"William Stein" <[EMAIL PROTECTED]> writes: > On 5/16/07, Michel <[EMAIL PROTECTED]> wrote: >> >> Consider >> >> sage: x,y,z=QQ['x','y','z'].gens() >> sage: u,v,w=W['u','v','w'].gens() > > Syntax error -- W is not defined. What is W? > >> sage: %timeit ((x+y+z)**3)(1/y,y,y**100) >> 10 loops, best of 3: 752 ms per loop <=== look here! >> sage: %timeit ((u+v+w)**3)(1/v,v,v**100) >> 10 loops, best of 3: 48.2 ms per loop <=== look here! >> >> Conclusion: rational functions over the rationals are unusably slow at >> this time. >> I even know what the reason is. Rational functions over W do not >> support reduce so it >> is not called during the intermediate computations which happens to be >> a blessing. >> >> William: I know you can easily fix this yourself. If you are too busy >> would you be interested in a patch? I think this kind of janitorial >> work is important to improve the predictability >> of sage behaviour. > > It's not at all clear to me without further discussion and benchmarking > what the right optimization is so that rational functions over fields are > sufficiently fast. I definitely wouldn't make any changes > to this without some discussion. Basically, are you suggesting that one > not call reduce until, e.g., printing? Or at least until some sort of > operations that require the thing be reduced? When do you want > to call reduce?
You can't let things pile up indefinitely; this was a problem with quotient fields that was fixed a while back. The intermediate values get large, especially when powering, so that the final reduce is ridiculously expensive. I'm not certain what the right behaviour is, but I don't like the idea of trying to have tricky behaviour to reduce only at the right moment. Always or never, I say. Nick --~--~---------~--~----~------------~-------~--~----~ 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://sage.scipy.org/sage/ and http://modular.math.washington.edu/sage/ -~----------~----~----~----~------~----~------~--~---
