API tools faq
paste
Advertisement
Guest User

Atari TLP first connection in over 30 years

a guest
Feb 6th, 2018
157
0
Never
Add comment
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 15.17 KB | None | 0 0
raw download clone embed print report
  1. 01/07/18 17:34:51 <joey_z> can't seem to hit shift+STOP in the learning phone
  2. 01/07/18 17:35:13 <joey_z> it's START+Shift+s
  3. 01/07/18 18:20:23 <joey_z> SteveS, seems the codes that the learning phone uses for the special keyboard characters aren't compatible with cyber1
  4. 01/07/18 18:20:44 <SteveS> bummer
  5. 01/07/18 18:21:23 <SteveS> I assume you got an account
  6. 01/07/18 18:21:24 <joey_z> not quite anyway
  7. 01/07/18 18:21:26 <joey_z> yeah
  8. 01/07/18 18:21:48 <joey_z> it seems that NEXT works, and that's about it
  9. 01/07/18 18:22:09 <SteveS> I can't imagine codes changing. I wonder The Learning Phone was only a prototype
  10. 01/07/18 18:22:33 <SteveS> even in the '80s it would have to have emulated a real terminal
  11. 01/07/18 18:22:56 <joey_z> well, it wouldn't surprise me if the codes for a real PLATO terminal, and the ones meant to be sent in 8 bits down the telephone line were different
  12. 01/07/18 18:23:29 <SteveS> could a plato terminal have used 7-bit ASCII
  13. 01/07/18 18:23:34 <joey_z> the real PLATO terminals operated in a bit of a weird mode IIRC, 9 bits maybe?
  14. 01/07/18 18:23:45 <SteveS> oh
  15. 01/07/18 18:24:20 <joey_z> at a weird baud rate
  16. 01/07/18 18:24:26 <joey_z> with a weird modulation scheme
  17. 01/07/18 18:24:28 <SteveS> well, yet another dream dashed
  18. 01/07/18 18:25:03 <SteveS> to be honest 320x192 didn't look very useful
  19. 01/07/18 18:25:53 <proteanthread> you could get 40*25 charcters out of 320*192
  20. 01/07/18 18:26:20 <joey_z> proteanthread, yes, but it's hard to do 512x512 in 320x192 :)
  21. 01/07/18 18:26:38 <SteveS> Atari's The Learning Phone, PLATO terminal emulator
  22. 01/07/18 18:28:08 <SteveS> I never saw it work. I thought maybe it scrolled
  23. 01/07/18 18:28:27 <MoparStephen> interesting - I wonder how VBXE could help?
  24. 01/07/18 18:28:33 <MoparStephen> I believe it has 620*200 mode
  25. 01/07/18 18:28:34 <SteveS> up and down to at least give 320x320
  26. 01/07/18 18:28:39 <MoparStephen> a scrolling mode perhaps?
  27. 01/07/18 18:29:09 <joey_z> SteveS, it just 'zooms out'
  28. 01/07/18 18:29:16 <SteveS> not much point. the idea was to revive an existing product
  29. 01/07/18 18:29:19 <joey_z> if you have 48K, you can zoom back in, and scroll around
  30. 01/07/18 18:29:25 <SteveS> oh
  31. 01/07/18 18:31:29 <joey_z> yes, originally PLATO IV terminals used *20* bit words :)
  32. 01/07/18 18:32:05 <joey_z> but that was before they even had microprocessors in them
  33. 01/07/18 18:32:09 <SteveS> that would be awkward with 8bit ASCII
  34. 01/07/18 18:32:26 <joey_z> I don't know if PLATO V was compatible or not
  35. 01/07/18 18:33:19 <SteveS> are the graphics done similarly to pen plotters?
  36. 01/07/18 18:33:58 <joey_z> some is
  37. 01/07/18 18:34:09 <SteveS> the java based JPTerm is weird. It looks like a pen plot but someone forgot to raise the pen much of the time
  38. 01/07/18 18:34:10 <joey_z> it supports both vector and character graphics
  39. 01/07/18 18:34:19 <joey_z> yeah, JPTerm is broke
  40. 01/07/18 18:34:30 <joey_z> normal not-java pterm works much better
  41. 01/07/18 18:34:43 <SteveS> I couldn't compile it
  42. 01/07/18 18:34:59 <SteveS> did bother trying the windows version
  43. 01/07/18 18:36:14 <joey_z> seems like PLATO terminals send 7 bit data to the far end
  44. 01/07/18 18:36:37 <joey_z> but looking in a network sniffer, the atari is definitely sending different codes compared to pterm
  45. 01/07/18 18:36:56 <joey_z> might just be parity though, actually....
