International Electronics Unlimited 16 bit microprocessor kit manual 1975

[mountainking]

I stumbled upon this manual on ebay, it only had a pic of the cover and it didn't really have any info about what it was, but it did say "microprocessor kit 1975" so I decided to grab it. I wasn't able to find out much about the kit, it looks like International Electronics Unlimited sold the kit but accept for some ads in Byte by that company I could find anything. The postage stamp says "Feb 20 75" which is pretty darn early for a micro kit and the manual is super "lofi" so I would really like to find out more about the kit. Notice that the manual using the term CROM, I've never seen that written any place else, though I suppose with this being so early in the microprocessor game that there were terms that hadn't yet been written in stone.

Anyone have any info or thoughts?

 

[Chuck(G)]

National Semiconductor IMP-16. (The NS PACE MPU is a direct descendant). PMOS stuff.
You can find data on the chip family here: http://www.bitsavers.org/components/national/imp/4200036A_IMP16P_Descr_1974.pdf

The IMP-16 did find some application; one that comes to mind is the early Sun engine analyzer. Here's the patent that mentions the IMP-16

Note that RAM on this thing used Intel 1101 chips--each with a whopping 256 bits.

 

[Eudimorphodon]

Yeah. To clear things up a little, the IMP-16 was a “bit-slice” design that broke up the “microprocessor” into multiple chips. “CROM” isn’t a goofy name for “system ROM” in the sense you’d use on a normal microprocessor, it’s a specialized component called the “Control and ROM”, which is basically the CPU’s sequencer and microcode store, riding herd over four 4-bit ALU chips which all together make a 16 bit processor.

 

[mountainking]

National Semiconductor IMP-16. (The NS PACE MPU is a direct descendant). PMOS stuff.
You can find data on the chip family here: http://www.bitsavers.org/components/national/imp/4200036A_IMP16P_Descr_1974.pdf

The IMP-16 did find some application; one that comes to mind is the early Sun engine nanalyzer. Here's the patent that mentions theIMP-16
Ad
Note that RAM on this thing used Intel 1101 chips--each with a whopping 256 bits.

I meant that I couldn't find anything about the actual kit and it's history. I am interested in the technical side as well but that info I could find.

There weren't a ton of microprocessor kits available in early 1975 so I would have figured that there would be some info about it and it's history. As far as the 256 bit RAMs, that was pretty common then, I've seen a quite a few micro projects from the early days that use two 256 bits RAMs to make 512 bits. Memory was crazy, stupid expensive back then so they designed projects that used the minimum amount of memory that you could actually do something interesting with, which back then was pretty basic stuff, I mean I have heard a bunch of people talk about how exciting it was when they were first about to program a microprocessor to make an led blink (eg the Cosmac Elf), but shortly before that the idea of the average person being able to own any type of computer was just a dream. This is why the Altair project/kit in PE caused such an massive reaction even though you couldn't really do much of anything with it.

 

[Chuck(G)]

I remember IEU mostly for back-of-the-magazine ads for various parts (ICs, resistors, etc.). You'd see their ads in early PE, 73 and I think BYTE. Not unsual that they offered a kit.

As far as "not being able to do something with it" Altair 8800--I built one in early 1976 and did plenty with it. Still have the blessed thing, cheap white wire and all.

 

[Eudimorphodon]

There weren't a ton of microprocessor kits available in early 1975 so I would have figured that there would be some info about it and it's history.

I don't think it's that odd that there's not much press about it, given the IMP-16 wasn't quite a "microprocessor" in the same sense as an 8080 or 6800 was. It was closer to a complete microprocessor than, say, more generic bit-slice ALU components like the 74181, in that the CROM to tie it all together was a pre-baked component that they'd sell you in a bundle that added up to a full CPU, but it still sort of straddles the line into being a "minicomputer-like" architecture. (It was apparently actually heavily inspired by the Data General Nova.) At the very least that makes it a bit more imposing to get started with than a simpler 8-bit microprocessor.

It would be interesting to know what this "kit" sold for, and what it actually added up to. What we can see of the manual doesn't seem to define what it included for I/O, if it came with some kind of system monitor or relied on some kind of a front panel, etc, etc.

 

[Chuck(G)]

I doubt that many were ever sold. I don't recall anyone ever telling me about one they purchased. The IEU card seems to be a cost reduced version of the IMP-16C. I'm not aware of a hobbyist program library or interest group. Even the PACE, which was pretty much the microprocessor version of the IMP, didn't garner much interest. NS were giving the chips away at conferences like Wescon.

DG did come out with the MicroNOVA chip in 1977, but other than in DG products, I don't think that mN601G was available for outside sales.

 

[mountainking]

I don't think it's that odd that there's not much press about it, given the IMP-16 wasn't quite a "microprocessor" in the same sense as an 8080 or 6800 was. It was closer to a complete microprocessor than, say, more generic bit-slice ALU components like the 74181, in that the CROM to tie it all together was a pre-baked component that they'd sell you in a bundle that added up to a full CPU, but it still sort of straddles the line into being a "minicomputer-like" architecture. (It was apparently actually heavily inspired by the Data General Nova.) At the very least that makes it a bit more imposing to get started with than a simpler 8-bit microprocessor.

It would be interesting to know what this "kit" sold for, and what it actually added up to. What we can see of the manual doesn't seem to define what it included for I/O, if it came with some kind of system monitor or relied on some kind of a front panel, etc, etc.

Yeah, I think you're correct, after thinking about it I realize that the tech used for the circuit wasn't the "new and exciting" 8008 or 8080 anything like that, it was tech that had existed for some time and was probably seen more like TTL, like you said. And with that all in mind it probably didn't sell in any great number and add that to fact that it wasn't of much significance to the history of the CPU/PC and you have a pretty mundane "bit of kit", as they say. Oh well!

But it did spark up this conversation almost 40 years later, and I for one learned some new bits (no pun intended ) of knowledge from it so if nothing else it has this to make it's existence somewhat significant, right?

Thanks,
Alan

 

[Chuck(G)]

One aspect of the IMP-16 chipset that was shared by the PACE was that both were PMOS, meaning that level translation had to be used to get to TTL levels. That caused the chip count to multiply and required something other than +5 from the power supply. PMOS also wasn't very fast.

 

[Eudimorphodon]

Yeah. The IMP-16 was specified to run with a 700Khz-ish multi-phase clock, and if you look at some of the docs that are on Bitsavers it looks like its instruction times were... not great. There's a decent chance it would actually be slower than an 8080, although its 16-bit-ishness might give it an advantage in certain niches. (Maybe that's why they decided it made a good brain for an engine analyzer.)