edgould1...@comcast.net (Ed Gould) writes: > https://www.youtube.com/watch?v=sTdWQAKzESA
IBM 5100 1973 at Palo Alto Science Center http://en.wikipedia.org/wiki/IBM_5100 enuf of 1130 emulation to run apl\1130 (SCAMP) product out in 1978 was enuf of 360 emulation (on PALM) to run apl\360 note that (at least low-end and mid-range) 360s & 370s were emulation on some native microprocessor ... so 5100 wasn't all that different. PASC also did the apl microcode assist for 370/145 ... apl with microcode assist on 145 ran almost as fast os on 370/168. some person also helped with the vm370 microcode assist for 138 & 148. Spring 1975, I got sucked into helping endicott do ecps for 138/148 (virgil/tully) ... it was sort of part of the mad rush to get out 370 products after the dearth during the FS period (which is also credited with giving clone processors a market foothold) http://www.garlic.com/~lynn/submain.html#futuresys Endicott had 6kbyte space for microcode and was to select the 6kbyte of highest used vm370 kernel pathlength. Typical 360/370 microcode emulation ran an avg. of 10 native instructions per 360/370 instruction. runs that measured elapsed time & frequency of kernel instruction sequences ... sorted by percent of total kernel time http://www.garlic.com/~lynn/94.html#21 kernel 370->native instructioins translated almost byte-for-byte ... so 6kbytes of highest used kernel instructions accounted for 79+ percent of total kernel time ... moved to microcode gained approx. 72% of kernel time. then they sucked in to spending a year off&on running around the world laying out 138/148 to the product administrators and business forecasters in the different countries ... going over details about how they stacked up against the competition. -- virtualization experience starting Jan1968, online at home since Mar1970 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
