PDP-8/E, Serial Number 10020

This is Mike's personal PDP-8/E.

Current Status:

Processor working OK

32kW of RAM memory working OK

Second serial port working OK.

RX8E and RX01 installed. The left drive works if you wiggle your the microcode ROMs.

TD8E and TU56 installed. The left drive works OK. The right drive has a bad head.

RX8E and RK05 installed. Works great!

The PDP-8/E, introduced in March of 1971, is the fastest of the non-microprocessor PDP-8 systems, and the last to offer a "blinky-light" front panel. It is the first PDP-8 to use a backplane bus (Omnibus) instead of wire-wrapping individual signals between flip-chips in dedicated slots. The Omnibus saved a lot of space, reduced manufacturing cost, and improved reliability. The PDP-8/E normally came with a 20 slot backplane, but could be expanded to 38 slots with any additional backplane. It could be expanded again to 74 slots with an expansion chassis containing two more 20 slot backplanes. DEC also made PDP-8/F and PDP-8/M variants where the chassis could only hold a single 20 slot backplane.


The KA8E External Positive I/O Bus interface let you connect peripherals from the PDP-8/I and PDP-8/L to the 8/E.


With the Memory Extension and Time Sharing module installed you could use up to 32kW of memory. The Time Sharing option enabled a 4k virtual system for each TSS-8 operating system user. This is now called a virtual machine. The Time Sharing option traps the CIF, CNF, IOT, OSR, and HLT instructions to prevent users from crashing the system.

Click on the image for a larger view.

From top to bottom: TU56 DECtape, SYKESdisk dual 8" diskette, PDP-8/E

From Datacut:

The PDP-8/E was originally puchased by Datacut (G & G Instrument Corp.) in 1974. It was their our computer and we used it to write Numerical Control (NC) part programs for G&G's NC milling machines in their machine shop. The system had an ASR-33 Teletype as an I/O device, which also punched paper tapes for data input to the NC machines. We found the TU56 DECtape to be somewhat slow and cumbersome so we added the Sykes dual floppy.


While we were producing these NC part programs for G&G and eventually other shops, we began developing programming software to make part programming easier. Eventually we developed a complete language based programming system, which they sold to their first customer in 1977. We supplied turnkey systems as a DEC OEM built around the next generation of PDP-8, the PDP-8A. Our software was written completely in PDP-8 assembler language, which gave it a "good bang for the buck" at the time.


Shortly after, DEC came out with its DECStation 78, which it marketed as a word processor. The computer was built into a CRT display and included dual floppy disk drives. This computer was built around an Intersil microprocessor that ran the PDP-8 instruction set. We bundled our system around the DECStation and added a plotter, tape punch, printer and furniture. About this time, this division split from G&G and became Datacut Inc.


We also continued marketing a system built around the PDP-8A and integrated the OMNI-8 multi-user operating system for PDP-8's developed by Network Systems Design out of Oshkosh WI.


The DECStation gave way to the DecMate I and then DecMate II, which had faster processors and integrated hard drives and screen graphics. Our software also evolved with interactive graphics.


By 1985, Datacut abandoned PDP-8 based hardware in favor of PCs. We first ported over our language based product to a system called GRAFX+ and then in 1988 began marketing a totally graphics based "point & shoot" CAD/CAM system called GRAFX II.


Click on the image for a larger view.

The front panel is simpler than some of the earlier PDP-8 systems.

This one has a problem with peeling paint and will need to be restored or replaced.