Wanted: Linpack benchmarks for SGI 4D/2x0s

[1k1mgm at kuhub.cc.ukans.edu]

Does anybody (S.G. itself, maybe?) have standard Dongarra/Linpack
benchmarks (N=100 vanilla, N=1000 tweaked and/or parallel) for
4D/2x0s?  I have what may be latest and best report from
Dongarra himself, done Feb. 7, 1990, and only Silicon Graphics
report is something old and slow--does a Model 2400 ring any
bells?  (I'm away from my desk...)

This would be useful, not critical, for grant proposal I'm writing.
I'll use Mips M/2000 numbers ~= 4D/210 if there's nothing
SGI-specific.

Christopher Gunn	Molecular Graphics & Modeling Lab
SPAN--KUPHSX::SYBYL	Department of Medicinal Chemistry
913-864-4428 or -4495	University of Kansas, Lawrence, KS  66045