[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Gordon Bell sighting



(Is Dan on this list?)

http://www.theregister.co.uk/990608-000003.html

>  Posted 16/06/99 11:29am by Mike Magee
>
>  Top Microsoft boffin votes for Russian Merced Killer
>
>  The Elbrus E2K processor, dubbed the Merced Killer, has received unexpected
>  endorsement from a senior Microsoft executive. 
>
>  Gordon Bell, who heads the Microsoft research unit, and developed DEC
PDP and Vax,
>  presented a table at this year's International Symposium on High
Performance Computing
>  which shows Merced in a poor light compared to the Russian chip. 
>
>  The slide Bell showed gave the E2K's clock frequency as 1.2GHz, compared
to Merced's
>  .8GHz, SPECint85/SPECfp95 as 135/350 compared to 45/70, Due size as 126
sq mm
>  compared to Merced's 300 sq mm, power dissipation as 35 watts compared
to 60 watts. 
>
>  For the E2K, bus bandwidth as 15Gbyte/sec, cache sizes as 64/256K,
GFLOPS as 10.2
>  and system ship in Q4 2001. 
>
>  Bell's slide showed these last four categories as non appplicable to the
Merced
>  microprocessor. 
>
>  By the way, has Merced taped out yet, anyone?  
>
>  See also 
>  Battle royal breaks out over Russian chip claim 
>  Intel uses Russian military technologies 
>  Russian Merced Killer to achieve 600MHz 
>  Russian chip makers get to .35 micron 
>  Moscow government to support Merced Killer 
...
>          The Register and its contents are copyright  1999, Situation
Publishing. All rights reserved.
>       Mail: 20-22 Maddox Street, London, W1R 9PG. Tel: +44 (0)171 499
2264 Fax: +44 (0)171 493 5922
>
>                     Editors: Mike Magee, John Lettice and Drew Cullen.
>              Managing Editor (site): Tony Smith. Managing Editor (news):
Sean Fleming
>            Reporters: Tim Richardson and Linda Harrison. Advertising and
Sales: Pranav J Osa. 

http://www.cs.clemson.edu/~mark/acs.html

>     However, there was a secret IBM project that operated in parallel
with the development and deployment of the Model 91
>     with a goal to produce a supercomputer that would advance scientific
computing with as much impact as did the IBM
>     Stretch. Indeed, like Stretch, it encountered difficulties and
criticisms, and even more so than Stretch its contributions and
>     achievements remain relatively untold. 
>
>     It is my hope that this set of web pages (and I hope later
publication) will help reveal more complete histories of the ACS-1
>     and the ACS-360 -- the amazing IBM supercomputers that were designed
and engineered in the 1960's -- and thus fill in
>     details that are only given as hints in John Cocke's Turing Award
Lecture and given as a brief section (7.5) in the wonderful
>     book on IBM history by Emerson Pugh, Lyle Johnson, and John Palmer. 

http://www.cs.clemson.edu/~mark/acs_technical.html

>          Out of this project, therefore, came numerous advances in
compiler technology, and, more importantly,
>          the foundations of the theory of program analysis and
optimization. 
>
>          -- Fran Allen, The history of language processor technology in
IBM (1981)