  46. 01/07/18 18:37:02 <joey_z> I wonder...
  47. 01/07/18 18:37:18 <SteveS> any chance there's a bug in Altirra's 850 emulation?
  48. 01/07/18 18:37:42 <joey_z> dunno, doesn't sound too likely
  49. 01/07/18 18:38:10 <joey_z> I think more people would be having issues if there was
  50. 01/07/18 18:39:12 <SteveS> yeah
  51. 01/07/18 18:40:28 <SteveS> still, I find the Learning Phone an interesting bit of history
  52. 01/07/18 18:47:44 <joey_z> well, for one thing, the learning phone is sending even parity
  53. 01/07/18 18:48:02 <joey_z> on the other hand, it also sends completely the wrong code for Shift+STOP
  54. 01/07/18 18:48:11 <joey_z> at least compared to pterm
  55. 01/07/18 18:48:30 <joey_z> codes for next, shift+next, stop are all correct (ignoring parity bit)
  56. 01/07/18 18:48:39 <joey_z> shift+stop is just plain wrong, I guess
  57. 01/07/18 18:48:55 <joey_z> pterm appears to send 0x05
  58. 01/07/18 18:49:19 <joey_z> learning phone sends 0x11 (stop is 0x01, seems 'shift' is indicated by 0x1X)
  59. 01/07/18 18:52:59 <joey_z> PLATO uses 11 bit words to send stuff to the far end
  60. 01/07/18 18:53:05 <joey_z> (reading the docs)
  61. 01/07/18 18:53:24 <joey_z> one parity bit normally, some bits used to determine what the event was
  62. 01/07/18 18:53:41 <joey_z> screen touch, key pressed, etc
  63. 01/07/18 18:55:09 <joey_z> I'm guessing the learning phone doesn't emulate the touch interface :)
  64. 01/07/18 18:56:05 <SteveS> how could an Atari send 11 bit words?
  65. 01/07/18 18:56:12 <SteveS> through a modem
  66. 01/07/18 18:56:37 <SteveS> and through common R: drivers
  67. 01/07/18 18:56:40 <joey_z> the atari can't
  68. 01/07/18 18:56:49 <joey_z> that's what I mean, this isn't *real* PLATO stuff
  69. 01/07/18 18:57:10 <joey_z> this is some 8-bit-ified variant of PLATO commands
  70. 01/07/18 18:57:17 <SteveS> I mean it could do if it took control of the serial bus
  71. 01/07/18 18:57:26 <joey_z> it can't though, not on the A8
  72. 01/07/18 18:57:35 <joey_z> I don't think you can bit bang the POKEY
  73. 01/07/18 18:58:51 <joey_z> I'm also guessing the atari doesn't emulate the pneumatically powered microfiche slide projector :)
  74. 01/07/18 18:58:59 <joey_z> lol
  75. 01/07/18 18:59:13 <joey_z> plato IV had that, dropped by the plato V terminals
  76. 01/07/18 19:05:02 <joey_z> SteveS, here's something about PLATO IV terminals on ARPANET: https://tools.ietf.org/html/rfc600
  77. 01/07/18 19:05:31 <joey_z> which highlights the issue with the 20 bit/11 bit asymmetric communications protocol
  78. 01/07/18 19:14:25 <SteveS> so the Atari would be copying the teletype mode
  79. 01/07/18 19:15:16 <SteveS> and so would the terminal emulators of today
  80. 01/07/18 19:15:19 <SteveS> I assume
  81. 01/07/18 19:16:47 <joey_z> probably not
  82. 01/07/18 19:17:18 <SteveS> does it work at all if Altirra's 850 emulation hasn't telnet mode enabled
  83. 01/07/18 19:17:20 <SteveS> ?
