[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Date: Thu, 1 Sep 88 12:11:40 EDT
From: apl_aimh@jhunix.HCF.JHU.EDU (Marty Hall)
Also: have there been MacIvory <-> microExplorer benchmarks? How
about Explorer II <-> XL 400 benches?
There are no public benchmarks available for either machine yet, as far
as I can tell. For MacIvory, benchmarks would be premature since the
delivered machines are supposed to run at a considerably faster clock
rate than the machines they had at AAAI last week. In the case of the
MicroExplorer, presumably benchmarks have been run, but to my knowledge
the results have not yet been made available to the public.
I'm not sure where this fallacy came from that the AAAI boards are
running at a slow clock rate. While it's true that future versions of
MacIvory could be made to run at a faster clock rate, we're shipping the
same boards we were showing at AAAI.
Which is not to say that there won't be speedups. There are still a lot
of things which can be sped up, and many of these will be sped up for
the release in December. Currently it's just too risky to put them in.
As for the current version, Symbolics officially says the performance is
about the same as a 3620. A number of people ran their own benchmarks
on MacIvories at AAAI, and the quoted performance ranged from 2/3 of a
3630 to 25% faster than a 3675. Somebody walked over with Gabriels from
a Explorer II+, and on those benchmarks the Explorer II+ came in from
1.5x to 2.5x the MacIvory. (Remember that's comparing their high end
with our low end.)
Which all goes to show you what good benchmarks are.