| Re: Impressions of IDL on PowerMac vs. Sparc [message #3535 is a reply to message #3471] |
Thu, 02 February 1995 12:52   |
walsteyn
Messages: 3
Registered: December 1993
|
Junior Member |
|
|
In <Pine.CVX.3.90.950131111953.24952A-100000@sstcx1.lanl.gov> Jeff Bloch <jbloch@sstcx1.lanl.gov> writes:
> We have just started playing with IDL on a Power Mac (Quadra 650 with Power
> Mac upgrade card, (6100/66 equivalent) with 40MB of memory) and comparing it
> to IDL running on a Sparc LX. We have found some very interesting speed
> differences. Simple large array operations are a factor of two FASTER on the
> Power Mac than on the Sparc, yet transcendental function array operations
> (sin, asin, tan, etc) appear to be a factor of two SLOWER on the Power Mac.
You might get a higher speed on the PowerMac if you would install
a new floating point math library (made by Apple). It's called MathLib.
Ask or read about it in a comp.sys.mac.* newsgroup. (I don't know the
details as I don't have a PowerMac. The only thing I do know is that
the MathLib in ROM is ``slow'' when it comes to transcendental functions...
The software patch, i.e., the MathLib extension, solves the speed problem.)
Good luck,
Fred. (walsteyn@fys.ruu.nl)
> Array operations using sqrt() also appear to be faster on the Power Mac.
> Operations using the convol() function are also much faster on the Power Mac.
> On the other hand, running the standard IDL demo on each shows the Power Mac
> running the demo ~40% slower. The Sparc LX did not page or swap during the
> tests.
> ------------------------------------------------------------ --------------
> Jeffrey Bloch Office: (505) 665-2568
> Astrophysics and Radiation Measurements Group ALEXIS Soc: (505) 665-5975
> Los Alamos National Laboratory FAX: (505) 665-4414
> Group NIS-2, Mail Stop D436 e-mail: jbloch@lanl.gov
> Los Alamos, NM 87545
> ------------------------------------------------------------ --------------
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
