[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: benchmark 1.5
From: |
Jarle Brinchmann |
Subject: |
Re: benchmark 1.5 |
Date: |
Thu, 11 Apr 1996 12:03:19 +0100 |
John L. Daschbach writes:
>
> First:
> benchmark.m v1.5 fails for me.
> error: called from `mean' in file
> `/msrc/home/d3h486/Lib/Octave/Contrib/statistics/mean.m'
This is because you are using the mean.m from the contributed
statistics package. If you use the bultin mean (ie. rename the mean
above or change your loadpath), then it will work.
---
Here are benchmarks for my 486DX4-100, running Linux 1.2.13 (portable
with 16 Mb RAM):
Speed of octave 1.1.1 on i486-unknown-linux relative to Sun Sparc 10/40
Matrix inversion (LAPACK) 0.50 +/- 0.5% (7 runs)
Schur decomposition (LAPACK) 0.44 +/- 0.1% (7 runs)
Differential equation (LSODE) 0.67 +/- 0.6% (7 runs)
Fourier transforms (FFTPACK) 0.51 +/- 0.2% (7 runs)
for loop 0.68 +/- 2.5% (7 runs)
-- Performance index (bm 1.5): 0.55
Cheers,
Jarle.