an earlier 8086 or
8088 development system? And the "957B" was a combination of the
earlier-rev cabling plus several ROMs for a then-available Multibus PROM
card as I suggested earlier?

I dunno ... but once the PROMs are dumped, it should bring some clarity.
There's likely all kinds of clear-ASCII that will correlate to some set
of Intel software.

Regards Herb

On 6/9/2026 6:36 PM, Mark Gent wrote:
> Having looked through the various databooks (I just *had* to know what
> it was) I'm now certain that this is part of one of the iSBC-957 series
> of debug packages.
> The best indicator that I could find so far is a photo of the iSBC-957A
> package on page 4-31 of the 1981 Intel Systems Data Catalog (also
> attached).  It only has 4 roms there but the rest matches.
> So it's likely a 957B or later.
>
> The description also details how the package can transfer data over a
> parallel connection using the parallel port on Multibus boards, which
> normally have 50 connections either as an edge connector or an
> IDC-compatible connector.
> And of course the number of lines on the PCB, 8 bi-directional data
> lines and 3 control lines, is all you need for that (so no, nothing
> close to SCSI).
>
> What it's missing is the accompanying manual and floppy disks.
>
> Sadly detailed information on the 957 packages seems to be scarce.
>
>

> ------------------------------------------------------------------------
> *From:* intel-...@googlegroups.com <intel-...@googlegroups.com> on

> behalf of Herbert Johnson <hjoh...@retrotechnology.com>

> *Sent:* Tuesday, 09 June 2026 22:35:29
> *To:* intel-...@googlegroups.com <intel-...@googlegroups.com>
> *Subject:* Re: intel-devsys A mystery Intel kit

>> https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com <https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com> <https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/intel-devsys/9f31f01b-a881-4ec6-891f-56773417c724n%40googlegroups.com?utm_medium=email&utm_source=footer>>.

>
> --
> Herb Johnson, New Jersey USA

> http://www.retrotechnology.com <http://www.retrotechnology.com> or .net

> preserve and restore 1970's personal computing
> email: hjohnson @ retrotechnology dot com
> or try later at herbjohnson @ comcast dot net
>
> --
> 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/6b94d885-5c59-46be-9d0d-72341c838f7c%40retrotechnology.com <https://groups.google.com/d/msgid/intel-devsys/6b94d885-5c59-46be-9d0d-72341c838f7c%40retrotechnology.com>.

>
> --
> 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
> <mailto:intel-devsys...@googlegroups.com>.
> To view this discussion visit

> https://groups.google.com/d/msgid/intel-devsys/VI0PR07MB119393A06E6B64A0CFA944A98F41D2%40VI0PR07MB11939.eurprd07.prod.outlook.com <https://groups.google.com/d/msgid/intel-devsys/VI0PR07MB119393A06E6B64A0CFA944A98F41D2%40VI0PR07MB11939.eurprd07.prod.outlook.com?utm_medium=email&utm_source=footer>.

[Jon Hales]

Hi All

Many thanks for the responses to my message.

I think Mark Gent is very close to the answer. 

Application Note AP-43 is The iSBC 957 'Execution Vehicle for Executing 8086 Program Code'. (November 1978).

Manual 9800743A has the title iSBC 957 Intellec-iSBC86-12 Interface and Execution Package User's Guide' (September 1978).

Given their date, these illustrate the connections to the MDS-800, the Series-II 210 (with disk drives) and the Series-II 220,230 (with at least 32Kb of memory). The 4 EPROMs were 2716s.

The components of that kit were very similar to the 'Mystery Kit' I showed in photos. An odd feature of the illustrations in the documents is that the iSBC86-12 is depicted outside the MDS chassis ('floating nearby'), with the cables connecting the serial and parallel interfaces connecting to the sockets on the rear panel of the MDS. The User's Guide states: "The iSBC 86/12 Single Board Computer should be installed in a chassis other than that of the Intellec System (my emphasis).

The Resident Processor Board (RPB-86, based on the 86/12) is equally concerned with allowing a 1977-78 MDS-800 or Series II to act as host for the development of 8086 applications, and in this case it would be fitted internally with no cabling.

It can be imagined that in 1983 there was continued interest by owners of 'Intellec' systems in developing for the 8086. By that time, the RPC (based on the iSBC 86/30) would have been available and the naming of the upgraded system as 'Series III' was presumably quite standard. It's not clear who would have had a need for the '1983 kit'.

Herb: Thank you for your comments about anti-static foam. The foam pad rebounds almost instantly from thumb pressure.

Open or leave? My sense is that there's little to be gained by reading the EPROMs. 

