dyer@spdcc.UUCP (06/07/87)
Here are a few first-pass results of my running a few flawed but familiar
benchmark programs on my AT both before and after the addition of the
Inboard 386/AT speedup card. For those of you who don't want to read
further, the results indicate a speedup of roughly 1.7 times a stock
8mhz PC/AT.
The Inboard used in this test had no memory of its own other than its cache.
I don't know what improvement, if any, would be shown by running these
programs out of the Inboard's own 32-bit memory--stuff for a future test.
No math coprocessor was installed (not that these benchmarks would be expected
to use them.) All tests were run on the same hardware and software environment,
with the exception of the replacement of the 286 for the 386 card, under SCO
XENIX 286 OS v. 2.1.3 with Development System v. 2.1.4. Note that this is a
test of the execution speed of 286 16-bit instructions; tests of the XENIX 386
Toolkit C compiler which generates 386 32-bit instuctions will follow as soon
as it arrives.
IBM PC/AT 8mhz IBM PC/AT with Intel Inboard 386/AT
at 16mhz, cache enabled
Drystone 1.0 no reg reg no reg reg
1278 1292 2293 2304
Drystone 1.1 1084 1094 1957 1963
Buchholz (sum of user & sys times in sec)
short cpu 0.3 0.2
medium cpu 3.3 1.9
long cpu *** *** (values out of range)
short I/O 0.9 0.6
I/O bound 3.1 1.9
long mixed 56.9 33.8
--
Steve Dyer
dyer@harvard.harvard.edu
dyer@spdcc.COM aka {ihnp4,harvard,linus,ima,bbn,halleys}!spdcc!dyer