19 Jun
2024
19 Jun
'24
3:36 p.m.
On 6/18/24 21:09, Murray McCullough via cctalk wrote:
Today I came across an obituary of Lynn Conway,
computer pioneer in the
field of VLSI(along with Carver Mead) and also in one called dynamic
instruction scheduling(used in supercomputing world). More to the point
Conway was transgender and suffered for this, an almost forgotten pioneer
in the microcomputing and supercomputing fields. Also, as a researcher at
IBM and Xerox Parc where she contributed to the first years of
microcomputing, the GUI and Ethernet protocol development. Eventually the
IEEE recognized her contributions as did IBM - better late than never!
WOW, sad news! Yes, Larry/Lynn Conway was right in the
middle of MAJOR developments in computers and ICs. I think
as Larry, he wrote a cycle-by-cycle simulator that was used
to test performance of computer designs in the 360 series
and later machines. The original ran on the 7094 (that
should date this.) Later, that simulator was used to
predict the performance of what became the 360/85, which was
really the prototype of the 370/165. The /85 was the first
IBM machine to use a cache (called a storage buffer by
IBM). The /85 was the first production machine from IBM to
use monolithic integrated circuits (MST4) and water cooling
The simulator was also used to evaluate the performance of
the never-built FS and ACS projects.
When I heard about Lynn and Carver's book and work in logic
synthesis, I thought they were nuts. Well, I was looking
about a year ahead, and they were looking TWENTY years into
the future!
Jon