In the summer of 2002 David Collier, the author of the DAI firmware,
was prepared to anwser some questions on the DAI history.
- How did the DAI PC come into being?
- A guy from TI UK, Peter Van Kuylenberg ( Cuylenberg? ) came to DAI
because they were totally pissed off with the timescales from TI USA with
making the 99/4 available. And even then they had no intention of making
it do anything but NTSC video.
He bunged DAI $30K or so to run up a prototype for a European unit which
would do PAL/SECAM, 220V etc.
It was taken to a board meeting in Dallas on Concorde, and scared the
board into advancing the timescales to pease TI (Europe). So $30K well
- What type of company was DAI before they started with the DAI
computer? Was the "Real World Card" system a complete system
(like the Acorn System 2 - 5), or just a bus definition and a
set of I/O cards?
- They had an 8080 CPU card which drove the bus, and on which the PC was
based. I did eventually make a Z80 card after DAI had gone down, when I
was working at DAI(UK) which continued afterwards.
Before the PC they did custom designs, and industrial control.
- Was there a concept of what the DAI computer should be able to do,
or was it just "more or less the same as the TI 99/4". Or were
there other computers used as "inspiration"?
- I don't think we had seen a 99/4, at least not in any detail. The graphics
capability seemed to spring straight from the fevered imagination of David
Lockey - he just told me what it could do, and I had to make a way to
present it to the user. Of course it was fiendishly difficult to get high
res graphics with almost no RAM. He produced a scheme which supported 4
colours on a screen line, with the opportunity of changing just one of the
4 on each new screen line, then expected me to write software algorithms
so the user didn't notice his feet were tied together.
- The internal connector was probably designed for the DOS
extension BIOS (and used by the MDCR). Was there an idea of what kind
of OS should be made for the DAI (CP/M or other)?
- Well Claude Simpson, the MD of DAI, HATED anything not invented at DAI.
Even though there was a perfectly good Intel disk-based development system
he made us design and build our own (the one you have a picture of )
He especially hated CPM, which was by then distributed by Vector
International, a company formed of ex-DAI employees, who were suing him.
Only when customer pressure got too much did he get someone to implement
CPM for it.
Some time later (2004) I heard from Richard King the following interested snippets of information:
- About the case:
- I worked for DAI from '76 thru '79, & was involved with building
the prototype. Getting it all to fit in a
TI Silent 700
case was a pain, but we did it!
- About the board:
- Re : the pcb : yes, it was done by hand. We had an offer from a local
company to do the layout, but they said 2 weeks. We were in a hurry, so
we did it ourselves in 5 days : that is, 120 hours, working around the
clock! I remember doing the dynamic RAM section between 4am & 8 am
one Saturday - I managed to forget all the power lines! It's much easier
In 2012 I got an email from David Lockey which could: add a memory that
David Collier saved the software on teletype paper tape. The result
was a large spool and because of the time to repunch, software
patching became almost an exercise in origami.