SCAMP: A Review 50 Years Later

In July 2023, circumstances came about that I came in contact with one of the department heads at the American History Smithsonian Institution in Washington, DC. After some discussions and planning, it was agreed I could examine their SCAMP archive. In particular, the Joseph D. George Engineering Notebooks, who was one of the key hardware engineers involved in the development of SCAMP and had kept a meticulous technical journal.

My understanding is SCAMP was donated in 1988 and Joe George contributed his notebooks some time later, possibly as late as 2003 (30th year anniversary) or perhaps late 1990s.

SCAMP @ Age 50

*** Click on images for enlarge view ***

The SCAMP prototype was demonstrated during the week of September 10th, 1973. The first demo was Tuesday, September 11th (New York), with the final demo Thursday, September 13th (Regency Hyatt Hotel in Atlanta). That makes 2023 now the 50th anniversary of that system. With permission (and gloves), we exercised the “pivot mechanism” used to close and re-open SCAMP.

When opening the system: To the right of the central controls (behind the cassette player) there is a slot in the front face, this is a “hold point” indentation — and is used to help lift the SCAMP back up/out. This subtle handle may be dark in some photos, but it is above the “IBM” logo.

When closing the system: Slightly lift the “hold point” mentioned above to get pressure off the “button” on the left side of SCAMP. Depress that button and the weight of SCAMP will sink itself down into the case (while the keyboard internal parts pivot to all slightly collapse down together). Then pull the case cover visor down (taking care that there is no “stop” on the visor, so avoid pulling it all the way out).

In either operation, the case was silent (no squeaks or noises). A very excellent design! Except that SCAMP is indeed near 50lbs.

Within the archived notes, the IBM Industrial Design are signed by “G. L. Jarvis” (Gerald). A wooden mockup is mentioned to ensure correct shape, then final material is designated as either LEXAN or NORYL FOAM. But the prototype is said to be POLYURETHANE FOAM. There were specific requirements from IBM for the system to sustain a “36-inch drop” (hence the soft edges), which was an aspect that didn’t transition over to the final IBM 5100.

The fold-out control panel has the following:

HEX DISPLAY: As on the IBM 5100 production system, a toggle switch to change the display to showing hex-values (aka registers-mode).
SINGLE INSTRUCTION: Toggle switch to go into RUN vs STEP mode. Followed by white button used to actually STEP the instructions.
ADDRESS: I’m not entirely certain about these dials, but I think they set what Device Address to use for certain components (like the CRT and Printer or Comm. card). There was a proposed idea to include an acoustic coupler towards the rear of the system, so these dials may adjust for variants of such devices during development. And/or one of these dials may select which address to IPL software into (initial program load).
The red button I believe is used to activate the IPL (initial program load) of data from the audio tape. In this regard, SCAMP is somewhat like the earlier Datapoint 2200.

As it turns out, the SCAMP does appear to have a sort-of power supply behind the black grill. The external power supply (in a silver box not shown) does handle some part of the AC wall-outlet portion of the power distribution, then goes into the “round three prong” connector at the back of the system. The other external connectors of SCAMP are the BNC video output, single DB25, and then a three-row connector that I am not familiar with.

According to an August 1973 letter from Jarvis (here) in regards to shipping dimensions of SCAMP, he mentions both a “fan – muffin type” of 3x3x1.5 inches and a “L-shaped” power supply.

Jarvis indicates that his original design included slide-cover plates to the rear external connectors. The blueprint-like drawings are excellent, but we did not have the space available to unfold them properly and did not want to risk any damage to them. In this context, “I.D.” means Industrial Design.

The following were inserts stapled into Joe George’s notebooks:

They are of particular interest because of the following:

A January 1973 letter mentioning the name “Howard Kazan” that is suspiciously similar to “Harry Katzan” (author if the “IBM 5100 Portable Computer” comprehensive 500+ page book)
This “diamond pin out shape” is one of the variety found on the IBM cards, and here is being referred to as “Dutchess.” It’s very exciting that we might have a confirmation on the pinout of those components.
Note this early proposed keyboard arrangement uses the word ENTER (which neither the SCAMP nor final IBM 5100 used).

JOE GEORGE NOTEBOOK (set of 8)

There were about 800 pages in total, so I’ve filtered down to the most interesting highlights. These will just be “snippets” of those highlights (as searchable context to reference in the future). The original images that I have are in the link further below, which can provide broader context.

Joe George Notebook Summary PART 1 (1972 – 1973)

Joe George Notebook Summary PART 2 (1974 – 1975)

SCAMP related PDF summary HERE

Paul Friedl notes (PDF):
https://github.com/voidstar78/SCAMP/tree/main/PJF

Joe George notes (PDF):
https://github.com/voidstar78/SCAMP/tree/main/JG_TechNotebook

As a sample, here is one of my favorite entries:

It is dated April 1st, 1974. I’m not sure John’s position in IBM, but apparently John feels there is a legal necessity that SCAMP should be left in the original (probably meaning “as demo’d”) condition. This does imply that variants of the SCAMP configuration were being built, causing confusion to product support.
Note that April 1974 is now seven months after the famous September 1973 demo in Atlanta. But also keep in mind that April 1st is traditionally April Fool’s Day.
The lower note on April 2, 1974 is more interesting: “Room with SCAMP unlocked and door open. I arrived at 8:15am – what kind of security is this?” Apparently with the primary demo accomplished, and project funding now secured, the secrecy and sacredness of SCAMP is already somewhat starting to wane. Or was it just a late-April Fools prank?

THE TAPE COLLECTION

The collection included a box of about 40 cassette “audio-data” tapes (of mixed brands). Most of them are, per their written labels, related to PRINTER samples (as in testing integration of the new dot matrix printer with SCAMP). One of the most interesting titles to me was “ALSOUP” (I suspect Al here is Alvin Ginsburg, one of the developers involved).

I believe original SCAMP only allowed one file per tape – meaning each tape is probably a single program/software only and not any freeform data (no letters or numeric sample data). I captured what I thought was the more interesting subset of the total. Here is an itemized description of the titles of the tapes shown further below.

SONY: BOOTSTRAP (no date)
SCOTCH: SCAMP TAPE 10/3/73 DUPLICATE COPY
TDK: Roy’s Printer APL Full Lookahead backup 4/28/75
SONY: ALSOUP 6/7/74
TDK: original 12/5/74 good tape tape of 3/4/74 backup
(IT?): AL’S DEMO 4/16/74
SCOTCH: SCAMP DEMO w/ TELLY APL updated 11/21/73
SCOTCH: SCAMP DEMO TAPE 3/5/74
TDK: copy of ALSOUP 6/9/75
SONY: 2 copies APL backup start 00C0 (no date)
(IT?): CERTIFIED DATA CASSSETTE (just an example of what an “official” certified data tape looks like)
(IT?): copy of sales aid demo 3/26/74 (black specs not a good sign)
SCOTCH: SCAMP BACKUP 10/8/73
SCOTCH: SCAMP APL 3 files 11/21/73 (Pat Smith, no other writing on that paper) [Patrick was one of the software developers on the project, the silver band is tying two tapes together]
SCOTCH: SCAMP1 APL backup 9/24/73
(IT?): 1/16/74 SCAMP PROGRAMS (also signs of mold, black specs) [ this is the tape that was set into the SCAMP tape deck when we reviewed it ]