• Please review our updated Terms and Rules here

New Datapoint 2200 owner.

I read about their work on the prior 3300 - basically not a product they wanted to make, but they knew it would be profitable and useful as one of the earliest glass teletypes. Then that gave them the capital to make the product they knew they could make, an actual programmable terminal. The Viatraon System 21 from end of '69 had a similar idea (and IIRC had a color screen), one of those got found a few years back (but I haven't kept up on whether it got successfully restored back to operational -- curious on how slow it was and how it was programmed; cause the first 1-bit serial processor of the Datapoint was pretty slow also).

But yes, prior minis and mainframes could be programmed. I meant "never intended" on something that didn't require a committee approval to get purchased (due to cost) I read about the farmers using the Datapoint - and programming it themselves, eventually including sending their weekly reports over modems rather than mailed tapes (that's all probably an over simplification of actual events). Still, "programmable terminal" -- it's a terminal, it just shows what it's told. Oh, but it's programmable also! I think cutting the screen in half to 8 rows (from the 3300 to 2200) was also a smart move - I can't remember if it was for cost reasons, or a deliberate style. But I remember ads/photos, like for transcribing its perfect since you can fit your full size page above the screen on a stylus, or it avoids the problem of a huge monitor blocking the receptionists view of customers walking in (or a practical size/weight for farmer barns).
 
The screen size was supposed by the size of a punched card as far as I understood. The book on Datapoint is quite well worth a read.

I think you miss the point. The idea they had was to create a terminal that coild support as many vatiants of hosts as possible. They already done the 3300 and it was quite limited, basically a glass tty. Now they wanted to be able to communicate with IBM, Univac and all the other mainframes. So if they just put a small mini inside the machine this would be possible. Thus a generic computer in a terminal. A programmable terminal is more than a device that show what it is told.

The PDT series from DEC is marketed as programmable terminals. They include a complete PDP-11 inside them, albeit the smaller LSI-11. https://www.bitsavers.org/pdf/datapro/programmable_terminals/Datapro_C21_Digital.pdf

The CPU of the originl 2200 was 8 bit bit serial. Not 1-bit serial. The ALU is 1 bit wide and data paths are 1 bit wide. Registera are 8 bits wide. From an instruction set view it is an 8 bit computer. The later version 2 is 8 bit parallell.

There are other examples of bit serial computers. PDP-8/S is one. The HP98x0 series all share the same 16 bit bit-serial implementation of the HP 21xx instruction set.
 
Yes, I think I recall reading in the Datapoint book how the screen size was based off prior punch cards, - and maybe that was initially intended as a selling appeal, but then ended up having other benefits (as sleek office equipment).

And sure, I get that point, I was just mock-dialoging how it would be advertised from the view of 1971. The term "personal computer" hadn't really come up yet, since the mindset was still that "a computer" was a huge room full of equipment -- so they compromised in calling it "programmable terminal", which folks initially maybe interpreted that as it could just be re-programmed to be a terminal for System A, B, C, etc. -- that by itself would be useful, since no wanted N terminals in their office to talk to N systems -- but not quite yet realizing what this truly meant: general purpose programming (which is what Man and Dog demonstrates -- a piece of software that has nothing to do with interfacing as a terminal to any other system, just some "general purpose" application that someone chose to write). Obviously academics and those familiar with computers understood all this (as those professional farmers saw the utility), but I mean from the general public perspective it was a gradual realization of what was coming.

I thought I recall something in the Datapoint book about it being described as a "1-bit serial processor" and that influenced Intel decision to adopt the little-endian convention? That's how I interpreted it as a "1-bit serial processor" (separate from the external 8-bit serial components).
 
voidstar78 said:
Yes, I think I recall reading in the Datapoint book how the screen size was based off prior punch cards, - and maybe that was initially intended as a selling appeal, but then ended up having other benefits (as sleek office equipment).

And sure, I get that point, I was just mock-dialoging how it would be advertised from the view of 1971. The term "personal computer" hadn't really come up yet, since the mindset was still that "a computer" was a huge room full of equipment -- so they compromised in calling it "programmable terminal", which folks initially maybe interpreted that as it could just be re-programmed to be a terminal for System A, B, C, etc. -- that by itself would be useful, since no wanted N terminals in their office to talk to N systems -- but not quite yet realizing what this truly meant: general purpose programming (which is what Man and Dog demonstrates -- a piece of software that has nothing to do with interfacing as a terminal to any other system, just some "general purpose" application that someone chose to write). Obviously academics and those familiar with computers understood all this (as those professional farmers saw the utility), but I mean from the general public perspective it was a gradual realization of what was coming.

I thought I recall something in the Datapoint book about it being described as a "1-bit serial processor" and that influenced Intel decision to adopt the little-endian convention? That's how I interpreted it as a "1-bit serial processor" (separate from the external 8-bit serial components).
Click to expand...
My memory from the book was that they sold it to the investors as a smart terminal on the basis that the investors wouldn't understand the concept of a small computer but the designers knew exactly what their were creating. (And note how Intel didn't see the point of the 8008. They marketed it to try and cover their costs. It was only when they realised it was selling that they saw processors as a product worth developing). PS Federico Faggin's autobiography briefly covers the story from the Intel side and is well worth a read if you haven't already.

I wonder how the display size was influenced by costs. A larger display would cost more. It would also require more memory and the early models used the expensive (and power hungry) shift memory for video memory.

You're probably aware there where two editions of the 2200, with the first using shift memory. This had an 1-bit ALU built from TTL logic. It used the shift memory and shift registers so everything, except the address bus, was 1-bit serial operation. Serial operation requires little-endian so you can start with bit 0.

