Re: CPU time for Solve under v.2.0 and 2.1
- To: mathgroup at yoda.physics.unc.edu
- Subject: Re: CPU time for Solve under v.2.0 and 2.1
- From: yoda at ele.crl.melco.co.jp
- Date: Tue, 17 Nov 92 15:18:09 JST
Dear Dave,
thanks for your prompt reply. Today I did further investigation about
LinearSole and Solve
for a linear matrix equation solving. A new test matrix is a 8 x 8 sparse
symmetric one.
Here's my conclusion:
(1) v.2.1 always gives much longer expressions than v.2.0 does. In my 8 x 8
matrix case,
the length ratio is over 10 fold!!!
(2) So even though the CPU time only for Solve and LinearSolve may be a bit
faster under v.2.1,
the total time including the following Cancel or Simplify operation is
much longer
under v.2.1.
Test results summary
v.2.0
--------------------------------
Solve Cancel Total
5 sec 7 sec 12 sec
--------------------------------
LinearSolve Cancel Total
6 sec 4 sec 10 sec
--------------------------------
v.2.1
--------------------------------
Solve Cancel Total
7 sec too long to wait !!
--------------------------------
LinearSolve Cancel Total
3 sec 62 sec 65 sec
--------------------------------
Is it possible to use v.2.0's Solve and LinearSolve under v.2.1
Mathematica?
The reason is the overall speed of v.2.1 except the Solve or LinearSolve
looks much faster!
Regards,
Kiyoshi
***************************************
* Kiyoshi Yoda, Ph.D *
* *
* Mitsubishi Electric Corporation *
* Central Research Laboratory *
* 8-1-1 Tsukaguchi-Hommachi *
* Amagasaki, 661 Japan *
* *
* yoda at ele.crl.melco.co.jp *
* domestic phone: 06-497-7130 *
* domestic fax : 06-497-7288 *
* international fax: +81-6-497-7288 *
***************************************