COSMAC FRED2 at Sarnoff Collection

Last updated June 8 2018, correction July 8 2019. Edited by Herb Johnson, (c) Herb Johnson, except for content written by others. Contact Herb at www.retrotechnology.com, an email address is on that page..

Introduction:

The Sarnoff Collection at The College of New Jersey (TCNJ) has several artifacts from the former RCA/Sarnoff Library. Among these is what appears to be a "FRED 2" style COSMAC 1801 computer, constructed and designed by Joseph Weisbecker and his associates in the mid-1970's. In Dec 2017, this computer will be displayed with several gaming prototypes and products of Weisbecker, as part of the Collection's new exhibit Playing with Innovation: The Games of Joseph Weisbecker. In October 2017, I examined and described the computer to Florencia Pierri, the Curator of the Sarnoff Collection; and produced the descriptions on this Web page.

Original documentation of FRED 2 prototypes is held by the Hagley Library, which holds the documenation and paper artifacts of the former RCA/Sarnoff Library. Access to FRED 2 documents has been requested. This page will be updated as more infomation becomes available. - Herb

Background about the COSMAC, Joe Weisbecker, & the Sarnoff Collection

Joseph Weisbecker and his RCA colleages designed the logic of the RCA "COSMAC 1802" microprocessor in the mid-1970's; the microprocessor was produced a few years later. Weisbecker and others built that logic, and later the microprocessor, into a series of prototype computers also used for program and product development. Many of those prototypes and their overall concepts, were known as "FRED". This FRED 2 computer contains the earliest 2-IC version of that microprocessor (briefly marketed as the 1801). IC's on one of the boards of this unit, date to 1973 and early 1974, consistent with the date when the microprocessor in use was produced as samples from the Solid State Technology Center or SSTC of RCA.

The Collection has an earlier COSMAC prototype, "System 00", which has the logic of the later COSMAC processor as individual TTL IC logic chips across several boards. The linked Web pages include documents about the System 00 and a "FRED" computer of similar design. Several "FRED" computers were built, to develop concepts of user interaction and display including gaming, and used as development tools themselves. collectively, they represent development of the FRED computing concept, the COSMAC microprocessor, and COSMAC based tools and products for engineers and consumers.

Here's a link to the Collection's home page. Check "Exhibitions and Programs" for links to this exhibit, and to other exhibits. A good number of RCA exhibits have been on display since 2014. The Sarnoff Collection at The College of New Jersey, has a cataloged collection of documents and artifacts from Joseph Weisbecker's COSMAC work, as part of the former RCA (Sarnoff) Library collection of products, tools, and artifacts. This link is to a database of cataloged items. Other pages and related sites show the exhibited collection. Some artifacts are not yet cataloged or in the online database. Also, a change to the online database system is ongoing as of late 2017.

Most of the former Sarnoff Library's paper documents, as business documents and publications of the RCA Corporation, are part of the Hagley Library in Delaware.

- Herb Johnson

My history with FRED 2

In Aug-Sept 2017, I (Herb Johnson) had access to a COSMAC computer while it was prepared for exhibition, with other Weisbecker gaming artifacts. The following are photos I took, and sketch I made of the overall layout of the computer. From the "FRED2" designation on the CPU board, I've tentatively identified this computer as a "FRED 2" prototype.

- Herb Johnson

Photos of the entire computer

A view of the FRED 2 from the front and top. On the front right, is a "front panel" of switches and a 16-button keypad. On the front right, is a Panasonic brand hand-held cassette recorder. Data and programs for the COSMAC were recorded and played-back as audio tones on a cassette tape. In the mid-1970's, this was an available and cheap means for data storage.

A view of the FRED 2 from the rear and top. On the rear right is the "card cage" of several logic cards, the logic and CPU of the computer. On the left rear is a rear-panel with controls and connections for video display, audio. There's a conventional AC outlet.

A sketch and photo of the layout of the FRED 2's major components, viewed from above with the front-panel at the bottom. The individual cards are identified by A or B and a number; these identifiers were hand-printed on the hand-wired cards. The Power Supply, of ordinary commerical design for the era, converts AC power to the various DC voltages needed for the logic. The various major components are photographed and discussed below.

Control Panels

A closer view of the front panel, left photo. Note the "JAW" for Joseph A Weisbecker. The right photo, shows the logic on the back of the keypad.

