Home > Previous Page >  Deep Thought Disc Drive System.
Archive Search  

Deep Thought Disc Drive System

This page is currently a WIP - Sorry for the delay due to ill health.

Dr Stuart Leask emailed the following details on 4/12/2018.

" The project was heavily influenced by the floppy disc interface for the ‘Junior’ microcomputer project*, documented in the magazine Elektor (still going strong) in November 1982.

* They documented a disc interface for their Junior micro, which used, instead of the very expensive disk interface chip (I think the usual one was a WD2795 - one chip cost 25 pounds – getting on for 100 quid in today’s money!), used a much cheaper UART. I recall that the idea for a quick ‘n dirty interface to what seemed at the time a magical data storage medium (compared to audio tape) was refreshingly simple:

1. The data input & output from a Shugart floppy drive was essentially serial data from the electromagnetic read/write head, reading a ‘stripe’ of data, which stripe being defined by how far from the centre the read/write head was.

2. 5.25” floppies had a hole punched near the middle; drives provided the output of an optical switch looking through this hole, which ‘appeared’ once per rotation.

3. The drives also had inputs to allow you to step the read/write head in and out (from the centre), allowing it to read/write the difference stripes of data on the disc. So you know when a stripe of data is about to start (from the optical switch seeing when the hole goes past), and you can say which stripe it is.

I recall that given the rotation speed of a 5.25” floppy in a Shugart and the data density, data came off the read/write head of the disc at about 1 bit every 10 microseconds, which isn’t enough time to do anything useful with a 4MHz 8 bit micro. BUT if the data from the read/write head is put through a UART (changing bits to bytes, and vice-versa), you had 90 microseconds per byte – which IS enough time to store the byte, look at the optical sensor, increment a counter or two etc. So the Z80 is used as the interface chip, with the help of the UART – a cheap solution.

I adapted the schematic of the Junior interface to suit the ACE, and Jeff wrote the software. I think we got 70kb to a standard floppy, which in those days was a tremendous amount of data, and as it only took a few seconds to read or write to a disc, it felt blisteringly fast. "

The DOS software was written by Jeff Shepherd - See here for the DOS ROM dissassembly

Screen Shots

Below are a few photos from some 5.25 floppy disk directory listings showing some files on disks. The last image shows the ACE DOS words (highlighted with a yellow box around them). Notice that the call to Deep Thoughts ROM from the ACE's patched ROM (We have named the "BootROM") lists the new DOS words after the Aces QUIT word.

Deep thought Disk listing 




Deep Thought DOS words

Deep Thought Hardware Images.

Stuart Leask's heavily modified Jupiter Ace
This image is looking down at Stuart Leask's heavily modified Jupiter Ace. The are many additions to the Ace but on this page its all about the Deep Thought disc board which can be seen clearly, its the brown vero-board pcb which sits over where the rubber keyboard mat would be.

Close up of the disc interface

Close up of the disc interface the chip with the sticker "DOS 4" is the 4k EPROM containing the DOS works and code.

Underside of the board.

Underside of the board.

Drive connection ribbon the interface can use two drives. If you want to take a closer look you can with these zipped up bigger images (12MB) you can down load.

Deep Thought Hardware Schematic

We do not know what code is in the eprom that controls the Jet-Disc, also we expected a disk controller chip and not the 6821 and 6850 chips. You can download a Hi-Res png image and Data sheets for the 6821 and 6850 chips in this zip file [10 meg]. If anyone has any images or a ROM dump of the EPROM or even any discs it used, then please consider lending them so we can preserve them.