Best regards

Jon

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/d9e80e05-1f65-4ea3-8734-d3c1b0fbbe98%40retrotechnology.com.

[Mark Gent]

Hi Jon and all,

First of all, personally I'd leave it and let the next owner decide what to do with the package, as I'm assuming that the next owner will be on this mailing list, so nothing will be lost (I myself would only open it if I could do it without leaving visible marks on the front, which I think might be possible when working very carefully from the back).

Jon, I see you've also found the documentation for the 957A kit.

Incidentally the 4 roms in there are 4 x 2K roms specifically for the 86/12A board, which means that the space requirements for the 957 kit is 8K.
All 10 roms in the kit you have are 8K in size.  That totals 80K if simultaneously inserted, which would be sufficient with the Cadillac configuration of RMX-I and then some.
So I'm sure that the roms are *not* meant to be inserted together but instead that each is to support a different target multibus board, which makes far more sense, even if the numbering system is a little unorthodox  compared to other board packages from Intel.

However the only way to be easily sure is to somehow get the fabled "iSBC 957B- iAPX 86/88 User's Guide", which is referenced by all documentation but sadly missing.

The 957B kit is totally independent of RMX I'm afraid.  It was apparently possible to use it for debugging RMX but I've never used it before as, well, we had in-circuit emulators and full RMX development kits.  Creating an RMX image from scratch is great fun BTW and you can easily create one with support for a serial port and a CLI, so you're hardly working blind.

With regards to the 80130 iOSP it's just a ROM containing the RMX-I kernel, a timer and an interrupt controller.  All 188/48 boards came with them but without the necessary support package to take advantage of it so we ended up inserting EPROMs on all boards with later version of RMX-I on them (*love* RMX, still do, I learned multithreading on it while the rest of the world was still using DOS).  I got the impression at the time that they were not a commercial success and that Intel was just using them up, and indeed later boards did not have them anymore.

The 80130 is still readily available, heck, I bought one fairly recently on eBay.  I've never seen an 80150, I presume they were announced but never made.

For those interested, I managed not too long ago to proudly get a copy of the "iOSP86 Support Reference Manual", which is in loose leaf format.  It's on my list to scan but it being loose leaf I should be able to do that soon if people are keen.  It's an inch thick though and I'm afraid that it didn't come with the necessary accompanying floppy disks so sadly it's still not possible to easily use the 80130 chip.

Best regards,

Mark G

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/CANPXPJ%3DM-cTaPePzES89rTFAhDL%3D9iEkTrXrzq%3D_p7PaCuR5pA%40mail.gmail.com.

[Jon Hales]

Hi Mark (cc Others)

Many thanks for your observations. And thank you for identifying the iSBC-957A kit as a predecessor of the 'Mystery Kit'. As I said, I find it hard to identify who the later version would have been aimed at - possibly for MDS-800 owners who were still using the system in 1983?

The 8k provided with the iSBC-957A appears to have provided little more than a monitor on the iSBC 86/12. The 80Kb of code in the later edition presumably made life a bit easier.

I have an iSBC 088/48 (from one of my Series 310) with an 80130. And I also purchased an 80130 on eBay in UK at a reasonable price - simply out of curiosity. My understanding is that it contains 'iRMX routines', presumably in order that a programmer could reference these routines rather than expanding the compiled code. You would be well-placed to correct me if that's not an accurate description.

I have also searched for an 80150, the capability of which seems rather more promising. By this I mean that it sounds like a full version of CP/M-86 in 16kb.

In 2022, there was a discussion on the VCF Forum about the fact that the CP/M-86 v1.1 source files assume the target is the Intel iSBC 86-12 (files are at http://www.cpm.z80.de/source.html). To quote from one of the comments in that thread:

 The BIOS.A86 claims to be for the SBC

;*********************************************
;* *
;* Basic Input/Output System (BIOS) for *
;* CP/M-86 Configured for iSBC 86/12 with *
;* the iSBC 204 Floppy Disk Controller *
;* *
;*********************************************

The link to the discussion, (including my reference to a teledisk image of an 8-inch boot disk for the iSBC 86-12) is: https://forum.vcfed.org/index.php?threads/cp-m86-for-ibm-pc-at-source-build-kit.1238685/

Best regards

Jon

To view this discussion visit https://groups.google.com/d/msgid/intel-devsys/61229862-62eb-4aaf-94df-ac684fa871a4%40gmail.com.

[Herbert Johnson]

> An odd feature of the illustrations in the documents is that the iSBC86-12 is depicted outside the MDS chassis.

