Re: [Numpy-discussion] Numpy performance boost
Dieter, thank you for contributing a numpy mandelbrot program - but no thanks for your disqualified for doing things differently comment here. The benchmarks game has been showing a spectral-norm program based on numpy as an interesting alternative for the last couple of years - http://shootout.alioth.debian.org/u64q/program.php?test=spectralnormlang=python3id=2 - simply because I thought numpy was interesting and wanted somehow to include a numpy program without taking on the chore of dealing with a whole bunch of numpy programs. The relevant point isn't that your numpy program is shown as an interesting alternative. The relevant point is that your numpy program is shown at all. best wishes, Isaac On Aug 31, 3:58 am, Dieter Weber die...@uellue.de wrote: Hi, just wanted to show an example of how python3 + numpy compares with just python3 and many other languages and language implementations:http://shootout.alioth.debian.org/u64q/performance.php?test=mandelbro... The python3 program using numpy is #6 and you find it with the interesting alternative programs on the bottom because it was disqualified for doing things differently. It is 6.3x slower than the fastest program and well ahead of all other interpreted languages. Thanks to all contributors for making numpy such a great piece of software! Greetings, Dieter ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] Numpy performance boost
- Original Message - From: Chris.Barker chris.bar...@noaa.gov To: numpy-discussion@scipy.org Cc: Sent: Wednesday, August 31, 2011 9:08 AM Subject: Re: [Numpy-discussion] Numpy performance boost On 8/31/11 3:58 AM, Dieter Weber wrote: just wanted to show an example of how python3 + numpy compares with just python3 and many other languages and language implementations: http://shootout.alioth.debian.org/u64q/performance.php?test=mandelbrot#about hmmm - it would be interesting to see what PyPy does with this. So do it! http://shootout.alioth.debian.org/help.php#languagex Here's the nightly snapshot with source code for all the programs - https://alioth.debian.org/frs/?group_id=30402 Have fun. ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] debian benchmarks
Sturla Molden sturla at molden.no writes: It is also the kind of tasks where NumPy would help. It would be nice to get NumPy into the shootout. At least for the sake of advertising http://shootout.alioth.debian.org/u32/program.php?test=spectralnormlang=pythonid=2 ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion
Re: [Numpy-discussion] [OT] Re: debian benchmarks
Pauli Virtanen pav at iki.fi writes: -snip- It's possible that in the cases where Lua wins, the Lua code is not completely equivalent to the Fortran code, or uses stuff such as strings for which Lua's default implementation may be efficient. Note - not Lua's default implementation but LuaJIT. At least in the mandelbrot example some things differ. I wonder if Lua there takes advantage of SIMD instructions because the author of the code has manually changed the inmost loop to process two elements at once? Note - the fastest Fortran mandelbrot program is written to use OpenMP, but those u32 measurements are when the programs are forced onto one core. ___ NumPy-Discussion mailing list NumPy-Discussion@scipy.org http://mail.scipy.org/mailman/listinfo/numpy-discussion