  84. 01/07/18 19:17:20 <joey_z> that would be great, if they all agreed on some standard at the very beginning
  85. 01/07/18 19:17:35 <joey_z> I am running, as far as I can tell, with telnet mode off
  86. 01/07/18 19:17:41 <SteveS> k
  87. 01/07/18 19:17:44 <joey_z> I don't think you want it, this *isn't* telnet
  88. 01/07/18 19:17:57 <joey_z> but altirra checks the box every time I open the 'modem' configuration
  89. 01/07/18 19:17:59 <SteveS> I had it on
  90. 01/07/18 19:18:10 <joey_z> but it does appear to be genuinely off if you uncheck it and then hit OK
  91. 01/07/18 19:18:27 <SteveS> but don't have an account
  92. 01/07/18 19:18:35 <joey_z> I'm making a spreadsheet of 'the learning phone' keycodes vs pterm keycodes
  93. 01/07/18 19:18:55 <joey_z> TLP is definitely doing parity, but it doesn't seem to cause an issue
  94. 01/07/18 19:19:04 <joey_z> they must have the far end set to ignore it
  95. 01/07/18 19:19:13 <SteveS> you've taken this further than anyone else has managed
  96. 01/07/18 19:19:20 <SteveS> in 33 years
  97. 01/07/18 19:20:09 <SteveS> you know though. I didn't even think about these being interface without a modem to terminal servers
  98. 01/07/18 19:24:09 <SteveS> I used a portable terminals in '78. synchronous 1200 baud modems. no not specifically plato terminals. I never saw one.
  99. 01/07/18 19:24:35 <SteveS> oh
  100. 01/07/18 19:24:47 <SteveS> I did see a plasma terminal and didn't know what it was
  101. 01/07/18 19:25:09 <SteveS> hmm
  102. 01/07/18 19:25:27 <joey_z> was it really big?
  103. 01/07/18 19:25:41 <joey_z> third pic: https://en.wikipedia.org/wiki/PLATO_(computer_system)
  104. 01/07/18 19:25:42 <Atari800XL> [WIKIPEDIA] PLATO (computer system) | "PLATO (Programmed Logic for Automatic Teaching Operations) was the first generalized computer-assisted instruction system. Starting in 1960, it ran on the University of Illinois' ILLIAC I computer. By the late 1970s, it supported several thousand graphics terminals distributed worldwide, running on nearly..."
  105. 01/07/18 19:26:06 <joey_z> no, nvm
  106. 01/07/18 19:26:12 <joey_z> that's a plato V
  107. 01/07/18 19:26:18 <joey_z> but still, one of the plasma models
  108. 01/07/18 19:26:26 <SteveS> I'd say it was a big as a TV
  109. 01/07/18 19:26:33 <joey_z> I thought they had a plato IV model
  110. 01/07/18 19:26:37 <SteveS> but long, long time ago. who knows
  111. 01/07/18 19:26:39 <joey_z> picture there somewher
  112. 01/07/18 19:26:50 <joey_z> if it wasn't a IV or V, then it wasn't a plato
  113. 01/07/18 19:26:55 <joey_z> all the later models were CRT
  114. 01/07/18 19:27:53 <joey_z> those were the ones after CDC took over
  115. 01/07/18 19:28:02 <joey_z> PLATO IV and V were UIUC in-house designs I think
  116. 01/07/18 19:28:16 <SteveS> my first Co-Op job was at MITRE. There was likely lot of cutting edge stuff. Plato not cutting edge in the late 70s but an amber screen was surprising tome
  117. 01/07/18 19:28:22 <SteveS> to me
  118. 01/07/18 19:29:10 <SteveS> MITRE was in Cost Plus mode
  119. 01/07/18 19:30:08 <SteveS> desperately in need of engineers and paying whatever price. OT even to interns
  120. 01/07/18 19:30:36 <SteveS> taking chances on people like me
  121. 01/07/18 19:30:48 <SteveS> i.e. far below 4.0 grade point average
  122. 01/07/18 19:31:30 <joey_z> maybe there's a key translation table in the learning phone that I could modify to 'fix' it
  123. 01/07/18 19:35:17 <Kamilion> messing with a physical device?
  124. 01/07/18 19:36:00 <Kamilion> or emulation?
