Notes on two presentations by Gordon Bell ca. 1998
|
From: Anne & Lynn Wheeler on 26 Mar 2010 15:10 Anne & Lynn Wheeler <lynn(a)garlic.com> writes: > now I did have a proposal that predated RP3 ... that would intermix blue > Iliad (first 32bit 801 chip design) and 370 boards ... big problem > getting to something like 96 boards (aka processors) per rack ... was > all heat issues. old post/reference > http://www.garlic.com/~lynn/2004m.html#17 mainframe and microprocessor re: http://www.garlic.com/~lynn/2010f.html#77 Notes on two presentations by Gordon Bell ca. 1998 in that time-frame, endicott was shipping vm/pc ... a 370 board for the xt/pc (xt/370) ... a couple motorola chips that did 370 subset at about 100kips. however, i was looking a 3chip-set from germany that did full 370 at 3mips (aka 30 times faster) ... and blue iliad was targeted at running 20 (801) mips. blue iliad was an enormous chip that ran really hot ... contributing to never getting finished. recent reference mentioning that endicott blamed me for 6month slip in shipping vm/pc ... I had done some thruput measurements and identified it was page thrashing quite a bit ... and that resulted in six month slip while they re-engineered with another 128kbytes of memory (to help with page thrashing ... however, I also provided them with enhanced page replacement algorithm especially in memory constrained environment and an improved filesystem) http://www.garlic.com/~lynn/2010f.html#8 What was the historical price of a P/390? http://www.garlic.com/~lynn/2010f.html#10 What was the historical price of a P/390? -- 42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Del Cecchi on 26 Mar 2010 20:15 "Anne & Lynn Wheeler" <lynn(a)garlic.com> wrote in message news:m3ljdekifx.fsf(a)garlic.com... > Anne & Lynn Wheeler <lynn(a)garlic.com> writes: >> now I did have a proposal that predated RP3 ... that would intermix >> blue >> Iliad (first 32bit 801 chip design) and 370 boards ... big problem >> getting to something like 96 boards (aka processors) per rack ... >> was >> all heat issues. old post/reference >> http://www.garlic.com/~lynn/2004m.html#17 mainframe and >> microprocessor > > re: > http://www.garlic.com/~lynn/2010f.html#77 Notes on two presentations > by Gordon Bell ca. 1998 > > in that time-frame, endicott was shipping vm/pc ... a 370 board for > the > xt/pc (xt/370) ... a couple motorola chips that did 370 subset at > about > 100kips. however, i was looking a 3chip-set from germany that did > full > 370 at 3mips (aka 30 times faster) ... and blue iliad was targeted > at > running 20 (801) mips. blue iliad was an enormous chip that ran > really > hot ... contributing to never getting finished. Blue idiot didn't get finished for a plethora of reasons. I finally came to the conclusion that the group doing it in Los Gatos were, to put it politely, too "research oriented". Rather than do something that worked they were always chasing rainbows or something. I mean, 8 level cascode current switch logic? Brad Dunham place and route? If they had delivered something the gravy train would have ended. snip
From: Anne & Lynn Wheeler on 26 Mar 2010 21:07 "Del Cecchi" <delcecchi(a)gmail.com> writes: > Blue idiot didn't get finished for a plethora of reasons. I finally > came to the conclusion that the group doing it in Los Gatos were, to > put it politely, too "research oriented". Rather than do something > that worked they were always chasing rainbows or something. I mean, 8 > level cascode current switch logic? Brad Dunham place and route? If > they had delivered something the gravy train would have ended. re: http://www.garlic.com/~lynn/2010f.html#77 Notes on two presentations by Gordon Bell ca. 1998 http://www.garlic.com/~lynn/2010f.html#78 Notes on two presentations by Gordon Bell ca. 1998 folklore is that one of the people that worked on blue iliad (including their last two weeks *after* they gave notice) ... went on to be principal architect of pa-risc and pa-wide word http://www.hpl.hp.com/news/2001/apr-jun/itanium.html some old email http://www.garlic.com/~lynn/2003e.html#email811006 http://www.garlic.com/~lynn/2003e.html#email811006b http://www.garlic.com/~lynn/2003e.html#email811113 http://www.garlic.com/~lynn/2003e.html#email811115 in this post http://www.garlic.com/~lynn/2003e.html#65 801 (was Re: Reviving Multics other old email mentioning 801, iliad, romp, etc http://www.garlic.com/~lynn/lhwemail.html#801 -- 42yrs virtualization experience (since Jan68), online at home since Mar1970
From: Del Cecchi` on 26 Mar 2010 23:40
Anne & Lynn Wheeler wrote: > "Del Cecchi" <delcecchi(a)gmail.com> writes: > >>Blue idiot didn't get finished for a plethora of reasons. I finally >>came to the conclusion that the group doing it in Los Gatos were, to >>put it politely, too "research oriented". Rather than do something >>that worked they were always chasing rainbows or something. I mean, 8 >>level cascode current switch logic? Brad Dunham place and route? If >>they had delivered something the gravy train would have ended. > > > re: > http://www.garlic.com/~lynn/2010f.html#77 Notes on two presentations by Gordon Bell ca. 1998 > http://www.garlic.com/~lynn/2010f.html#78 Notes on two presentations by Gordon Bell ca. 1998 > > folklore is that one of the people that worked on blue iliad (including > their last two weeks *after* they gave notice) ... went on to be principal > architect of pa-risc and pa-wide word > http://www.hpl.hp.com/news/2001/apr-jun/itanium.html > > some old email > http://www.garlic.com/~lynn/2003e.html#email811006 > http://www.garlic.com/~lynn/2003e.html#email811006b > http://www.garlic.com/~lynn/2003e.html#email811113 > http://www.garlic.com/~lynn/2003e.html#email811115 > > in this post > http://www.garlic.com/~lynn/2003e.html#65 801 (was Re: Reviving Multics > > other old email mentioning 801, iliad, romp, etc > http://www.garlic.com/~lynn/lhwemail.html#801 > I don't know about architecture folks, but the circuit folks were pretty overambitious, to be polite about it. Those interested, there was a paper about it at ISSCC although they never did get hardware that I am aware of. To be historical, there was supposed to be a 5100 followon that would use Blue Iliad. The manager of the group got all upset when we proposed a bipolar gate array implementation alternative. He eventually went to boca and presided over the System/23 and later went to DEC. Dennis Roberson. Wonder what happened to him? del |