System/1 Build Log

An ongoing chronology of System/1's construction, and any other related musings.

8th March, 2015

Just a quick administrative note today, really; I've not had time to work on anything System/1 related for a while, but over the past week (having finally shifted the nasty case of writer's block that was interfering with the documentation-writing I'd been tasked with at work recently) I sat down and filled in various backdated entries in this build log.

They're probably not worth going back to read — mostly the fairly dull trials and tribulations of fumbling around with logic design without a clue — but at least they're there for completeness now. The new entries span the period between 20th June and 17th December last year.

I've also written up the six boards I've built to date, and they're linked from the main page; not hugely happy with the board photographs I've included, so might reshoot those at some point.

With half of the backplane filled, the system is beginning to take shape; I ended up building a (disgracefully shoddy) stand for the backplane, so that I can hook my breakout board up to it and try testing several boards working in concert. Before that can happen, though, I'll need to wire up some of the interconnections between boards; time to get to grips with wire-wrapping...

System/1 as completed so far; that register file board is going to need some support!

I think I've decided to tackle the control unit next, at least as far as drawing up an interim schematic so I can calculate some propagation delays. A comment spotted lurking in the EEVblog forum recently had me looking at the 8272 floppy diskette controller and wondering whether I could interface it to System/1 when completed; the answer seems to be that I'd need to be able to poll it for new data every 27µs, which wouldn't be possible with the 4MHz crystal I've played with so far — it would need to run at 8MHz or better before that requirement could be met. Adding an 8272 would mean polluting the design with an ASIC, though, so maybe I'll just stick to IDE hard drives if I ever get as far as attaching mass storage to the contraption!