  125. 01/07/18 19:40:06 <joey_z> Kamilion, emulation
  126. 01/07/18 19:40:39 <SteveS> the ultimate goal is to get it working on real hardware
  127. 01/07/18 19:45:02 <Kamilion> through a real 1050's RS232 port?
  128. 01/07/18 19:45:19 <SteveS> 850
  129. 01/07/18 19:45:24 <SteveS> 1050 is a disk drive
  130. 01/07/18 19:49:39 <joey_z> aha!
  131. 01/07/18 19:49:44 <Kamilion> er, sorry, yeah, 850
  132. 01/07/18 19:49:53 <joey_z> I found a key combo that produces the right key code to do Shift+STOP
  133. 01/07/18 19:49:55 <joey_z> (0x05)
  134. 01/07/18 19:50:13 <joey_z> it's the keycode produced by what the manual calls 'shift+sub'
  135. 01/07/18 19:50:32 <joey_z> which is start+shift+=
  136. 01/07/18 19:50:46 <joey_z> or start+pipe '|' if you're on a PC keyboard
  137. 01/07/18 19:50:51 <joey_z> OK, lets give this another try...
  138. 01/07/18 19:50:56 <SteveS> can't type CTRL+"E" to get 0x05?
  139. 01/07/18 19:51:09 <joey_z> SteveS, no, it's NOT ASCII
  140. 01/07/18 19:51:40 <joey_z> the codes and ASCII have a lot of shared ground, but beyond letters, not much matches
  141. 01/07/18 19:52:28 <SteveS> K
  142. 01/07/18 19:52:29 <joey_z> yep, I'm logged in on atari TLP
  143. 01/07/18 19:52:37 <joey_z> start+shift+= worked
  144. 01/07/18 19:52:52 <joey_z> which is start+| on a PC KB
  145. 01/07/18 19:53:24 <SteveS> not easy to type
  146. 01/07/18 19:53:36 <SteveS> if it were an A8
  147. 01/07/18 19:53:57 <SteveS> two handed
  148. 01/07/18 19:54:06 <SteveS> or 3 finger contortion
  149. 01/07/18 19:54:10 <SteveS> lol
  150. 01/07/18 19:56:36 <joey_z> unfortunately, I think 8005 is the 'color' port, and TLP isn't filtering the color codes
  151. 01/07/18 19:56:57 <joey_z> so this is less usable than I would have hoped
  152. 01/07/18 19:58:00 <SteveS> yeah. A8 *hi-res*, 1.5 colors
  153. 01/07/18 19:58:33 <SteveS> guess it's supposed to emulate the Plasma terminal's color
  154. 01/07/18 19:59:22 <SteveS> JPTerm was picking port 5004 (I think)
  155. 01/07/18 20:00:42 <joey_z> 5004 isn't compatible
  156. 01/07/18 20:00:58 <joey_z> must use the 20bit/11bit system converted into some 8 bit variant
  157. 01/07/18 20:01:07 <joey_z> only 8005 produces any sort of usable result
  158. 01/07/18 20:01:17 <joey_z> AFAIK they don't have a non-color 'ASCII' mode port
  159. 01/07/18 20:01:45 <joey_z> I believe they called 8005 the 'color ASCII mode' port or something like that
  160. 01/07/18 20:04:02 <joey_z> 'Color is only available under the ASCII port 8005 when connecting, but it achieves faster perceived output due to buffering design at the mainframe. '
  161. 01/07/18 20:04:56 <joey_z> there's a version of adventure which appears to be fairly colorless so far
  162. 01/07/18 20:07:13 <joey_z> adventure as in zork varient
  163. 01/07/18 20:19:59 <SteveS> Plato generation 2 terminal cost $12,000 in 1972
  164. 01/07/18 20:20:54 <SteveS> oh that can't be right
  165. 01/07/18 20:21:48 <pusakat[m]> if the terminals were connected directly to serial ports of mainframe or mini, no need for modems, right?
