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

Re: XL1200 at SLUG90 [really XL1200 performance]

In message entitled "XL1200 at SLUG90" on Jun  1, Kim Cook writes
> We will have an XL1200 available at SLUG90 (Stanford University,
> June 20 - June 22).  This is an invitation to bring your 
> application to SLUG and try it out on the XL1200.  There 
> are a few restrictions on the applications we will schedule
> You are also welcome to just come and sit down and work at the XL1200
> and see for yourself the performance improvement.

We ran several unscientific benchmarks on an XL1200 at the recent AI Systems
in Gov't Conference, comparing various parameters. These benchmarks were
things like A* and RTA* programs, simple (too simple?) graphics things
like drawing thousands of lines, consing up big lists, etc.

Again, these were unscientific benchmarks, but we got 4-5X speedup
on all of them, as compared to our 3630. We were not comparing
apples with apples, since our 3630 had slightly less memory and
quite a bit less swap space (although both had an FPA), but since
the 4-5X speedup was so consistent, I still thought it would be
of interest to this group.

					- Marty Hall
hall@aplcen.apl.jhu.edu                 Artificial Intelligence Lab
hall%aplcen.apl.jhu.edu@cunyvm.bitnet   AAI Corp
..!uunet!aplcen!hall                    PO Box 126
(301) 683-6455                          Hunt Valley, MD 21030