On Fri, May 6, 2016 at 10:05 AM, Herb Johnson via vcf-midatlantic < vcf-midatlantic@lists.vintagecomputerfederation.org> wrote:
Thanks to Bill Sudbrink for starting this thread, and to other participants for their response and encouragements for some kind of "lecture" on this subject. Bill notes he got questions about IMSAI/Altair front panels. I'm often asked, why are S-100 cards so big? Why the "ugly boxes of big boards"? It's all related.
As I see it, I'd talk about how early microcomputing (1970's, pre and early CP/M era) was mostly what I'd call "resource scarce". The contrast would be post IBM-PC, certainly 1990's, I'd call "resource adequate", and later "resource rich".
In the early 70's, everything was expensive in terms of then-current income, and then-current technology. Memory. Circuit boards - big, to hold all the TTL chips needed. Low speed of processors. Speed and real cost of storage and display. And 100 pins? Sort of circumstantial, but it's a way to bring out all the decoded 8080 lines and features, so other cards didn't need to decode them, and to manage a front panel.
As Dave McGuire and others pointed out, front-panels were necessary as I/O devices; "storage" was pencil and paper as others noted. Good work could be done, as writing 1K or 2K or 4K programs was well within the capacity of one person, working in his/her head, in octal (or hex, or binary). (I have modern examples of this, in 1802 coding.)
But, once there were ROM monitors, terminals or text-video, and some mass storage (cassette and diskettes), a front panel was an extra expense, time consuming, complicated. It went away. The Ithaca Intersystem DPS-1 I exhibited, across from Bill's Cyclops display, was the last S-100 front-panel system (I think), produced in 1980.
Herb
thank Herb for putting things into perspective, it helps. Which is why I keep your website on my short list, when it comes to rejuvenating my memory cells :) Another perspective which I think is often neglected in history books is the emotion. We had a school computer, Honeywell H1646 in the 70s This is the timesharing version of the H-316, except it had dual H-316 minicomputers in a rack. The H-316 is one of the many systems used as a IMP for Arpanet. Plus the associated hard drives, Mag tap, Line printers, and terminals in the Lab. So you would spend all day here, without any astonishment because that was the norm of the day. But then you come home, working on a homebrew S-100 machine, And you experience that enormous elated feeling because you just upgraded your machine, with a parallel ascii terminal keyboard for the first the time, another homebrew project and you can start typing in your code Dan