Hi Williams, On Jan 22, 8:52 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > On Jan 21, 2008 11:36 PM, Francois <[EMAIL PROTECTED]> wrote: > > > > > > > On Jan 22, 8:04 pm, mabshoff <[EMAIL PROTECTED] > > dortmund.de> wrote: > > > > > On the funny side of things my regular copy fails the calculus.py > > > > test > > > > because it takes too much time, my ebuild one passes it without > > > > problem. > > > > Ok, that is odd. Does it timeout or hang? > > > Timeout: > > sage -t devel/sage-main/sage/calculus/calculus.py *** *** > > Error: TIMED OUT! *** *** > > *** *** Error: TIMED OUT! *** *** > > [182.4 s] > > > OF course the ebuild one may have some extra optimizations from > > the Gentoo C(XX)FLAGS that I didn't kill and the test is run by root > > and not a regular user. > > > For BLAS I use ATLAS-3.8.0 I don't think there is any differences > > between the sources used by Gentoo and sage (tell me if I am > > wrong) you have the Gentoo patch and we also have the > > pentium-M patch. The only difference I can think of is one was > > compiled by gfortran and the other by g95. > > The calculus.py module likely doesn't use ATLAS at all. The only thing > that would have much of an impact on speed would be Maxima, probably. > > What hardware are you using exactly? On a 2.6Ghz machine I get: > > teragon:calculus was$ sage -t calculus.py > sage -t calculus.py > [44.3 s] > > What happens if you do > > sage -t --verbose calculus.py > > i.e., do you visibly see any doctests taking a really long time? For > me the main > time takers are some plots (this should be optimized soon, actually.) > > I seem to vaguely recall maybe you are using a different version of Maxima > than the one in Sage? Maybe the newer Maxima is simply faster at some > key operations, or maybe we compile clisp or maxima in some stupid > way that makes it slower (that would be _very_ interesting to get to the > bottom of, it is the case). > At the moment I conduct tests on my home machine rather than on a computer at work (Massey university in New Zealand) but apparently it should be similar: AMD Athlon(tm) XP 2600+
Tonight the test passed both in verbose and non-verbose form. Must have been a glitch or something going on on my computer at the time. I used maxima 5.13 as 5.14 will fail this test amongst other, but yes for some reason when I tested 5.14 (outside of sage) it felt faster but I haven't try to quantify it - so take that with a big grain of salt. Francois --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