Not odd at all. That's called "embedded controller development". The MDS
system is not the controller under test. What's under test is another
Multibus system or at least a Multibus CPU board. It's only odd in
retrospect, after embedded controllers became disk-based, you could
debug on-the-system, and so on. Wasn't always so.

The cited 1978 "execution package and vehicle" are a likely predecessor
to this 1983 flavor of remote executing (debugging) some 8086/88
Multibus system. The two-ROM version with the same cabling, may be a
variant that did not need ROM based debugging or OS operation.

All the gathered information in this discussion, suggests a series of
Intel development tools, in transition from earlier (ROM dependent)
product to disk-based products.

> It's not clear who would have had a need for the '1983 kit'.

It's likely transitional, for reasons cited or not, and thus a one-off.

The ROM software apparently supports use of parallel I/O in some kind of
debug/control way of a stand-alone Multibus 8086/88 processor which is
(probably) diskless, possibly a modified (not stock) Intel Multibus CPU
board, and is operating some kind of embedded application.

The "modification" is suggested by the additional non-ROM chips, and
some tiny adapter board. Otherwise, how does parallel I/O become a
processor debugging tool?

Who needed it then? I was an embedded systems engineer; early days
development often was ROM based. Or some early-days INtel tools found
one last use by piling on the ROMs.

> Open or leave? My sense is that there's little to be gained by reading the EPROMs.

Consistent with "who needed it then?" is "who needs it now?". Jon,
nobody "needs" any of this stuff now. Yet here we are. Don't be annoyed
Jon, that when you call some prize to everyone's attention, some show
interest in opening the gift.

There's NOTHING to be gained by keeping the package sealed, as far as I
as an engineer is concerned. Fortunately the package itself is stable,
if it were unstable then recovery for preservation would be the "gain".

But there's no immediate need to open it. And now we've collectively
identified the likely Intel products associated with this package.

So it's a matter of time and happenstance, that 1) someone else will
have this set of ROMs but in use OR they will uncover the documents
referenced but not yet available. Then 2) they will find our discussion
and post accordingly. Then we'll have reason to visit this artifact again.

Jon, thanks for calling out this set of ROMs and cables. It's your
artifact, your choice as to what's done with it.

Regards Herb Johnson

On 6/10/2026 5:10 AM, Jon Hales wrote:
> Hi All
>

> Many thanks for the responses to my message.
>
> I think Mark Gent is very close to the answer.
>
> Application Note AP-43 is The iSBC 957 'Execution Vehicle for Executing
> 8086 Program Code'. (November 1978).
>
> Manual 9800743A has the title iSBC 957 Intellec-iSBC86-12 Interface and
> Execution Package User's Guide' (September 1978).
>
> Given their date, these illustrate the connections to the MDS-800, the
> Series-II 210 (with disk drives) and the Series-II 220,230 (with at
> least 32Kb of memory). The 4 EPROMs were 2716s.
>
> The components of that kit were very similar to the 'Mystery Kit' I
> showed in photos. An odd feature of the illustrations in the documents
> is that the iSBC86-12 is depicted outside the MDS chassis ('floating
> nearby'), with the cables connecting the serial and parallel interfaces
> connecting to the sockets on the rear panel of the MDS. The User's Guide

> states: "The iSBC 86/12 Single Board Computer should be installed *in a
> chassis other than that of the Intellec System* (my emphasis).

>
> The Resident Processor Board (RPB-86, based on the 86/12) is equally
> concerned with allowing a 1977-78 MDS-800 or Series II to act as host
> for the development of 8086 applications, and in this case it would be
> fitted internally with no cabling.
>
> It can be imagined that in 1983 there was continued interest by owners
> of 'Intellec' systems in developing for the 8086. By that time, the RPC
> (based on the iSBC 86/30) would have been available and the naming of
> the upgraded system as 'Series III' was presumably quite standard. It's
> not clear who would have had a need for the '1983 kit'.
>
> Herb: Thank you for your comments about anti-static foam. The foam pad
> rebounds almost instantly from thumb pressure.
>
> Open or leave? My sense is that there's little to be gained by reading
> the EPROMs.
>
> Best regards
>
> Jon

[Jon Hales]

Thank you, Herb. That's all useful information.

I was looking at the situation in 1978 from a different perspective. I assumed Intel wanted to encourage engineers to familiarise themselves with 8086/8088 coding, potentially for their embedded production/application projects. I got the impression from the 1978 manuals I referenced that the means of transmitting code to the iSBC 86/12 over the parallel interface was cumbersome and the ROM code on the iSBC 86/12 essentially a monitor that displayed on the Series II.

