peter@stca77.stc.oz (Peter Jeremy) (02/05/91)
The first commercial APL product was APL\360, released by IBM and running on the IBM 360 computers. The back of the APL\360 manual describes a number of useful APL functions supplied in a workspace with APL\360, including execution times. Does anyone know what system these examples were run on? I am interested because when I timed the same functions on APL.68000 on my Amiga, they ran roughly twice as fast. -- Peter Jeremy (VK2PJ) peter@stca77.stc.oz.AU Alcatel STC Australia ...!uunet!stca77.stc.oz!peter 240 Wyndham St peter%stca77.stc.oz@uunet.UU.NET ALEXANDRIA NSW 2015
sethb@Morgan.COM (Seth Breidbart) (02/08/91)
In article <1226@stca77.stc.oz> peter@stca77.stc.oz (Peter Jeremy) writes: >The first commercial APL product was APL\360, released by IBM and >running on the IBM 360 computers. The back of the APL\360 manual >describes a number of useful APL functions supplied in a workspace >with APL\360, including execution times. >Does anyone know what system these examples were run on? I am >interested because when I timed the same functions on APL.68000 >on my Amiga, they ran roughly twice as fast. When I worked for the IBM APL group at Yorktown Heights in 1969, APL\360 was running on a 360/50. It supported several dozen users, on a machine with less CPU capability than the 386 I have at home. Of course, 30K workspaces and no filesystem made that much easier. On the other hand, it was the only system that IBM at the time would allow to run unattended. I don't know if the original implementation was on that machine or a predecessor, so the timings may have been made on a lesser 360. Seth sethb@fid.morgan.com
cortesi@informix.com (David Cortesi) (02/08/91)
In article <1226@stca77.stc.oz> peter@stca77.stc.oz (Peter Jeremy) writes: >The first commercial APL product was APL\360, released by IBM and >running on the IBM 360 computers. The back of the APL\360 manual >describes a number of useful APL functions supplied in a workspace >with APL\360, including execution times. ...Does anyone know what >system these examples were run on? The product was developed in the IBM Philadelphia Scientific Center. During that era, if memory serves, that site had a 360/75 as its main computing resource. That was a big machine with a fast floating point unit, soon superceded by the 360/85. But the manual, or that appendix of it, might have been prepared elsewhere. The Palo Alto development center, where APL\360 the product was maintained and enhanced, had a 360/50 as its main workhorse. I am a little surprised there is no indication of what machine produced the timings. The wide range of price & performance over the 360 line was one of its advantages, and whoever wrote the manual would have been well aware that different users would see different results depending on the machine in use. I seem to remember APL\360 could be coerced into running (standalone!) on a 360/30, though a 360/40 with DOS or a /50 with OS/MFT was a more typical host.
sethb@Morgan.COM (Seth Breidbart) (02/13/91)
In article <1991Feb8.003825.15723@informix.com> cortesi@informix.com (David Cortesi) writes: >In article <1226@stca77.stc.oz> peter@stca77.stc.oz (Peter Jeremy) writes: >>The first commercial APL product was APL\360, released by IBM and >The product was developed in the IBM Philadelphia Scientific Center. Actually, it was developed by the APL group at Yorktown Heights. The APL group was originally (before my time) part of IBM Research, and was later moved to the Data Processing Division. Adin Falkoff was the manager of the group, and in 1969 or 1970 he was also made the head of the New York Scientific Center in New York City. They looked for a single location that could hold both groups, and ended up moving to Philadelphia. (I believe that having a frisbee field nearby was a requirement for the location chosen.) Seth sethb@fid.morgan.com
norm@hydro.wbst845e.xerox.com (Norm Crowfoot) (02/15/91)
We were a beta site for APL\1130 during the summer of '68. Had some special ties to Yorktown and got a very early version of the interpreter: some "features": o single character names for everything o one and two dimensions, from 1 to 256 for each dimension Wow, what a long time ago. What a strange language. I thought I was the only in the little research lab that didn't understand it (I was young), the other guys were to afraid to ask. -- | | Norm Crowfoot Voice: 716/422-2458 |\ | KBS Competency Center Internet: crowfoot@arisia.xerox.com | \ | Xerox Corp - MS: 0845-20C XNS: ncc:wbst845e:xerox | \| 780 Salt Road FAX: 716/422-2479 | | Webster, New York 14580 /|\ / | \