Intel created the 8008 to implement the 2200 first edition instruction set (with a few tweaks). The 8080 was an improved version of the 8008 (different bytecodes implementing an extended instruction set). The 8086 was designed to be source code compatible with the 8080 (and 8085). So modern Pentiums are still Datapoints at their core.
 
Hello Mattis and other 2200 folks,

A while back I laid out a pcb based on the EPROM boot board schematic shared here. (Thanks!) The 2200 that I am working with is the serial first version, so I needed to make a few changes to the design, for instance making the data output open collector. I also used the REBOOT signal to reset things instead of TAPE MOTION. I used a 27256 EPROM for the boot program, with DIP switches to select one of four programs. I tested the board off and on over a period of a few weeks, but could never get it to boot the 2200, even though booting from several tapes that I made is pretty reliable now. Finally today I discovered that covering the End of Tape sensor in cassette 1 and also pushing back on the middle cassette 1 switch (which I think is for Cassette in Place) causes the 2200 to boot from the EPROM. I have not figured out how these signals prevent booting, although it makes sense that they might. This board will be helpful for verifying any new programs, as the 2200 seems pretty picky when it comes to tapes. Once I know a program is good, I can focus on trying to get a good tape copy of it.

Bob Grieb
 
SRE28 said:
I worked for Datapoint as a field tech for 11 years. Started in 1973 (Employee #52). Spent a year (1978) in San Antonio writing diagnostics for Datapoint equipment (in Assembler) I knew Gordon, Vic, and Harry (he was called the Wizard). Worked up to a regional support tech in Chicago.. Then I worked for a Datapoint customer that had a LOT of Datapoint equipment for 5 years. Wrote a lot of Databus code for several companies. I have been looking for a couple of years to find a museum to donate my Datapoint equipment too. Recently after seeing a article in Ask Woody about the National Electronics Museum in Maryland, I contacted Bob Roswell and he was very interested to have a Datapoint 2200. So I dug out the 2200 I placed in my attic in 1987 in working condition, along with (2) 8600 processors and an external disk drive. Problem is the CRT on the 2200 is no good (See the picture). I don't want to power it up for fear of damaging the display circuits. I was wondering if anyone has a Datapoint 1100/2200/5500/6600 CRT? I believe all models used the same CRT. The Display board was updated slightly for the 5500/6600 but I believe the CRT was the same, I know they used the same 15KV for all. I was hoping to donate a working 2200 to the museum. I have a collection of tapes, including the DIAG2200 tape that I wrote) I also have a box of extra parts for the 2200/5500. Willing to trade for a CRT.
Click to expand...
..:WOW - impressive vita! I had a Datapoint EntryPlus system in Germany, running on EISA tower with SCO 2.13, for data entry at 24 terminals with "my girls" typing like hell on the very good keyboards, especially for that time (1993ff). Now, I run the system (still on SCO 2.13) on a VM, as the EISA PC quit service due to its age. I just want to "play around" with it, for the good old times sake.

My question: The EntryPlus data entry system used to connect to Windows 3.1 / 95 / XP PCs with the DP Keyboards attached to them, using the Win7350.exe as the terminal. Since I set up the machine as VM, I can't figure out how to properly connect to the VM running DP EntryPlus. Telnet itself works, so the connection is solid, but I get in Win7350.exe just a "disconnect", and with regular Telnet I get a bunch of sequences that I can't decrypt to understand what I need to change where. ==> Do you know what would cause this, or where to check for settings? The EntryPlus system is installed from a tar, then the install command set it up in /usr/E. In Telnet, I get 3x "E@@@IA@ ", then "NOT A DATAPOINT ENTRYPLUS TERMINAL. LOAD ABORTED."

I saw on the DP website that there might be another terminal program, "RMS Terminal", which looks much like that Win7350.exe, but it seems they don't have that available for download...

Thank you very much in advance!
Ben
 
Sorry I never worked on a DP system running SCO. I must have been big in Europe. Every DP System I worked ran either DOS in the early days, or RMS. Then there was the dedicated ARC file server running on a 5500/6600. There was a terminal program for Windows 3.1 / 95 /98 PCs. I think there was a set of diskettes with this in stuff I gave to Dave
 
Someone has created a adventure game for the Datapoint 2200 in assembler in 2026.


The creator apparently used my simulator for creating the game so it is not just me that is using the simulator.
I will write the tap-files to a cassette and see how it looks on my 2200 later this summer.
 
I don't think that game is an "Original" I remember playing a VERY similar game back in the 70's. It was my first exposure to an "Intelligent" program. Back then is was amazing...
 
There has been loads of variants of very similar adventure games since ever.

The source code is provided by the author in this zip-archive https://drive.google.com/drive/folders/1OzaglpR7Bp8WvBuAXAflVGSeeSWGGMYR

The code is a byte code interpreter for the "GNOMECODE" language which is used to describe the game itself. It might be the case that the actual GNOMCODE is taken from some other game, but the interpreter is most likely new and original.
 
MattisLind said:
There has been loads of variants of very similar adventure games since ever.

The source code is provided by the author in this zip-archive https://drive.google.com/drive/folders/1OzaglpR7Bp8WvBuAXAflVGSeeSWGGMYR

The code is a byte code interpreter for the "GNOMECODE" language which is used to describe the game itself. It might be the case that the actual GNOMCODE is taken from some other game, but the interpreter is most likely new and original.
Click to expand...

Very excellent to see! Mattis I mentioned your work in a recent VCF SW OS talk! And a Datapoint 6600 was sold during that VCF - hope it found a good home. It seemed in fair condition and weighed 65 lbs, had two top load tape decks.
 
You must log in or register to reply here.
Back
Top