Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!att!tsdiag!scr1!pechter From: pechter@scr1.UUCP (Bill Pechter) Newsgroups: comp.os.misc Subject: Re: Jupiter & the 8600 Message-ID: <347@scr1.UUCP> Date: 2 Jun 89 16:47:43 GMT References: <24994@agate.BERKELEY.EDU> <7510@bsu-cs.bsu.edu> <25101@agate.BERKELEY.EDU> Reply-To: pechter@scr1.UUCP (Bill Pechter) Organization: CONCURRENT COMPUTER,OCEANPORT NJ Lines: 49 In article <25101@agate.BERKELEY.EDU> mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer) writes: >In article <7510@bsu-cs.bsu.edu> dhesi@bsu-cs.bsu.edu (Rahul Dhesi) writes: >I've heard from multiple sources that much of the jupiter hardware >wound up inside the 8600. The usual thing sited as evidence is that >the 8600 & 8650, unlike the rest of the VAX line, and like the 10/20 Yup, it has keep-alive failure dumps like the KL's and such and some 36 bit bus sizes and other remnants are rumored to remain. I think that is more a function of having some members of the same team i working on both machines, especially when the Jupiter began going south and the 8600 was much delayed and in need of additional talent. The architecture - seperate I BOX, E Box and M Box were very KL-ish The following are unsubstantiated rumors. I have no internal information from my time in DEC Field Service to back the following statements. The 8600 was originally the 11/790, the 11/810 and the 11/860 -- as the development took a longer time the name of the machine changed. It was build in Marlboro at DEC by the Large Computer Group (LCG) instead of the old 11 and Vax guys. The machine was heavily simulated on large DEC gear with the gate arrays being emulated in software. I heard an 11/44 was used to emulate a single chip. The machine was the first Vax with very hairy pipelineing. They under-estimated how hard the new ECL machine would be to produce. Both the Jupiter and the 8600 were a drain on LCG. Only one machine survived. The word at DEC about the Jupiter is they couldn't get enough speed out of it to make it viable in the market. Another rumor is there was a dual 36/32 bit machine on paper that would run TOPS/VMS depending on what microcode you loaded. A nice idea that should've happened if it was true. There were rumors as to 2 Jupiters -- neither of which were working out well so they were scrapped. I don't think any real working models ever were produced. (Unlike the late, lamented PDP11/74-MP multi processor PDP11/70s.) Disclaimer: the above information was a collection of raw rumors and gossip about products I never saw (Jupiter). I did work on 8600's for Field Service and once worked on a KB11-C (one of the last 11/74 single processor units. -- Bill Pechter -- Home - 103 Governors Road, Lakewood, NJ 08701 (201)370-0709 Work -- Concurrent Computer Corp., 2 Crescent Pl, MS 172, Oceanport,NJ 07757 Phone -- (201)870-4780 Usenet . . . rutgers!pedsga!tsdiag!scr1!pechter ** MS-DOS is CP/M on steroids, bigger bulkier and not much better **