The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main,alt.folklore.computers as well.
[email protected] (Ted MacNEIL) writes: > I remember two models -- KP-26 & KP-29. > I had my own (as probably did others) card 'programmed' for certain stops, so > I didn't have to space over. > I guess that was the precurser to tab stops on a 327x terminal. based on whether or not, 327x terminal was being used as upgrade to keypunch card entry ... or used as upgrade to 2741 selectric terminal (i.e. 026 keypunch stops or typewriter tab settings). we got into argument about whether 3278 was purely a keypunch entry machine ... or possibly might also be used for interactive computing. part of the problem was that some cost cutting in the transition from 3272/3277 to 3274/3278 .... moved a lot of function that had been in the 3277 head back into the 3274 controller (cutting cost of manufacturing 3278). this significantly increased processing overhead delay in the 3274 and made it much less suitable for interactive computing. the interactive delays didn't really show up for keypunch entry work ... and TSO response was so slow that that the transition from 3272/3277 to 3274/3278 wasn't noticable. old post with some measurements http://www.garlic.com/~lynn/2001m.html#19 3270 protocol part of the issue was that with electronics in the head of 3277, it was possible to tweak the operation to make it much more acceptable for interactive computing. one was with a little bit of soldering in the keyboard it was possible to alter the "repeat delay" and the "repeat rate" timing. also it was possible to add a small fifo to eliminate the problem that if you happened to hit a key at the same time a system write to the screen was going on ... the keyboard would lock up (and the keystroke was lost). the keyboard then had to be unlocked before continuing (was terribly annoying). there were also early arguments about whether program function keys were going to be dropped from 3278 ... or if it was possible to get them as option. -- 40+yrs virtualization experience (since Jan68), online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
