Image Map Image Map
Page 1 of 13 1234511 ... LastLast
Results 1 to 10 of 127

Thread: Honeywell 200 resurrection

  1. #1

    Default Honeywell 200 resurrection

    Hi there. This is my first posting - could be my last.

    The Honeywell 200 seems to be extinct; correct me if I'm wrong. I intend to try to provide a remedy, a replica built from original Honeywell parts from around 1969, more a pastiche though really as they're too modern. I have almost enough components, but if you have any that might be suitable in your museum or even just remember what a Honeywell 200 looked like please respond.

    By the way, don't believe that the H-200 killed the IBM 1401 like they say in the histories. I think the 1401 died of natural causes.

    Rob S

  2. #2
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    25,230
    Blog Entries
    20

    Default

    I always liked the H200's futuristic (for the time) styling. You've got quite a job ahead of you.

    The 1401 (and 1410 and 1441 and the 7010) were in use well into the 1970s and many lower S/360 and S/370 systems were run in 1401 emulation mode after IBM withdrew support for the older systems. I don't think that IBM ever considered the H200 as any particular threat. Once IBM got its foot in the door of a customer, it was very difficult to dislodge.

  3. #3

    Default

    I agree with you. I have read a document written by Dr. William Gordon, head of the H-200 development team, setting out the original proposal for the design of the H-200 and in it emulating the 1401 is treated as almost an afterthought. It's just the pat historical records that call the H-200 the 1401-Killer. I understand that it did cause the IBM management to set a very rigid date for the launch of the 360 as damage-limitation though. Our company went the same internal emulation path with successive generations of Honeywell machine. I had old programmes from the H-200 days still running under not one but two levels of emulation when we were using a Honeywell model 66. If you're wise you stay with what you know best.

    The visual style of the H-200 is just one reason why there should still be one somewhere. I knew someone who had the control panel from one as a souvenir but he threw it away when he moved house. I intend to build as precise a replica of that panel as I can but I only have less than thirty of those original iconic illuminated push-buttons and the panel needs over fifty, so I'll have to improvise a bit. After doing that designing the logic should be a walk in the park.

    Rob S

  4. #4
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    25,230
    Blog Entries
    20

    Default

    Are those button caps simply acrylic? If so, it should certainly be possible to machine or cast them and fit them to contemporary illuminated pushbutton bodies.

    I never got to see a 200 up close; my first contact with Honeywell iron was via a friend who'd taken a job in Phoenix (at the old GE computer plant) in the early 70s with them. I got to take a stroll through the ill-fated water-cooled 66/85 prototype they were working on--really, just a bunch of free-standing racks sitting inside of an enclosure that you could walk into. The console the engineers were using was very impressive--Nixie-tube bar-graph displays in particular caught my eye.

  5. #5

    Default

    Yes, inch long chunks of three quarter inch square translucent plastic, don't know which kind but acrylic seems likely. The problem I see is getting or casting acrylic with the same colour and density of translucence as the originals. I assume one adds some ingredient to the clear mix to cause the effect. Perhaps a professional encapsulating firm would know how to do it. Not all the buttons need to light up, but the whole style of the panel would be ruined if they didn't all look the same and I don't want to abandon the originals entirely and use all replicas. Provided that I grouped old and new buttons separately it might look acceptable.

    Good news regarding memory for the machine. Someone in Amsterdam has offered me several 4k H-200 memory banks with supporting driver boards. That will solve one problem.

    I'm being very pragmatic about this project. At its conception I christened the planned computer "Honey Pi" after the Honeywell 200. I changed Honeywell to Honey because it won't be entirely Honeywell and the 200 to Pi because the project may be irrational and never end. Its signature demonstration programme is to compute Pi beyond the Feynman point (767 decimal places) using only 2k bytes of memory and no hardware multiply or divide, printing the result progressively while doing the calculations. I have already written and tested the programme on a PC-based emulator and the H-200 would take 14 minutes 1.9 seconds to run it to 770 places, so all I need now is to build the real computer, which will take me much longer.

    When I've made some significant progress to report I'll open the honeypi.org.uk website to document it so that I don't fill this forum with details, but building that will just add to my work.

  6. #6
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    25,230
    Blog Entries
    20

    Default

    I'd definitely talk to a plastics firm. You can tint casting resin, but it's a tricky business. Don't overlook the many model-making fora that are online--there's a wealth of information there.

    What are you going to do about peripherals for your system?

  7. #7

    Default

    My view is that the equipment should be visually interesting, hence the reason why the Pi Factory, sorry didn't mention before that's what I call my progressive Pi algorithm, starts printing immediately and not fourteen minutes after everyone has got bored and left. I originally worked on a punched card H-200, but that's impractical in our spare bedroom, so I'll try interfacing to a Dataplex machine that I have. That has two magnetic card reader/writers, you know the cards the same size as a punched card? They hold 12.5k bytes apiece, equivalent to around 150 punched cards, and when they slide back and forth through the reader with the read head going sideways to change tracks they look pretty stupid. I suppose it didn't take long for someone to make the magnetic media round, spin it around instead and call it a floppy disk. The Dataplex sits easily on a desktop, so it won't cause any domestic issues. It was designed by Englishmen. Here's the American patent application http://www.freepatentsonline.com/3860793.pdf but don't blame them for the magnetic card idea because I think that was an American invention first. Our company used the Dataplex for word processing for quite a few years. As an alternative to the Dataplex I could get really modern and use eight inch floppies, single-sided hard-sectored ones of course. I have a few packs of those and a drive.

    I haven't decided about hard copy output yet. Printing devices suffer too much wear to last. I have all the H-200 electronics to interface to an IBM Selectric golfball typewriter but I doubt that I'll find one. It's more likely that output will be to a screen of some sort until something come along. If I get as far as worrying about hard copy output no doubt somebody out there will realise that I'm being seriously successful and offer something. I've got a Wang daisy-wheel printer but who in the world understands their weird dual coaxial communication protocol now? If I want to use that I'll have to hack straight into the interface microprocessor board and bypass their connection method altogether.

    The other possible medium is paper tape, although it's no fun to handle. I have a box of sixteen reels to hand just in case. Again the machinery would be entertaining as I have a punch and readers from a Flexowriter and interfacing the H-200 to relays would be a good puzzle. One of the readers is a Selectadata device with the two speed gearbox for fast searches. I stripped down and rebuilt the punch not so long ago because the parity checking contacts were filthy. Of course the Flexowriter itself is long gone, worn out printing machinery you see.

    That's me off the grid for the evening now. We've got a big time difference between us here you know.

  8. #8
    Join Date
    Jan 2007
    Location
    Pacific Northwest, USA
    Posts
    25,230
    Blog Entries
    20

    Default

    Sign up to the cctalk mailing list--there's a lot of older gear (including TTYs and Flexowriters) lurking among the members there--and much of it on your side of the pond. The list's web page is here. Be patient--the list is sporatically administered.

    Note my avator--an NCR CRAM magnetic-card storage device. I learned to despise that beast.

  9. #9
    Join Date
    Dec 2005
    Location
    Toronto ON Canada
    Posts
    6,807

    Default

    Sounds like a worthy project for sure! I've got a couple of those mag card drives (and one card ) in case you need any parts...

  10. #10

    Default

    I took a look at cctalk but at the moment I think I should be doing some of this work that I'm promising as there's plenty here without getting more. I'll bear it in mind though so thanks for that.

    I remember hearing about those CRAM devices at my company but it was told to me more as a joke than a sign of progress. I suppose like a lot of things, including the H-200, it was great when it worked. I feel for you.

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •