Loading ...
Sorry, an error occurred while loading the content.

RE: [cosmacelf] Hardware Multiplication

Expand Messages
  • Chuck Bigham
    If you need to do incredibly serious math, there are serial interface math processors available. I’ve always wanted to have an excuse to use one of these:
    Message 1 of 20 , Jan 30, 2013
    • 0 Attachment
      If you need to do incredibly serious math, there are serial interface math processors available. I’ve always wanted to have an excuse to use one of these:
       
       
      Chuck
       
      From: William Donnelly
      Sent: ‎January‎ ‎28‎, ‎2013 ‎6‎:‎12‎ ‎PM
      To: cosmacelf@yahoogroups.com
      Subject: [cosmacelf] Hardware Multiplication
       
       

      There are probably many hardware multiplication circuits around,
      I haven't checked yet, and division, too. I just came across this and
      thought I would share. It would be a cool add-on project for an 1802.
      Since it uses (I assume/hope) fairly common non-discrete components,
      it's likely to be something that will not be unavailable for years.

      This review is from BYTE magazine, January 1976 issue, page 90.
      I don't think I have this book, and I don't remember reading it.
      I don't know if they mention the 1802 inside.

      Microprocessors : New Directions for Designers,
      edited by Edward A. Torrero,
      Hayden Book Company, 7975 Rochelle Park, New jersey.
      $8.95.

      Microprocessors is a paperback collection of reprints, originally published in Electronic Design
      from 1973 to 1975. Edward A. Torrero has put together a book aimed principally at the design
      engineer, emphasizing the practical aspects of microcomputer design.

      The first section deals with an overview of the micro's impact on industry. Representative
      articles are "Focus on Microprocessors," a truly global assessment of the current situation and
      potential of microcomputers for the design engineer; and "Smart Machines in Industrial Electronics,"
      a look at micros in numerical control and related equipment.

      The second section, "Microcomputer Basics," includes articles on microprocessor selection for
      various applications, the capabilities and various techniques of microcomputer input and output,
      the choice between random logic and micros, an analysis of micro instruction sets, some basics of
      software construction, and an excellent worked-through example of microprocessor control of traffic
      lights. This section also features a detailed look at the internal operation of, and external
      support required for, the 8008 CPU chip.

      But man is never satisfied with what he's got — the third section of Microprocessors deals with
      limitations of microcomputers and how to push them back a bit. The first article, "Speeding
      Microcomputer Multiplication," shows how to build a CPU-complementary multiplication circuit for
      the IMP-16C that allows multiplication of two 16-bit unsigned operands in 23 microseconds. Using
      only 16 standard SSI and MSI circuits, this CPU addition reduces multiplication time by a factor of
      30 from software, and by a factor of 7 from the optional National-supplied multiply instruction.


      Other articles include an investigation of the "microprogrammability" of some micros for special
      applications, construction of a system that allows a conventional mini to debug a microsystem,
      implementation of an external push-down stack for the 8008 to assist in interrupt handling, and
      specifications for providing the clock and drive signals for the 8080, something apparently
      missing from the 8080's application data.

      The articles of the fourth and last section cover an ample range of applications. The first is a
      close-up of a Motorola 6800 CPU used as a controller of a simple drum printer; the second describes
      the use of an ACIA (Asynchronous Communications Adapter) in a minicomputer context. The last three
      articles deal with uses of micros in instrumentation, phase locked loop motor control, and
      "intelligent" networks of computers.

      From this (rather exhaustive) list of articles, it should be clear that Microprocessors is
      certainly a useful book for its intended audience, the electronics design engineer. It also fills
      two other important functions: for the hobbyist hardware expert, it provides a fair overview of the
      microcomputer field; for the person more acquainted with software, it gives a good introductory
      look at the field from a hardware perspective, an introduction that I found very useful.

      Chris Ryland
      25 Fallen St.
      Cambridge MA 02138

      – Bill

       

    • bill rowe
      ooh, spi - expensive though. To: cosmacelf@yahoogroups.com From: chuck@bramblyhill.com Date: Wed, 30 Jan 2013 16:17:37 +0000 Subject: RE: [cosmacelf] Hardware
      Message 2 of 20 , Jan 30, 2013
      • 0 Attachment
        ooh, spi - expensive though.
         

        To: cosmacelf@yahoogroups.com
        From: chuck@...
        Date: Wed, 30 Jan 2013 16:17:37 +0000
        Subject: RE: [cosmacelf] Hardware Multiplication

         
        If you need to do incredibly serious math, there are serial interface math processors available. I’ve always wanted to have an excuse to use one of these:
         
         
        Chuck
         
        From: William Donnelly
        Sent: ‎January‎ ‎28‎, ‎2013 ‎6‎:‎12‎ ‎PM
        To: cosmacelf@yahoogroups.com
        Subject: [cosmacelf] Hardware Multiplication
         
         

        There are probably many hardware multiplication circuits around,
        I haven't checked yet, and division, too. I just came across this and
        thought I would share. It would be a cool add-on project for an 1802.
        Since it uses (I assume/hope) fairly common non-discrete components,
        it's likely to be something that will not be unavailable for years.

        This review is from BYTE magazine, January 1976 issue, page 90.
        I don't think I have this book, and I don't remember reading it.
        I don't know if they mention the 1802 inside.

        Microprocessors : New Directions for Designers,
        edited by Edward A. Torrero,
        Hayden Book Company, 7975 Rochelle Park, New jersey.
        $8.95.

        Microprocessors is a paperback collection of reprints, originally published in Electronic Design
        from 1973 to 1975. Edward A. Torrero has put together a book aimed principally at the design
        engineer, emphasizing the practical aspects of microcomputer design.

        The first section deals with an overview of the micro's impact on industry. Representative
        articles are "Focus on Microprocessors," a truly global assessment of the current situation and
        potential of microcomputers for the design engineer; and "Smart Machines in Industrial Electronics,"
        a look at micros in numerical control and related equipment.

        The second section, "Microcomputer Basics," includes articles on microprocessor selection for
        various applications, the capabilities and various techniques of microcomputer input and output,
        the choice between random logic and micros, an analysis of micro instruction sets, some basics of
        software construction, and an excellent worked-through example of microprocessor control of traffic
        lights. This section also features a detailed look at the internal operation of, and external
        support required for, the 8008 CPU chip.

        But man is never satisfied with what he's got — the third section of Microprocessors deals with
        limitations of microcomputers and how to push them back a bit. The first article, "Speeding
        Microcomputer Multiplication," shows how to build a CPU-complementary multiplication circuit for
        the IMP-16C that allows multiplication of two 16-bit unsigned operands in 23 microseconds. Using
        only 16 standard SSI and MSI circuits, this CPU addition reduces multiplication time by a factor of
        30 from software, and by a factor of 7 from the optional National-supplied multiply instruction.


        Other articles include an investigation of the "microprogrammability" of some micros for special
        applications, construction of a system that allows a conventional mini to debug a microsystem,
        implementation of an external push-down stack for the 8008 to assist in interrupt handling, and
        specifications for providing the clock and drive signals for the 8080, something apparently
        missing from the 8080's application data.

        The articles of the fourth and last section cover an ample range of applications. The first is a
        close-up of a Motorola 6800 CPU used as a controller of a simple drum printer; the second describes
        the use of an ACIA (Asynchronous Communications Adapter) in a minicomputer context. The last three
        articles deal with uses of micros in instrumentation, phase locked loop motor control, and
        "intelligent" networks of computers.

        From this (rather exhaustive) list of articles, it should be clear that Microprocessors is
        certainly a useful book for its intended audience, the electronics design engineer. It also fills
        two other important functions: for the hobbyist hardware expert, it provides a fair overview of the
        microcomputer field; for the person more acquainted with software, it gives a good introductory
        look at the field from a hardware perspective, an introduction that I found very useful.

        Chris Ryland
        25 Fallen St.
        Cambridge MA 02138

        – Bill

         


      • William Donnelly
        That s cool. ~$20 isn t too bad. (long-term sustainability could be an issue) Something fun to play around with or use in a project that requires speed. I wish
        Message 3 of 20 , Jan 30, 2013
        • 0 Attachment
          That's cool. ~$20 isn't too bad. (long-term sustainability could be an issue)

          Something fun to play around with or use in a project that requires speed.

          I wish they would have had those, or something like them, available back
          in the 1970's or so. Maybe they did, or if they had, they probably would
          have been really expensive, and out of the range of most hobbyists.

          That's why I always thought it would be cool/fun to interface an 1802 to
          a common and/or scientific calculator to get some nice speed and "ease" of
          access and use for some mathematical functions. I actually took apart at least
          one pocket calculator and semi-designed on paper the interfaces for "pressing keys"
          and "reading the 7-segment LED outputs". As usual, I never built it. I still plan on
          doing it, just to do it, and engage and indulge in some "old-style" hobbyism.

          I had a TI-57 I got as a high school graduation present. That's about all my parents
          could afford, and not even that, really, so I was surprised to get it. It was programmable,
          but barely so, with very small instruction memory and such, so I almost made it
          explode trying to eke out every little ounce of power trying to get it to do stuff.
          And it lost any program when you turned the power off, and other "cheap" stuff like that.

          At "my level", what I really needed was at least a TI-58, and ideally a TI-59. I would have
          been in Seventh Heaven on Cloud Nine if I had gotten one of those. But they were
          WAY out of our budget. A old-school 'friend' of mine got a decked out TI-59, I think "just because".
          I had moved, not too far away -- he was part of our "geek group" at my old HS -- he wasn't
          stupid, just... I don't what -- in his own way he had a bit of a hard family life. His TI-59 It was
          a beautiful thing. His family wasn't wealthy, but they definitely weren't poor.
          Of course, he was kind of an 'idiot' and couldn't even really use it, much less program it.
          So he let me play with it a little. He was your typical spoiled kid. An only child, too, I think.
          Or his brother(s) and/or sister(s) were MUCH older than he was and long gone.
          He got a new 1979 Hi-Performance V-8 Trans Am for graduation. It was a beautiful car.

          When I lived out in the boonies with no access to a computer or one at the school,
          my TI-57 is what I played with so I wouldn't go insane. (after having 3 years of access
          to a DG Nova 2 minicomputer at my old HS --- I'm not complaining, or anything, and at least
          I'm not resentful or bitter -- in too many ways I just had a really sad, harsh, destitute young life
          and people should feel really sorry for me ;o) )

          http://en.wikipedia.org/wiki/TI-57

          Anyway, it ended up being my interface calculator of choice, so I would have had
          trig functions and all kinds of mathematical computing power. I designed a better,
          more powerful, "real" BASIC for the 1802, and planned on using the TI-57 to do
          all of the math. It was a pipe dream, but entertaining and a learning experience.
          And more or less kept me out of trouble. (and from going out on dates and such)
          All that stuff is probably still in my papers somewhere.

          I also always wanted to interface a Selectric typewriter to an 1802, or any computer,
          to get really nice output that wasn't dot matrix. Back then that would have been a BIG thing.
          I'm sure people have done it by now, and probably even way back then. Getting the school
          Typing teacher to let me take apart and interface to one of her typewriters would have been
          a Herculean feat unto itself.

          As usual, goods times, good times.

          – Bill

          On 1/30/2013 8:17 AM, Chuck Bigham wrote:
           
          If you need to do incredibly serious math, there are serial interface math processors available. I’ve always wanted to have an excuse to use one of these:
           
           
          Chuck
        • bill rowe
          I haven t gotten to floating point yet but I don t think 32 bit multiply and divide will be much slower than integer so 25-50 ms maybe. Could you press the
          Message 4 of 20 , Jan 30, 2013
          • 0 Attachment
            I haven't gotten to floating point yet but I don't think 32 bit multiply and divide will be much slower than integer so 25-50 ms maybe.  Could you "press the keys" quickly enough to get under that? 

            I guess the calculator would have all kinds of other functions though which could blow the 1802 away because the time to iterate to a sqrt or sin or whatever would overwhelm the keying time.

            In the early eighties I acquired a mechanical nightmare called a flexowriter(?) with an eye to spiffy I/O for my elf but the project never went anywhere.  I don't know what I did with it in the end but it was a beast - I bet it weighed 100 lbs.


            To: cosmacelf@yahoogroups.com
            From: william@...
            Date: Wed, 30 Jan 2013 13:08:30 -0800
            Subject: Re: [cosmacelf] Hardware Multiplication

             

            That's cool. ~$20 isn't too bad. (long-term sustainability could be an issue)

            Something fun to play around with or use in a project that requires speed.

            I wish they would have had those, or something like them, available back
            in the 1970's or so. Maybe they did, or if they had, they probably would
            have been really expensive, and out of the range of most hobbyists.

            That's why I always thought it would be cool/fun to interface an 1802 to
            a common and/or scientific calculator to get some nice speed and "ease" of
            access and use for some mathematical functions. I actually took apart at least
            one pocket calculator and semi-designed on paper the interfaces for "pressing keys"
            and "reading the 7-segment LED outputs". As usual, I never built it. I still plan on
            doing it, just to do it, and engage and indulge in some "old-style" hobbyism.

            I had a TI-57 I got as a high school graduation present. That's about all my parents
            could afford, and not even that, really, so I was surprised to get it. It was programmable,
            but barely so, with very small instruction memory and such, so I almost made it
            explode trying to eke out every little ounce of power trying to get it to do stuff.
            And it lost any program when you turned the power off, and other "cheap" stuff like that.

            At "my level", what I really needed was at least a TI-58, and ideally a TI-59. I would have
            been in Seventh Heaven on Cloud Nine if I had gotten one of those. But they were
            WAY out of our budget. A old-school 'friend' of mine got a decked out TI-59, I think "just because".
            I had moved, not too far away -- he was part of our "geek group" at my old HS -- he wasn't
            stupid, just... I don't what -- in his own way he had a bit of a hard family life. His TI-59 It was
            a beautiful thing. His family wasn't wealthy, but they definitely weren't poor.
            Of course, he was kind of an 'idiot' and couldn't even really use it, much less program it.
            So he let me play with it a little. He was your typical spoiled kid. An only child, too, I think.
            Or his brother(s) and/or sister(s) were MUCH older than he was and long gone.
            He got a new 1979 Hi-Performance V-8 Trans Am for graduation. It was a beautiful car.

            When I lived out in the boonies with no access to a computer or one at the school,
            my TI-57 is what I played with so I wouldn't go insane. (after having 3 years of access
            to a DG Nova 2 minicomputer at my old HS --- I'm not complaining, or anything, and at least
            I'm not resentful or bitter -- in too many ways I just had a really sad, harsh, destitute young life
            and people should feel really sorry for me ;o) )

            http://en.wikipedia.org/wiki/TI-57

            Anyway, it ended up being my interface calculator of choice, so I would have had
            trig functions and all kinds of mathematical computing power. I designed a better,
            more powerful, "real" BASIC for the 1802, and planned on using the TI-57 to do
            all of the math. It was a pipe dream, but entertaining and a learning experience.
            And more or less kept me out of trouble. (and from going out on dates and such)
            All that stuff is probably still in my papers somewhere.

            I also always wanted to interface a Selectric typewriter to an 1802, or any computer,
            to get really nice output that wasn't dot matrix. Back then that would have been a BIG thing.
            I'm sure people have done it by now, and probably even way back then. Getting the school
            Typing teacher to let me take apart and interface to one of her typewriters would have been
            a Herculean feat unto itself.

            As usual, goods times, good times.

            – Bill

            On 1/30/2013 8:17 AM, Chuck Bigham wrote:
             
            If you need to do incredibly serious math, there are serial interface math processors available. I’ve always wanted to have an excuse to use one of these:
             
             
            Chuck

          • David W. Schultz
            ... It looks like a micro-controller, possibly of the PIC family, programmed to do math and other things. I wouldn t consider that useful for even serious
            Message 5 of 20 , Jan 30, 2013
            • 0 Attachment
              On 01/30/2013 10:17 AM, Chuck Bigham wrote:
              >
              >
              > If you need to do incredibly serious math, there are serial interface
              > math processors available. I’ve always wanted to have an excuse to use
              > one of these:
              >
              > http://www.micromegacorp.com/umfpu-v3.html
              >

              It looks like a micro-controller, possibly of the PIC family, programmed
              to do math and other things.


              I wouldn't consider that useful for even serious math. It will do a
              floating point (32 bit) multiply in 9 microseconds. Which is slow
              except in comparison to the 1802. I picked up a very inexpensive STM32F4
              Discovery board (aka Cortex-M4) to play with last year and it will do
              that in 1 clock cycle at 168Mhz or 6 nanoseconds.



              --
              David W. Schultz
              http://home.earthlink.net/~david.schultz
              Returned for Regrooving
            • David W. Schultz
              ... That reminded me of something so I dug into the archives and found a photocopy of a Radio-Electronics article from December 1978 titled NOM Card For The
              Message 6 of 20 , Jan 30, 2013
              • 0 Attachment
                On 01/30/2013 03:08 PM, William Donnelly wrote:
                >
                >
                > That's cool. ~$20 isn't too bad. (long-term sustainability could be an
                > issue)
                >
                > Something fun to play around with or use in a project that requires speed.
                >
                > I wish they would have had those, or something like them, available back
                > in the 1970's or so. Maybe they did, or if they had, they probably would
                > have been really expensive, and out of the range of most hobbyists.
                >

                That reminded me of something so I dug into the archives and found a
                photocopy of a Radio-Electronics article from December 1978 titled "NOM
                Card For The 1802".

                It interfaced the $18 MM57109 chip to the 1802.

                --
                David W. Schultz
                http://home.earthlink.net/~david.schultz
                Returned for Regrooving
              • William Donnelly
                Whatever speeds you come up with for software routines will be fine , because it is what it is. The calculator key entry and display output access would
                Message 7 of 20 , Jan 30, 2013
                • 0 Attachment
                  Whatever speeds you come up with for software routines will be "fine", because
                  it is what it is.

                  The calculator key entry and display output access would definitely be considerations
                  when interfacing to a calculator. I never got to the point of figuring out what
                  the speeds would be. And there are other considerations, like: when is the
                  output valid? ; how fast can you 'type'? ; etc.

                  You don't see it much these days, but many of the old calculators would blank
                  the display and "eventually" show you a result for the more complex functions,
                  like Square Root, trig and "anti-trig" functions, logarithms, X^Y, etc. Now-a-days
                  it would probably drive people crazy, but back then it was more than fast enough.
                  Amazingly, magically fast, actually. (computers speeds and other things like that
                  are messing with our brains, which is probably not the best thing, and might lead
                  to issues like problems dealing with the stress of our increasingly fast-paced world)

                  Although I did know a few people who could beat a person using a calculator with
                  a slide rule. (either always, or over the short run) And some abacus users can beat
                  calculators and computers of today, although that usually has to take into account
                  the speed of data entry of the calculator user / keyboardist.

                  Slide rules were JUST slightly before my time, so I never really learned to use them.
                  My older brother did, though, because there basically was no such animal as a
                  hand calculator in his day, and definitely not a portable hand calculator. Unless maybe
                  if you had A LOT of money. I hear stories and read histories that make me go, "Hmm...",
                  because sometimes I definitely don't remember the world being like "that". But there
                  are things you have to take into account, like geographical locations, socio-economic
                  statuses, whether and how much one was a "techy" and interested and aware and
                  well-read about those things, what your father did for a living, etc... But even nice
                  "scientific" desk calculators were pretty expensive. Really nice slide rules weren't cheap.
                  I think a slide rule was my brother's 8th grade graduation present for HS.

                  I used to work at a print shop when I was just starting out working post-HS, and did a lot of
                  typesetting on a photo-typesetting machine (early "computerized" word processor-like
                  machines before there were "real" word processing programs, and before/during hard-wired
                  (expensive) WP systems, although there were all kinds, because it was a burgeoning industry).

                  And, of course, ink jet and laser printers were just a gleam in someone's eye. Or they were
                  way too expensive for a small business to own. I don't think the print shop even owned
                  their typesetter, I think it was leased. (remember back when you didn't OWN your home
                  telephone; Ma Bell, or Ohio Bell, or the service company in your area owned it and was
                  effectively leasing it to you as part of the telephone service, and it was actually illegal
                  (or so we were told, or it might have been an urban legend) to plug in a phone that was
                  not theirs and officially sanctioned, etc...?)

                  I was chosen to typeset mostly because I could type, type (relatively) quickly, be accurate,
                  and was just generally good at using any kind of electronic or computerized hardware,
                  and my end results tended to be high quality. I could type pretty fast with few errors.
                  I forget what the average was, but at my best I was at least 50 to 60 WPM, probably
                  faster, and even faster in bursts. Even today, with my reverted "hunt and peck"-like, pseudo-correct,
                  too-many-years-typing-as-a-programmer style, I can still type pretty fast with few errors,
                  if I have the right keyboard. I've gotten more than one "Geez" 's when people have seen
                  me type. And I think I type too slow anymore.

                  Anyway, if there was a really long text, they would hire-in a "professional keyboardist",
                  who could type well in excess of 100 WPM. I think he was rated over 150, but I may be
                  misremembering. It's amazing to see someone like that type, though. Very mind-boggling.
                  You almost can't follow their fingers. They fly across the keyboard and almost look like a blur.
                  They aren't really even "thinking" -- they just type what they see. In and Out. Now we know
                  a lot of that is "muscle memory", and some of it is "stored in the spinal column". And, like
                  pretty much all things, it also takes a person with a certain kind of built-in talent to be really good at it.

                  I've seen some "secretaries" who can zoom on desktop adding machines, too. At one place
                  I worked, the calculator died, and you couldn't get that model anymore, and she couldn't
                  find a used one, or one with the same keyboard layout. She was highly upset and her data entry
                  speed went way down, as did her error rate go up. I don't think she ever really recovered from that.

                  My keyboard died several months ago from a drink spill, which has NEVER happened to me
                  in near 30 years of computer work. I used to silently berate people who spilled stuff on their
                  keyboards. What the Hell were they thinking / doing??? Then it happened to me AGAIN a few
                  months later! So I bought a cheaper, nice keyboard, that is water-resistant.

                  Although you can more or less adapt to any keyboard, some are better than others, some much better.
                  It's amazing how a millimeter, or so, or fraction thereof, difference in the spacing of keys, horizontally
                  and/or vertically,  can make you faster or slower, have more or less errors, and be more or less
                  productive. So now I'm in search of a really good keyboard. Plus there's the feel of the key
                  when it's pressed, the reaction to being pressed, how hard it has to be pressed, the sound
                  it makes, etc. And some other factors.

                  Flexowriters were very cool. Only uppercase alpha characters, though.
                  They were basically Teletypes. I guess several companies made teletype competitors
                  and such. I want to have at least a Teletype, and ideally a PTR and punched
                  card reader. A card punch like I used to have, too, if I ever get the space and money.
                  But it's getting so difficult and expensive to have those worked on now. And they have
                  a tendency to go out of whack if you look at them the wrong way. I remember trying to help
                  work on our teletypes and PTR and some other hardware when I was in HS to try to save
                  money from our non-existent budget. (I think our teacher paid for much of it from his own
                  pocket, god bless his soul and all that) But really complex mechanical and electro-mechanical
                  machines tend to freak me out a little, so I've never been good at understanding them
                  and working on them. I've also noticed that my 3D spacial "thing" isn't particularly strong
                  for some reason. I guess it's just naturally one of my weak points.

                  Sorry I'm writing such long e-mails. I'm flu-fuzzy and such. It could be worse, it could be
                  required reading and their could be a required test at the end.

                  – Bill

                  On 1/30/2013 1:27 PM, bill rowe wrote:
                   

                  I haven't gotten to floating point yet but I don't think 32 bit multiply and divide will be much slower than integer so 25-50 ms maybe.  Could you "press the keys" quickly enough to get under that? 

                  I guess the calculator would have all kinds of other functions though which could blow the 1802 away because the time to iterate to a sqrt or sin or whatever would overwhelm the keying time.

                  In the early eighties I acquired a mechanical nightmare called a flexowriter(?) with an eye to spiffy I/O for my elf but the project never went anywhere.  I don't know what I did with it in the end but it was a beast - I bet it weighed 100 lbs.

                • ted_rossin
                  I agree. The pin names sound very PIC like. The 18F series PICs have a single instruction cycle 8x8 hardware multiplier. With a tiny amount of code you can
                  Message 8 of 20 , Jan 31, 2013
                  • 0 Attachment
                    I agree. The pin names sound very PIC like.

                    The 18F series PICs have a single instruction cycle 8x8 hardware multiplier. With a tiny amount of code you can implement 16x16 or 32x32 faster than it takes to transfer the operands into and results out of the chip. There are some tiny parts (18F1330) that also have the multipliers and they do have I2C interfaces. Interfacing to the 1802 through the ports using 8 bits would be quicker but will require some glue logic. Either way, the PICs cost about $4 to $6. The programmer will set you back $35 however.

                    --- In cosmacelf@yahoogroups.com, "David W. Schultz" wrote:
                    >
                    > On 01/30/2013 10:17 AM, Chuck Bigham wrote:
                    > >
                    > >
                    > > If you need to do incredibly serious math, there are serial interface
                    > > math processors available. I’ve always wanted to have an excuse to use
                    > > one of these:
                    > >
                    > > http://www.micromegacorp.com/umfpu-v3.html
                    > >
                    >
                    > It looks like a micro-controller, possibly of the PIC family, programmed
                    > to do math and other things.
                    >
                    >
                    > I wouldn't consider that useful for even serious math. It will do a
                    > floating point (32 bit) multiply in 9 microseconds. Which is slow
                    > except in comparison to the 1802. I picked up a very inexpensive STM32F4
                    > Discovery board (aka Cortex-M4) to play with last year and it will do
                    > that in 1 clock cycle at 168Mhz or 6 nanoseconds.
                    >
                    >
                    >
                    > --
                    > David W. Schultz
                    > http://home.earthlink.net/~david.schultz
                    > Returned for Regrooving
                    >
                  • urrossum@att.net
                    ... I also had this calculator (I had to buy it myself, with money from a student job in a lab on campus); no way could I afford the 58 or 59. At the time,
                    Message 9 of 20 , Jan 31, 2013
                    • 0 Attachment
                      > I had a TI-57 I got as a high school graduation present. That's about
                      > all my parents
                      > could afford, and not even that, really, so I was surprised to get it.
                      > It was programmable,

                      I also had this calculator (I had to buy it myself, with money from a student job in a lab on campus); no way could I afford the 58 or 59. At the time, though, I was impressed that it had most of the fundamental constructs needed for actual programming, though as you said it was extremely limited (it did have indirect references, though, so you could sort of do arrays - with the 8 available registers!)

                      > And it lost any program when you turned the power off, and other "cheap"
                      > stuff like that.

                      Yeah - that part sucked.

                      > Anyway, it ended up being my interface calculator of choice, so I would

                      I worked as a summer job at a company that made the first programmable implantable pacemaker. It was programmed through the skin by using an inductive pendant that would couple to a coil in the pacemaker. Data was sent to the pacemaker by pulsing a sine-wave carrier. By alternately shorting and opening the coil in the pacemaker while a continuous wave was emitted by the pendant, a "downlink" channel was implemented for pacemaker-to-programmer communications, without requiring any power from the pacemaker's extremely limited battery. I don't know what the processor was, but I sure wouldn't be surprised to find that it was an 1802 (this was in ~1978).

                      The programming unit was battery powered, because there were very strict requirements about anything that was plugged into the wall in the operating room. It also didn't have a printer of any sort, so to get a printed record from the device (remember, this was before PCs even existed), the programmer (device, not person!) was carried out of the operating room, and plunked down on a TI57/58/59 printer! This charged the batteries in the programmer, and more importantly provided a printed record of the setup of the pacemaker, and its status (battery level, etc.).

                      As a result, this company (PSI - Pacesetter Systems, Inc.) requested, and got, the protocol needed to send data to the printer. (I managed to get a copy of this protocol (which I still have, scanned, if anyone wants it [Bill!]), and always intended to use it to control "other stuff" from my 57. Never got around to it, though...

                      This wouldn't have solved the problem of remotely "pushing the keys", but it would have made it much easier to get the results back to the 1802. The interface is pretty simple: two streams of serial data, one for control and one for 6-bit character data, and a bi-phase clock. A BUSY line allows the calculator's output to be paced.

                      > I also always wanted to interface a Selectric typewriter to an 1802, or
                      > any computer,
                      > to get really nice output that wasn't dot matrix. Back then that would
                      > have been a BIG thing.

                      When I was a senior in college, I lucked into a Xerox Diablo - a truly fantastic (for the time) daisy-wheel printer. Continuous 30 CPS (the Selectrics could only do 15 CPS), plus micro-positioning so that I could plot graphs (or, in theory, do true proportional print and kerning of the output). Later on, I had a CP/M system with WordStar that made terrific use of this printer.

                      I'd still like to land a TeleType KSR/ASR-33, though. I had one of those too, in perfect working and cosmetic order, which I abandoned to rust away in a garage in San Pedro because I didn't have room for it in my new place. What an idiot!

                      > As usual, goods times, good times.

                      Yup.
                      ~~
                      Mark Moulding
                    • William Donnelly
                      That s cool. That printer was one of the crappiest printers ever. But at the time it was magnificent. Just getting hardcopy output was cool. (and they had
                      Message 10 of 20 , Feb 1, 2013
                      • 0 Attachment
                        That's cool.

                        That printer was one of the "crappiest" printers ever.
                        But at the time it was magnificent. Just getting hardcopy output was cool.
                        (and they had those little magnetic cards for loading and saving data and programs;
                        I always thought those would be cool to have for an 1802 so you could save and
                        load your programs; but the mechanism would be 'infinitely' more expensive than
                        the 1802 system itself; I guess some of that older-tech magnetic bar scanner stuff
                        might be somewhat cheap now, ore even the newer stuff, relatively speaking)

                        I think the printer was a 'static' (?) dot-matrix printer or whatever they were called
                        that used that glossy-slick-oily "adding machine" paper that was once used (and still
                        is for some applications) for some desktop calculators, and cash registers, and such.
                        I later heard that there was something in/on that paper that was carcinogenic,
                        unless I'm misremembering or mixing it up with something else.
                        Which is partly why they quit using it. (for the most part)
                        I don't know if some of the stuff they use now is the same, or similar, or what.

                        If you have that interface info, I would like a copy, or maybe put it on the
                        downloads somewhere. Unless it is already online somewhere, then you could link to it.
                        Too much of that stuff has disappeared into the ether, never to be seen again.

                        My printer died last year and I'm going to get a flat bed X-in-1 so I can scan stuff.
                        My brother has a really nice high-resolution graphics designer-like scanner, so
                        I can always use it for stuff that needs to be really nice resolution. In case someone
                        needs something nicely-scanned. It's kind of an art and a science to get a nice scan
                        out of something that has some quality to it.

                        After I finally get over this flu thing, I hope to get back into doing stuff and having
                        something to show for my work. I really need to get my MC built so I can play
                        with it. I wanted to make a mini-doc on how to do it from start to finish, but I'm
                        probably going to have to give up on that, or it will never get done. My brother
                        and sister are getting a new DSLR video camera to make a documentary with, so
                        maybe I'll try to use that and just hunker down and do it. It would be pretty cool
                        to see it being built from beginning to end, from parts to finished product.
                        I was hoping that would encourage others to do it when they saw how "easy"
                        it is to do it. Plus a video that goes step-by-step would make it much less intimidating
                        for the beginner electronic project-maker.

                        – Bill

                        On 1/31/2013 2:03 PM, urrossum@... wrote:

                        > I had a TI-57 I got as a high school graduation present. That's about
                        > all my parents
                        > could afford, and not even that, really, so I was surprised to get it.
                        > It was programmable,

                        I also had this calculator (I had to buy it myself, with money from a student job in a lab on campus); no way could I afford the 58 or 59. At the time, though, I was impressed that it had most of the fundamental constructs needed for actual programming, though as you said it was extremely limited (it did have indirect references, though, so you could sort of do arrays - with the 8 available registers!)

                        > And it lost any program when you turned the power off, and other "cheap"
                        > stuff like that.

                        Yeah - that part sucked.

                        > Anyway, it ended up being my interface calculator of choice, so I would

                        I worked as a summer job at a company that made the first programmable implantable pacemaker. It was programmed through the skin by using an inductive pendant that would couple to a coil in the pacemaker. Data was sent to the pacemaker by pulsing a sine-wave carrier. By alternately shorting and opening the coil in the pacemaker while a continuous wave was emitted by the pendant, a "downlink" channel was implemented for pacemaker-to-programmer communications, without requiring any power from the pacemaker's extremely limited battery. I don't know what the processor was, but I sure wouldn't be surprised to find that it was an 1802 (this was in ~1978).

                        The programming unit was battery powered, because there were very strict requirements about anything that was plugged into the wall in the operating room. It also didn't have a printer of any sort, so to get a printed record from the device (remember, this was before PCs even existed), the programmer (device, not person!) was carried out of the operating room, and plunked down on a TI57/58/59 printer! This charged the batteries in the programmer, and more importantly provided a printed record of the setup of the pacemaker, and its status (battery level, etc.).

                        As a result, this company (PSI - Pacesetter Systems, Inc.) requested, and got, the protocol needed to send data to the printer. (I managed to get a copy of this protocol (which I still have, scanned, if anyone wants it [Bill!]), and always intended to use it to control "other stuff" from my 57. Never got around to it, though...

                        This wouldn't have solved the problem of remotely "pushing the keys", but it would have made it much easier to get the results back to the 1802. The interface is pretty simple: two streams of serial data, one for control and one for 6-bit character data, and a bi-phase clock. A BUSY line allows the calculator's output to be paced.

                        > I also always wanted to interface a Selectric typewriter to an 1802, or
                        > any computer,
                        > to get really nice output that wasn't dot matrix. Back then that would
                        > have been a BIG thing.

                        When I was a senior in college, I lucked into a Xerox Diablo - a truly fantastic (for the time) daisy-wheel printer. Continuous 30 CPS (the Selectrics could only do 15 CPS), plus micro-positioning so that I could plot graphs (or, in theory, do true proportional print and kerning of the output). Later on, I had a CP/M system with WordStar that made terrific use of this printer.

                        I'd still like to land a TeleType KSR/ASR-33, though. I had one of those too, in perfect working and cosmetic order, which I abandoned to rust away in a garage in San Pedro because I didn't have room for it in my new place. What an idiot!

                        > As usual, goods times, good times.

                        Yup.
                        ~~
                        Mark Moulding


                      • urrossum@att.net
                        ... I keep seeing small 20- or 30-column thermal printer mechanisms on the surplus market for cheap; I bought one a while back, and it looked like it would be
                        Message 11 of 20 , Feb 1, 2013
                        • 0 Attachment
                          > I always thought those would be cool to have for an 1802 so you could
                          > save and
                          > load your programs; but the mechanism would be 'infinitely' more
                          > expensive than
                          > the 1802 system itself; I guess some of that older-tech magnetic bar
                          > scanner stuff
                          > might be somewhat cheap now, ore even the newer stuff, relatively speaking)

                          I keep seeing small 20- or 30-column thermal printer mechanisms on the surplus market for cheap; I bought one a while back, and it looked like it would be pretty easy to use. That would actually be a pretty good fit for a little 1802 system; good for assembly listings, too.
                          >
                          > I think the printer was a 'static' (?) dot-matrix printer or whatever
                          > they were called
                          > that used that glossy-slick-oily "adding machine" paper that was once
                          > used (and still
                          > is for some applications) for some desktop calculators, and cash
                          > registers, and such.

                          No, this was a simple thermal printer mechanism. I've used something in the past ("Findex" portable computer: http://tinyurl.com/bee2zsk ) with that glossy paper. As nearly as I can tell, it's conductive, and the pixels are made by blasting a little spark ("static") through the top surface of the paper.

                          > If you have that interface info, I would like a copy, or maybe put it on the
                          > downloads somewhere. Unless it is already online somewhere, then you
                          > could link to it.
                          > Too much of that stuff has disappeared into the ether, never to be seen
                          > again.

                          I pretty sure that this data exists nowhere else on the Internet, as I scanned it myself from a faded photocopy of a photocopy. I couldn't see any easy way to put it up on bitsavers.org, so I've uploaded it to the root of the Files section. I don't believe this should ultimately live there, since it's not 1802-related, but maybe it will propagate to somewhere useful...

                          > My printer died last year and I'm going to get a flat bed X-in-1 so I
                          > can scan stuff.

                          Those are dirt cheap now; for less than $100 you can get one with flatbed and sheet feeder, wireless networking, etc. Check out the reviews carefully though - some of them (like the Brothers) use up ink even when you're not printing anything, and the ink is where the real cost lies.
                          ~~
                          Mark Moulding
                        • Ray Sills
                          I remember that once upon a time there was a printer that used a silver-like coating on the paper (over a black layer) so that as the printing head passed over
                          Message 12 of 20 , Feb 1, 2013
                          • 0 Attachment
                            I remember that once upon a time there was a printer that used a
                            silver-like coating on the paper (over a black layer) so that as the
                            printing head passed over a print point (pixel) a small current would
                            zap the silvery stuff, revealing the black color underneath. It had a
                            fairly nice dot-matrix type font, and it may have been able to do
                            modest graphics.

                            The best thing about it was that the reflectivity of the silver
                            coating was so high, that it made -perfect- photocopies... and the
                            photocopies were usually easier to read than the original.

                            There was a software author, Paul Pieseck, (I think that's his name)
                            who was the owner of Cuddly Software. We wrote an operating system
                            for the 1802, plus a number of utility programs. When the business
                            foundered, and he decided to close up shop, he sent me a bunch of his
                            documents and tape files, and released it all to the pubic domain.
                            Anyway, Paul used that silvery paper in his printer.

                            73 de Ray
                          • Ray Sills
                            Make that ... Make that: HE wrote an operating system for the 1802 ..... etc. 73 de Ray
                            Message 13 of 20 , Feb 1, 2013
                            • 0 Attachment
                              Make that
                              On Feb 1, 2013, at 5:42 PM, Ray Sills wrote:
                              > <snip>

                              > There was a software author, Paul Pieseck, (I think that's his name)
                              > who was the owner of Cuddly Software. We wrote an operating system
                              > for the 1802, plus a number of utility programs.

                              Make that: HE wrote an operating system for the 1802 ..... etc.

                              73 de Ray
                            • William Donnelly
                              That s right, thermal. Those would be kind of nice to mess around with for the 1802, but I (more) meant the mag card reader-writers, in case that wasn t clear.
                              Message 14 of 20 , Feb 1, 2013
                              • 0 Attachment
                                That's right, thermal. Those would be kind of nice to mess around with for
                                the 1802, but I (more) meant the mag card reader-writers, in case that wasn't clear.
                                Even though it is fairly old tech these days, it would still be retro and cool to be
                                able to run a tiny mag card through a reader to load your programs, and to save them
                                and data, etc. It's on my ToDo list, but toward the bottom. For those young and old
                                who aren't familiar with this, check out wikipedia. They even had a video of reading
                                the mag cards. http://en.wikipedia.org/wiki/TI-59_/_TI-58

                                I did "design" a printer for the 1802 that used paper cash register tape and felt tip pins
                                and solenoids and motors to make a "cheap" 1802 printer. Unless you got all
                                of the parts from surplus houses, it probably wouldn't be non-expensive, though.
                                And it wouldn't be a very fast printer, even by CRAV standards.
                                I used to love getting those surplus electronics catalogs in the mail every month and
                                then pouring over them to see what kind of cool things they had, and imagining
                                what I could make with them. Later, I resurrected the design to design a (Calif) Lotto
                                paper slip printer. (never built) Kind of overkill, but would be cool. You type in your
                                numbers and it fills in the play slap for you.

                                Thanks for the reminder about printer caveats.
                                It always seems I buy the printer whose ink cartridges cost more than
                                any others for any other printer. So now I look for that.

                                My sister has a Canon PIXMA PM210, and you can't not only not print at
                                all if your color cartridge is empty, that is, can't print in black and white until
                                you can get a new color cartridge, BUT, you cannot do black and white copies
                                OR even scan something on the scanner. (or send/receive FAXes) WTF???

                                I like to give the designers the benefit of the doubt that they didn't
                                purposely design it that way to 'sell more ink cartridges', but you never know.
                                That is how they "make their money" and justify / rationalize selling printers
                                for sub-$100 prices. Plus being made in Asia, usually.

                                It's like that one car that you can't change the front spark plug unless you remove
                                the front tire, and another you have to actually jack the engine out of the car.
                                I've heard of a couple of others like that, like having to remove the oil pump, etc.
                                Which is why car designers should be forced work on the cars they design,
                                just to make sure something like that doesn't happen. Supposedly they were
                                "accidents of design". Which is plausible. Or at least plausible deniability.
                                I had a car once that you had to remove the air conditioner pump and unloosen
                                the alternator and swing it out of the way to replace some other unit.
                                I forget what it was and which car that was.

                                – Bill

                                On 2/1/2013 1:06 PM, urrossum@... wrote:
                                 



                                > I always thought those would be cool to have for an 1802 so you could
                                > save and
                                > load your programs; but the mechanism would be 'infinitely' more
                                > expensive than
                                > the 1802 system itself; I guess some of that older-tech magnetic bar
                                > scanner stuff
                                > might be somewhat cheap now, ore even the newer stuff, relatively speaking)

                                I keep seeing small 20- or 30-column thermal printer mechanisms on the surplus market for cheap; I bought one a while back, and it looked like it would be pretty easy to use. That would actually be a pretty good fit for a little 1802 system; good for assembly listings, too.
                                >
                                > I think the printer was a 'static' (?) dot-matrix printer or whatever
                                > they were called
                                > that used that glossy-slick-oily "adding machine" paper that was once
                                > used (and still
                                > is for some applications) for some desktop calculators, and cash
                                > registers, and such.

                                No, this was a simple thermal printer mechanism. I've used something in the past ("Findex" portable computer: http://tinyurl.com/bee2zsk ) with that glossy paper. As nearly as I can tell, it's conductive, and the pixels are made by blasting a little spark ("static") through the top surface of the paper.

                                > If you have that interface info, I would like a copy, or maybe put it on the
                                > downloads somewhere. Unless it is already online somewhere, then you
                                > could link to it.
                                > Too much of that stuff has disappeared into the ether, never to be seen
                                > again.

                                I pretty sure that this data exists nowhere else on the Internet, as I scanned it myself from a faded photocopy of a photocopy. I couldn't see any easy way to put it up on bitsavers.org, so I've uploaded it to the root of the Files section. I don't believe this should ultimately live there, since it's not 1802-related, but maybe it will propagate to somewhere useful...

                                > My printer died last year and I'm going to get a flat bed X-in-1 so I
                                > can scan stuff.

                                Those are dirt cheap now; for less than $100 you can get one with flatbed and sheet feeder, wireless networking, etc. Check out the reviews carefully though - some of them (like the Brothers) use up ink even when you're not printing anything, and the ink is where the real cost lies.
                                ~~
                                Mark Moulding


                              • Rick Cortese
                                I had a TI58, I didn t dare ask for a TI59 because I had previously burned most of my bridges with IT by the time they came out but the guy I shared an office
                                Message 15 of 20 , Feb 1, 2013
                                • 0 Attachment
                                  I had a TI58, I didn't dare ask for a TI59 because I had previously burned most of my bridges with IT by the time they came out but the guy I shared an office with didn't have any such qualms: He ordered a TI59 and the printer. We really used the heck out of them. Pharmaceutical research so we wrote all sorts of programs for data reduction and hard copy print out of everything from Beer's Law plots to release rate data. I recall my office mate reduced his note taking to just taping TI printouts into his laboratory notebook which saved time but he ended up with a lab notebooks that were triple the thickness of a normal notebook.

                                  We also had a ton of those Diablo daisy wheel printers. They were actually much more then just a printer, they functioned as a full terminal. The IT department set us up with Hayes modems, probably 300 baud, and we would dial into the DEC mainframe using the Diablo printers. The graphics programs and print outs with the Diablo were tedious at best. Quality and accuracy was OK but at 30 cps it took a long time to print out a graph.

                                  One of the oddball coolest things about the Diablo: One of my friends in IT, yes, I managed to not offend everyone in the department, had a Diablo set up in her house so she could access the mainframe from home. One night I was over there for diner and she set my son up on the terminal with Hunt the Wumpus playing on the mainframe. The printer was mounted on a wheeled cart so every time my son shot an arrow the printer would do a little song and dance
                                  The arrow bounces<cr>
                                  The arrow bounces<cr>
                                  The arrow bounces<cr>


                                  From: William Donnelly <william@...>
                                  To: cosmacelf@yahoogroups.com
                                  Sent: Friday, February 1, 2013 3:32 PM
                                  Subject: Re: [cosmacelf] Re: Hardware Multiplication

                                   
                                  That's right, thermal. Those would be kind of nice to mess around with for
                                  the 1802, but I (more) meant the mag card reader-writers, in case that wasn't clear.
                                  Even though it is fairly old tech these days, it would still be retro and cool to be
                                  able to run a tiny mag card through a reader to load your programs, and to save them
                                  and data, etc. It's on my ToDo list, but toward the bottom. For those young and old
                                  who aren't familiar with this, check out wikipedia. They even had a video of reading
                                  the mag cards. http://en.wikipedia.org/wiki/TI-59_/_TI-58 wrote:
                                   


                                  > I always thought those would be cool to have for an 1802 so you could
                                  > save and
                                  > load your programs; but the mechanism would be 'infinitely' more
                                  > expensive than
                                  > the 1802 system itself; I guess some of that older-tech magnetic bar
                                  > scanner stuff
                                  > might be somewhat cheap now, ore even the newer stuff, relatively speaking)

                                  I keep seeing small 20- or 30-column thermal printer mechanisms on the surplus market for cheap; I bought one a while back, and it looked like it would be pretty easy to use. That would actually be a pretty good fit for a little 1802 system; good for assembly listings, too.
                                  >
                                  > I think the printer was a 'static' (?) dot-matrix printer or whatever
                                  > they were called
                                  > that used that glossy-slick-oily "adding machine" paper that was once
                                  > used (and still
                                  > is for some applications) for some desktop calculators, and cash
                                  > registers, and such.

                                  No, this was a simple thermal printer mechanism. I've used something in the past ("Findex" portable computer: http://tinyurl.com/bee2zsk ) with that glossy paper. As nearly as I can tell, it's conductive, and the pixels are made by blasting a little spark ("static") through the top surface of the paper.

                                  > If you have that interface info, I would like a copy, or maybe put it on the
                                  > downloads somewhere. Unless it is already online somewhere, then you
                                  > could link to it.
                                  > Too much of that stuff has disappeared into the ether, never to be seen
                                  > again.