  166. 01/07/18 20:23:59 <joey_z> pusakat[m], I don't know about the later ones, but the early PLATO terminals had the modem built in
  167. 01/07/18 20:24:16 <joey_z> they had two versions, bell 202 for phone lines, and a leased line version
  168. 01/07/18 20:35:19 <joey_z> the learning phone does emulate the touch screen thing
  169. 01/07/18 20:35:31 <joey_z> appears to use PMG for the cursor
  170. 01/07/18 20:36:38 <joey_z> to put it lightly though, this thing would prove hard to use compared to a real PLATO for certain thing
  171. 01/07/18 20:36:38 <joey_z> s
  172. 01/07/18 20:37:01 <joey_z> runnig at such reduced resolution would make certain PLATO 'lessons' pretty hard to use
  173. 01/07/18 20:37:06 <joey_z> others would probably be fine
  174. 01/07/18 20:37:32 <Kamilion> they probably havn't catered to a 40 column terminal since the late 90s
  175. 01/07/18 20:37:45 <joey_z> Kamilion, that's not really the issue
  176. 01/07/18 20:37:57 <joey_z> PLATO came out in the late 60's/early 70's I think
  177. 01/07/18 20:38:04 <joey_z> but it was originally made for 512x512
  178. 01/07/18 20:38:14 <joey_z> atari is only 320x192, so they scale it, and it looks pretty bad
  179. 01/07/18 20:38:16 <Kamilion> 512x512... characters?
  180. 01/07/18 20:38:19 <Kamilion> oh.
  181. 01/07/18 20:38:20 <Kamilion> gotcha.
  182. 01/07/18 20:38:29 <joey_z> 64 column text
  183. 01/07/18 20:38:30 <Kamilion> some kind of bitmapped display mode.
  184. 01/07/18 20:38:33 <joey_z> or graphics
  185. 01/07/18 20:38:39 <joey_z> or, really, a mix of both
  186. 01/07/18 20:38:54 <Kamilion> but the server side handles the text rendering out of pixels?
  187. 01/07/18 20:38:59 <joey_z> nope
  188. 01/07/18 20:39:11 <joey_z> you can draw characters, or you can draw lines
  189. 01/07/18 20:39:28 <joey_z> you can also redefine characters, and get some limited tiled graphics
  190. 01/07/18 20:39:38 * Kamilion scratches his head
  191. 01/07/18 20:39:51 <Kamilion> huh.
  192. 01/07/18 20:39:54 <joey_z> locally, it drew the characters as bitmapped graphics
  193. 01/07/18 20:40:13 <joey_z> the plato terminals had a font ROM with also a font RAM for the redefinable characters
  194. 01/07/18 20:40:23 <Kamilion> wow, so the thing I've been yelling about missing over serial ports has been around for a long time...
  195. 01/07/18 20:40:33 <joey_z> well, not really
  196. 01/07/18 20:40:40 <joey_z> plato isn't over a normal serial port :)
  197. 01/07/18 20:40:43 <Kamilion> i know DEC terminals with later VT encodings can use sixels for graphics
  198. 01/07/18 20:40:44 <joey_z> that's half the battle
  199. 01/07/18 20:40:59 <Kamilion> see: https://github.com/saitoha/PySixel
  200. 01/07/18 20:41:13 <joey_z> plato used 20 bit incoming, 11 bit outgoing over a 1260 baud modem
  201. 01/07/18 20:41:24 <joey_z> far from normal serial :)
  202. 01/07/18 20:41:46 <Kamilion> heh, and not something that your average UART is going to be fine with
  203. 01/07/18 20:41:53 <joey_z> it also had a 16x16 IR sensor 'touch screen'
  204. 01/07/18 20:41:59 <Kamilion> but that's totally doable with softserial on an arduino
  205. 01/07/18 20:42:04 <joey_z> the plato IV terminal was all discrete logic
  206. 01/07/18 20:42:13 <joey_z> they're also unobtainium
  207. 01/07/18 20:42:17 <joey_z> don't ask me where to get one
  208. 01/07/18 20:42:19 <Kamilion> i figured
  209. 01/07/18 20:42:29 <joey_z> that doesn't mean I don't know where to get one, it just means I'm not telling ;)
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement
Public Pastes
Advertisement
We use cookies for various purposes including analytics. By continuing to use Pastebin, you agree to our use of cookies as described in the Cookies Policy.  OK, I Understand
Not a member of Pastebin yet?
Sign Up, it unlocks many cool features!