Thank you CJ, this too adds texture to bare fact.
I still mean to write some posts on "how to compute without a digital computer", but it'll have to wait until after the holidays.
This will be interesting, I shall look forward to that.
No sequencing, no real-time clock, no A/D or D/A conversion, no worries about cycle time or memory allocation. No programming-language issues, no naming of variables, no compiler faults, no software to debug.
I'll admit I was still a bit doubtful, until you wrote this, particularly:-
no compiler faults, no software to debug.
and I say Amen to that!
Modern software, at least the stuff that drives my kit, comes
delivered with bugs and very often the 'workaround list', until that is superseded by an even longer list.
The software on the original EMIScanners was written in BASIC and was amazingly efficient as well as being surprisingly bug free and 'bomb proof'. Of course it wasn't very quick and it was possible to figure out almost exactly what was happening at any given millisecond.
Modern software - on our kit anyway - is doing fifteen things at once while processing huge dollops of data - much of it computational - and no-one, not even the software writers, can tell you anything more than approximately what is happening at
any time.
I think, perhaps, my respect for the analogue designers comes from my relative weakness on the core electronics. I've been a 'System Engineer' for too long and some days I never even get a screwdriver out.
Looking forward to more guys.
Roger.