Message 75495 - Python tracker

Message75495

Author	mark.dickinson
Recipients	christian.heimes, fredrikj, mark.dickinson, pernici
Date	2008-11-04.15:37:52
SpamBayes Score	1.8995648e-05
Marked as misclassified	No
Message-id	<1225813076.42.0.335183613096.issue3944@psf.upfronthosting.co.za>
In-reply-to

Content
I've opened a separate issue (issue 4258) for the idea of using 30-bit long digits instead of 15-bit ones. I'll remove the patch from this issue. Pernici Mario's idea applies even better to base 2**30 longs: one can clump together 16 (!) of the multiplications at once, essentially eliminating the overhead of shifts almost completely.

Content

I've opened a separate issue (issue 4258) for the idea of using 30-bit
long digits instead of 15-bit ones.  I'll remove the patch from this 
issue.

Pernici Mario's idea applies even better to base 2**30 longs:  one can
clump together 16 (!) of the multiplications at once, essentially
eliminating the overhead of shifts almost completely.

History
Date	User	Action	Args
2008-11-04 15:37:56	mark.dickinson	set	recipients: + mark.dickinson, pernici, christian.heimes, fredrikj
2008-11-04 15:37:56	mark.dickinson	set	messageid: <1225813076.42.0.335183613096.issue3944@psf.upfronthosting.co.za>
2008-11-04 15:37:52	mark.dickinson	link	issue3944 messages
2008-11-04 15:37:52	mark.dickinson	create