                                  I pretty sure that this data exists nowhere else on the Internet, as I scanned it myself from a faded photocopy of a photocopy. I couldn't see any easy way to put it up on bitsavers.org, so I've uploaded it to the root of the Files section. I don't believe this should ultimately live there, since it's not 1802-related, but maybe it will propagate to somewhere useful...

                                  > My printer died last year and I'm going to get a flat bed X-in-1 so I
                                  > can scan stuff.

                                  Those are dirt cheap now; for less than $100 you can get one with flatbed and sheet feeder, wireless networking, etc. Check out the reviews carefully though - some of them (like the Brothers) use up ink even when you're not printing anything, and the ink is where the real cost lies.
                                  ~~
                                  Mark Moulding



                                • Richard
                                  Cuddly Software? I have two programs from him. One is a game called Starship and the other one is a debugger called CSTP. I even have CSTP still on its
                                  Message 16 of 20 , Feb 1, 2013
                                  • 0 Attachment
                                    Cuddly Software? I have two programs from him. One is a game called 'Starship' and the other one is a debugger called CSTP. I even have CSTP still on its original tape (Scotch Highlander).

                                    At some time my ASCII keyboard was broken. I had accidentally messed up the voltages and killed the keyboard encoder. Without it I could not use CSTP anymore. so I tried to modify it to take its input from the hex keyboard.

                                    It did not work. Finding all bytes wich might be input instructions was easy enough, but I had no disassembler or even a listing and changed some bytes which were no input instructions.

                                    That's why I decided to call him and ask for some help. He talked me through the modifications and sent me a hex dump of CSTP. On the silver thermo paper you mentioned.

                                    Some time later my poor parents got the telephone bill and I got into some trouble. They grounded me and made me spend some time in my room with the Elf and CSTP. Harsh punishment that really made me feel sorry for what I did :-)

                                    CSTP was the first software I bought and I still use it today whenever I write something for the Elf. I load it into the emulator and then can debug my program with it. I always wanted to post it to the files section, but did not know where to ask for permission. Is there a chance that it's ok?
                                  • Ray Sills
                                    HI Richard: Ah yes, CSTP!! I had forgotten the name. Paul did some nice programming. I don t know what he did after he closed down Cuddly Software, but
                                    Message 17 of 20 , Feb 2, 2013
                                    • 0 Attachment
                                      HI Richard:

                                      Ah yes, CSTP!! I had forgotten the name. Paul did some nice
                                      programming. I don't know what he did after he closed down Cuddly
                                      Software, but hopefully he found an occupation that would benefit from
                                      his talents as a programmer. He loved to roll up his sleeves and dig
                                      into machine language, especially for the 1802.

                                      I'm positive that it's OK to post his material in the files section.
                                      I'm glad he said he was releasing his 1802 work into the public
                                      domain. He did not have to do that. Some programmers will issue an
                                      unrestricted license to use code or documentation for non-commercial
                                      hobbyist use, retaining control, should someone make a derivative
                                      commercial application... just in case. But Paul placed his material
                                      into public domain, meaning that it's totally unrestricted for anyone
                                      for any purpose. It's possible that I still have his letter somewhere
                                      in my files. Next time I'm browsing through my paper files, I'll keep
                                      an eye out for it.

                                      Originally, I purchased CSOS and CSTP... but Paul sent me extra copies
                                      of that and Starship. I'm fairly sure I still have those blue
                                      Highlander tapes. I used them with my VIP... never had an ELF. Paul
                                      was a contributor to the VIPER newsletter, too.

                                      73 de Ray

                                      On Feb 2, 2013, at 2:38 AM, Richard wrote:

                                      > Cuddly Software? I have two programs from him. One is a game called
                                      > 'Starship' and the other one is a debugger called CSTP. I even have
                                      > CSTP still on its original tape (Scotch Highlander).
                                      >
                                      > At some time my ASCII keyboard was broken. I had accidentally messed
                                      > up the voltages and killed the keyboard encoder. Without it I could
                                      > not use CSTP anymore. so I tried to modify it to take its input from
                                      > the hex keyboard.
                                      >
                                      > It did not work. Finding all bytes wich might be input instructions
                                      > was easy enough, but I had no disassembler or even a listing and
                                      > changed some bytes which were no input instructions.
                                      >
                                      > That's why I decided to call him and ask for some help. He talked me
                                      > through the modifications and sent me a hex dump of CSTP. On the
                                      > silver thermo paper you mentioned.
                                      >
                                      > Some time later my poor parents got the telephone bill and I got
                                      > into some trouble. They grounded me and made me spend some time in
                                      > my room with the Elf and CSTP. Harsh punishment that really made me
                                      > feel sorry for what I did :-)
                                      >
                                      > CSTP was the first software I bought and I still use it today
                                      > whenever I write something for the Elf. I load it into the emulator
                                      > and then can debug my program with it. I always wanted to post it to
                                      > the files section, but did not know where to ask for permission. Is
                                      > there a chance that it's ok?
                                      >
                                    • William Donnelly
                                      Thanks for people like Paul. – Bill
                                      Message 18 of 20 , Feb 2, 2013
                                      • 0 Attachment
                                        Thanks for people like Paul.

                                        – Bill
                                      • Kevin
                                        ... By graph , you really mean nudie ASCII art, right?
                                        Message 19 of 20 , Feb 2, 2013
                                        • 0 Attachment
                                          --- In cosmacelf@yahoogroups.com, Rick Cortese wrote:
                                          >
                                          > We also had a ton of those Diablo daisy wheel printers...Quality and accuracy was OK but at 30 cps it took a long time to print out a graph.


                                          By "graph", you really mean nudie ASCII art, right?
                                        • Charles Richmond
                                          ... Kevin, I m sure he was talking about his poster-size printout of Mr. Spock from Star Trek. :-) -- +----------------------------------------+ ...
                                          Message 20 of 20 , Feb 2, 2013
                                          • 0 Attachment
                                            On Feb 2, 2013, at 2:35 PM, Kevin wrote:

                                            > --- In cosmacelf@yahoogroups.com, Rick Cortese wrote:
                                            > >
                                            > > We also had a ton of those Diablo daisy wheel printers...Quality
                                            > and accuracy was OK but at 30 cps it took a long time to print out
                                            > a graph.
                                            >
                                            > By "graph", you really mean nudie ASCII art, right?
                                            >
                                            Kevin, I'm sure he was talking about his poster-size printout of Mr.
                                            Spock from Star Trek. :-)

                                            --
                                            +----------------------------------------+
                                            |.....Charles and Francis Richmond.......|
                                            |........................................|
                                            |..plano dot net at aquaporin4 dot com...|
                                            +----------------------------------------+
                                          Your message has been successfully submitted and would be delivered to recipients shortly.