The 16 keys of the keypad, represent hexidecimal codes and data. The various buttons and toggle switches, provide means to enter and review data, and to run and stop the microprocessor. The COSMAC microprocessor instructions or data, are entered as a series of hexidecimal values (base 16, 0 thru 9 and A thru F). The five red rectangles on the lower left, are a five-digit hexidecimal numeric display. Three hex digits are sufficient for memory addresses to four thousand bytes (4K) of memory. Two digits represent the 8-bit data value as two hex digits.

Rear controls and connectors. The rear back-panel has controls and connections for video display, audio. There's a narrow vertical blue slot, likely with connections to insert a small electronics card. There's a conventional AC outlet, and below it is likely the connector to an AC power cord.

Cassette recorder player

A visitor, looking at the exhibit's FRED 2, was confused about the function of the cassette recorder / player. Did you "play the cassette" as part of the game? Maybe to include audio or music, as part of the game-play? The answer is, the cassette tape would not have audio at all. The function of the data cassette and recorder, is as a digital storage device, like a flash drive; or an older floppy disk drive and diskette.

The cassette holds the game-program itself, as binary information. The bits for the COSMAC's instruction codes, are represented as audio tones. The tones are encoded or decoded by electronics in the FRED 2. What sounds like bird-song noise to human ears, is binary data to or from the computer's electronics

So, the cassette "stores" a gaming program and its data; that's "loaded" into the computer, and then "run" to start and play the game. It was "saved" to cassette, when the game program was developed, on that or another FRED 2 computer.

In older computing terminology, the cassette contains one or more "executable program files". In modern personal-computing terminology, the cassette is storage for the "app", the application, which must be loaded into the computer before it's run. Here's a linked Web page, on the technical features of audio cassettes for data storage. There's links there to other COSMAC computers which use audio cassette data storage. As I learn more I'll have cassette storage information these RCA development system "FRED"s on this linked Web page.

Boards

The above view of the FRED2. The right-hand side shows the B-cards, which includes the power supply and the front and rear control panels, and electronics for tone and cassette tape. The left-hand side shows the A-card cage, the boards with the microprocessor, memory, and other computing logic.

A-section boards

The A-board section, represents much of the "logic" of the computer. It includes the CPU board, memory board, and other logic (likely including a digital video display). Details described below are means to determine some of these functions.

The A-board card cage, two views. Boards are numbered from right to left. Visible cards are A12 (a two-card set for RAM), A3, A7, A8 and A9. Card A6 is removed and card A4 is the short CPU board, in blue. "Vector" is the brand name of the empty protoboards. The cards were hand-wired and hand labled.

View inside card cage, left photo. NOte slot number 5 and dots for other slot numbers. Slot 4 is the blue board with two processor chips. Slot 3 "card A3" has a 28-pin chip in socket. At far right is slot 2 with "card A12", a stacked card for both slots. On the far left, you can see the wiring for card A7.

View from above into card cage, right photo. The blue CPU board A4 shows multiple large chips. Two of those chips constitute the COSMAC 1802 microprocessor. That blue board looks similar to the CPU board included with the COSMAC Microtutor.

I was informed by Andy Modla, about a likely clock frequency for the the FRED 2. Andy Modla is a former RCA employee who worked on COSMAC development with Joe Weisbecker in the 1970's. Andy had access in 2017 to FRED 2 - related documents at the Hagley Library. In private email to me in Jan 2019, Andy said this regarding those documents. "..after looking more carefully I see a clock line label for .625 uSec. Now this implies a clock frequency of 1.6 MHZ."

The center of this image, shows card A3, with a 28-pin white ceramic and gold chip. The chip reads: "RCA-SSTC TCC152A - E 7440 CC". SSTC is the RCA Solid State Tech Center and identifies this as a prototype chip. 7440 dates the chip to 1974, week 40. TCC152A is the designation for the chip. Card A4 has two large IC's, the microprocessor chip-pair, which likely are also SSTC originated IC's. Later information, suggests this chip is an SSTC prototype "video processing" chip.

The top center of the photo, shows "FRED 2" behind the edge of the perforated board A3, etched on the board A12 board-pair. Here's a closeup of that etched label, as "FRED 2 2K RAM". Also on that etched board, are the initials B J CALL. Billie Joe Call was one of Weisbecker's associates, an engineer and programmer of COSMACS and FREDs.

