I've been checking over the benchmarks as a lead up to the 2010.09 release. We're in a good way compared to both 4.4.4 and 4.5.1 for most non-trivial tests. * pybench is 10.9 % faster than 4.4.4 and 7.7 % faster than 4.5.1. * linpack is 46.4 % faster than 4.4.4 and the same as 4.5.1. * ffmpeg h.264 video decode (with hand written assembler versions turned off) is 15.4 % faster than 4.4.4 and 1.2 % faster than 4.5.1.
All results are statistically invalid and against poor workloads, but I'll work on that.
See http://ex.seabright.co.nz/helpers/benchcompare for more.
-- Michael
Michael Hope wrote:
I've been checking over the benchmarks as a lead up to the 2010.09 release. We're in a good way compared to both 4.4.4 and 4.5.1 for most non-trivial tests.
- pybench is 10.9 % faster than 4.4.4 and 7.7 % faster than 4.5.1.
- linpack is 46.4 % faster than 4.4.4 and the same as 4.5.1.
- ffmpeg h.264 video decode (with hand written assembler versions
turned off) is 15.4 % faster than 4.4.4 and 1.2 % faster than 4.5.1.
Michael, Do we plan to do size comparison for these benchmarks?
See http://ex.seabright.co.nz/helpers/benchcompare for more.
On Wed, Sep 08, 2010, Michael Hope wrote:
See http://ex.seabright.co.nz/helpers/benchcompare for more.
<type 'exceptions.IndexError'> at /helpers/benchcompare
values is an empty list when calling square() in templates/benchcompare.html; the web backtrace seems off by some lines for some reason
Ah, this is from me mucking about on a live server. Will fix.
-- Michael
On Wed, Sep 8, 2010 at 8:51 PM, Loïc Minier loic.minier@linaro.org wrote:
On Wed, Sep 08, 2010, Michael Hope wrote:
See http://ex.seabright.co.nz/helpers/benchcompare for more.
<type 'exceptions.IndexError'> at /helpers/benchcompare
values is an empty list when calling square() in templates/benchcompare.html; the web backtrace seems off by some lines for some reason
-- Loďc Minier
All better now. I've also added OGG Vorbis encode/decode tests and bzip2 compress/decompress tests and expanded the number of runs. Over the next week I'll look at re-running the 4.4.4 and 4.5.1 builds to create a baseline.
-- Michael
On Wed, Sep 8, 2010 at 8:59 PM, Michael Hope michael.hope@linaro.org wrote:
Ah, this is from me mucking about on a live server. Will fix.
-- Michael
On Wed, Sep 8, 2010 at 8:51 PM, Loïc Minier loic.minier@linaro.org wrote:
On Wed, Sep 08, 2010, Michael Hope wrote:
See http://ex.seabright.co.nz/helpers/benchcompare for more.
<type 'exceptions.IndexError'> at /helpers/benchcompare
values is an empty list when calling square() in templates/benchcompare.html; the web backtrace seems off by some lines for some reason
-- Loďc Minier
linaro-toolchain@lists.linaro.org
-
Loïc Minier -
Michael Hope -
Yao Qi