I thought I had reviewed that ticket, but apparently I have tried it out but not yet reported.
As wjp says on the ticket, to try it out you only need to install the new spkg and the second patch. But to adopt this spkg as part of Sage proper would need a vote here. I suggest that wjp helps that process by collecting some data (before and after). For example: sage: time P = prime_range(10^8) CPU times: user 1.83 s, sys: 0.50 s, total: 2.32 s Wall time: 2.33 s sage: len(P) 5761455 but this does not use the new PrimeGen class. I tried this (withe the new spkg + patch): sage: pg=Primes().pg sage: pg.reset() sage: N=pg.count(10^8) sage: pg.reset() sage: time P=[pg.next() for _ in range(N)] CPU times: user 4.98 s, sys: 0.03 s, total: 5.01 s Wall time: 5.02 s which is slower but it's using a more stupid method to collect the primes that in prime_range(). John 2009/7/26 Minh Nguyen <nguyenmi...@gmail.com>: > > Hi folks, > > Ticket #3925 > > http://trac.sagemath.org/sage_trac/ticket/3925 > > is adding Bernstein's primegen C program as a standard SPKG. A patch > on that ticket modifies the file > > SAGE_ROOT/spkg/standard/deps > > At the moment, that file is not under revision control so one would > need to apply the patch by hand. If that file come under the revision > control of Mercurial, then all files ending with ".spkg" (all the > standard SPKG's in that directory) would need to be ignored by > Mercurial. Any thoughts on this matter? > > PS: You're welcome to review the ticket as it speeds up certain number > theoretic computations. > > -- > Regards > Minh Van Nguyen > > > > --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