View inside card cage. "Card A9" and "Card A8" are on the left, "Card A7" is partially revealed in the center. On the far right, a view of RAM card A12 in slots 1 and 2.

Program ROMs?. Later inspection of the FRED 2 boards without disassembly, did not uncover any ROM IC's. It's likely that, like other early COSMAC systems, programs were loaded in under the COSMAC "LOAD" mode of operation; no program ROM is needed to initiate that process.

B-section boards

Left: a portion of card B2. Right: view of card B3, showing "TONE" subassembly. B2 and B3 likely provided logic and signal electronics to support the "analog" features of the computer, such as audio tones and encoding/decoding data on the audio cassette drive. Tones stored on audio tape, represented binary values of programs, stored or recalled as files.

A view of the power supply. This was a standard product of the era, produced by many companies. It provides the DC voltages to power the logic and other electronics.

How were these boards made?

Each board was either hand-wired, or was produced from etched copper based on a "layout" design. I'll explain how a board of IC's can be hand wired, so each wire properly placed by hand.

Card A6, removed, shown component-side up. (the back of the card is shown below). Those black rectangles, are individual CMOS logic chips. Each chip performs a few simple binary logic functions: AND, OR, NOT; or a two-state logic element called a "flip flop". RCA was first to produce Complimentary Metal Oxide Surface or C-MOS semiconductor logic devices. These operate with microamps of current and have other attributes, as compared to the "transistor-Transistor Logic" or TTL chips of the same era. These devices were produced in 1973 and 1974 based on the year/week 3-digit codes printed.

The CD40--AE printing designates the standard logic function of each chip. The same chips in the CD4000 series are still produced and used in the 21st century, four decades later. See this text file for a layout of the IC's on this A6 board.

Card A6, removed, shown wire-side up. (the chip side of the card is shown above). This was hand-wired to follow the schematic design, connecting the socketed IC's pin by pin. The top edge and left edge, show the 1-2-3 and A-B-C identifying each chip. Thus chips are designated "B4" or "D2". The schematic would show both the IC part number and the letter/number designation. The bottom edge of the board with the gold tabs, is the "edge connector" to the "bus connectors". On this side they are identified as A, B, C, and so on. On the chip side, they are identified as 1, 2, 3...22.

FRED documentation

Some documentation of FRED or FRED-related systems, is held by the Sarnoff Collection and the Hagley Library. They have terms of access and terms of use within their facilities, and limits on distribution. Some of that material has become generally available. Other COSMAC Web pages I have, links in this document, give a narrative or link-to a narrative about all that. Those Web pages point to such materials as were available at the time of that writing. There's few RCA COSMAC development documents available from private hands today. There are some documents as technical articles of the mid-1970's; those are owned and managed by their publishers.

Documents something like this FRED 2 system, are referenced on this linked Web page about FREDs and audio cassette storage. There's an earlier "FRED" manual from 1972-74; but that's for a pre-microprocessor system such as the "System 00" owned by the Sarnoff Collection. I'm sorry COSMAC documents are scattered and fragmentary; their use and distribution is limited by the circumstances I've just described. - Herb Johnson

What were the "FREDs" used for?

uncapped COSMAC 1802 microprocessor

Described above, is a CPU board, with a paired chip-set COSMAC microprocessor. The FRED 2, used two chips to represent the COSMAC microprocessor. Later it was reduced to a single 40-pin IC. On Sept 13th I spotted among the odd bits at the Collection, "a big IC chip"; an uncapped 40-pin DIP integrated-circuit package, The photo on the right shows most of the IC; the photo on the left is a closer view of the "die", the silicon inside. I guessed it was the COSMAC 1802, or maybe an 1801. The upper right area of the die, suggests the 16 registers. A Google search finds a matching die image at the visual6502.org Web site of the COSMAC 1802.

In mid-Feb 2018, the Sarnoff Collection's curator, found a display artifact. It was a collection of COS/MOS integrated circuits and FET transistors, in a display case. Among them was an entry for a COSMOS microprocessor - but the chip was missing. I think the "uncapped" 1802 is a likely candidate. I told the curator, and she'll eventually retrieve the chip and see if it "fits". - Herb

This page and edited content is copyright Herb Johnson (c) 2019. Copyright of other contents beyond brief quotes, is held by those authors. Contact Herb at www.retrotechnology.com, an email address is available on that page..