The following message is a courtesy copy of an article that has been posted to bit.listserv.ibm-main as well.
[email protected] (Barry Merrill) writes: > You reminded me that back in 1976 when I had just joined > Sun Oil, and we were a big IBM datacenter, I requested a > new feature in our 3270s that led to several phone calls > from the IBM'er responsible to answer the request > (and my first call from IBM Japan!). > He sincerely examined the request in several calls, > but finally convinced himself that it couldn't be done, > or couldn't be done with then-current technology. > > All I wanted was a key that would take my cursor back to > where it was just before it's current location (i.e., > after I had hit something and didn't know what I had > done, and wanted to go back to where I had been!). we had something of a battle with kingston claiming that 3274/3278 was downgrading from 3272/3277 for purpose of interactive computing ... and "lost" when they claimed that 3274/3278 was purely being targeted for data entry ... and not for interactive computing. post with some of the analysis http://www.garlic.com/~lynn/2001m.html#19 3270 protocol as noted in the above ... TSO didn't care since they never really claimed to be interactive computing or considered subsecond response time. japan did do (ascii, glass teletype) 3101/topaz ... old reference going into some references http://www.garlic.com/~lynn/2007e.html#15 we had some number of early 3101 "-1" ... and were looking to get copy of the ROM for "-2" and burning our own ROMs to field upgrade from "-1" to "-2". prior to getting 3101 at home, I had cdi miniterm ... some old pictures including home cdi miniterm and home 3101 (not a very good picture) http://www.garlic.com/~lynn/lhwemail.html#oldpicts one of the pictures shows an old "HYPER" 3270 logon screen. this was project to move 300 people from the IMS group out of STL to an offsite bldg with their 3270 CMS service back to STL datacenter. They had looked at doing "remote" 3270 ... but the human factors were intolerable. I did HYPERChannel device driver for them ... basically channel extender over T1 microwave link ... with 300 "local" 3270s at the remote end. The transmission time wasn't seriously affected ... but a side-effect of moving the 3274 controllers directly off the mainframe channels, resulted in overall 10-15% increased system thruput ... i.e. the HYPERChannel controllers had significantly lower channel busy for the same amount of 3270 data ... vis-a-vis 3274 controllers, the significant reduction in (3270) channel busy resulted in overall system thruput increase. -- 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