The RPB was a modified iSBC 86/12 that was to be installed in the Series II alongside the IPC/IOC, usually with additional RAM. This seems a more satisfactory basis for code development. For example, being able to use the Series II disks. And with an ICE unit as a superior approach to debugging a remote board.

It's only recently that I became aware that the iSBC 80/10 (1975) and other early Multibus iSBC parts were intended to enable application developers to use 'off-the-shelf' boards as an alternative to developing 'bespoke' embedded systems. This was made clear in the 1975 story in the Intel 'timeline' web pages.

Best regards

Jon

[Mark Gent]

Don't forget that systems at the time, unlike modern systems that fit on your fingernail, did not have a serial or parallel ports out-of-the-box: they had to be added to the board, making it more expensive and requiring more real estate so people only ever added the bare minimum, if at all.

You could develop with an In-Circuit Emulator but these were astonishingly expensive pieces of equipment, so having anything that provided a cheap software-based communications channel to your embedded system was gold.  The fact that the code worked logically correctly on a development system is no guarantee whatsoever that it will work when you plug your eprom into the socket on the target system, even if that is a standard Multibus board.  Don't forget that Multibus boards are highly configurable, e.g. you can normally change the whole memory map using just (wire-wrap) jumpers, let alone when using customisations using soldered jumper wires as you frequently see on these boards.

So, you plug your fully prepared eprom with your fully debugged software in your target board, power on and nothing happens.  Then what?

If you didn't have something like the 957 kit you could only resort to something like having the software blink some LED if it reached some point (e.g. I actually wrote some code that used a single output to drive a speaker so that it played "Monty Python's Flying Circus" in a loop, which turned out to be vital for finding an issue with dram refresh as after a couple of hours it started skipping parts, which was very easy to recognise even from 2 rooms away).

With something like the 957A and B you could remotely debug a system from the development box without having to buy an ICE while still being able to use a well supported toolchain all the way.  Stuff you can nowadays take for granted with on-chip breakpoints, plenty of RAM, Flash ROM and serial ports as well as standard JTAG, but it wasn't always so.

-Mark

--
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/CANPXPJm9ne%3DCvAye8pn%3DSUNTDQHKKO3q-w_JqbG9yBy%3DHm9xZg%40mail.gmail.com.

[Herbert Johnson]

On 6/10/2026 12:39 PM, Mark Gent wrote:
> Don't forget that systems at the time, unlike modern systems that fit on
> your fingernail, did not have a serial or parallel ports out-of-the-box:
> they had to be added to the board, making it more expensive and
> requiring more real estate so people only ever added the bare minimum,
> if at all.

Mark makes a great case at a point in time, thanks Mark! One other
factor may have been, the absence of a disk-based system, either in the
embedded system or the development/debugging system. My prior thought
was that all the ROMS represented an operating system and debugger that
otherwise (and later in the era) would be loaded from disk into RAM.

I like Mark's idea, that the ten ROMS were multiple pairs of ROMS per
Intel target (embedded) processor board. Time will tell if that was so
for that product.

> It's only recently that I became aware that the iSBC 80/10 (1975) and other early Multibus iSBC parts were intended to enable application developers to use 'off-the-shelf' boards as an alternative to developing 'bespoke' embedded systems. This was made clear in the 1975 story in the Intel 'timeline' web pages.- Jom Hales

I'd love to give my lecture, on the history of introducing
microprocessor computing, at points in time. One point of time, was
introducing microprocessors on the factory floor, and in products and
instruments. That's all called "embedded computing" today.

There was no use case, certainly no software, for what became personal
computing later, or home video gaming, or business desktop computing,
and so on. Such things were not believable, until they happened - then
they became obvious.

Intel's Multibus began on an Intel 8080 teaching tool, a single board
evaluation board that had a ROM monitor and some expansion capability
left unnamed; until it caught on as a product standard. Just as MITS
produced an 8080 microprocessor set-of-boards on an "Altair bus" which
became the S-100 bus. Just as Motorola's expansion connector on their
6800 evaluation kit became "Versabus". All about the same time.

This is one base case for why it's useful to have museums and
collections of vintage computing. It's my base case for my work, about
the early days when it was embedded computing and the beginnings of
desktop computing. I kept myself from posting that lecture today.

Regards Herb

[Herbert Johnson]

On 6/10/2026 1:11 PM, Herbert Johnson wrote:
> became the S-100 bus. Just as Motorola's expansion connector on their
> 6800 evaluation kit became "Versabus".

ExorBus, sorry. The Motorola 6800 D2 kit has an "EXORciser compatible"
edge connector.

https://www.retrotechnology.com/restore/exorset.html#hard