snewpgen improvement very much appreciated :)

jasong · 2007-11-23, 00:58

I recently downloaded a new copy of the Linux newpgen(static version). When the new version is sieving(the Linux version is totally text-based) it only adds lines if it finds a factor. If all it needs to do is update how high p is, it just overwrites the line. Originally, it just kept adding lines, and if there was more than about 5 minutes between the time you looked and the last factor, you had no idea how efficient the sieving was at that point.

This improvement is very much appreciated. So appreciated in fact that I want him to know how thrilled I am. If I could write him a short note, preferably a PM rather than an email(email is more personal, but he might feel like his privacy is being invaded) I could tell him how happy I am that it's fixed. (I'm not stalker happy, it's just the way things were bugged the crap out of me)

paulunderwood · 2007-11-23, 02:12

I just got the latest newpgen, did:

Code:

./newpgen -ho

and saw:

Code:

-ol :         Append the removed k's or n's in a file.

So you could run newpgen up to say p=10M, then stop it and restart it with logging. Use:

Code:

tail -f your_log_file

HTH

jasong · 2007-11-23, 03:10

lol, Paul, I think you may have been in a turkey coma when you wrote that. The whole point of my post was that that particular problem has been SOLVED by the version I downloaded and I wanted to thank the programmer.

paulunderwood · 2007-11-23, 04:06

2007-11-23, 00:58	#1
jasong "Jason Goatcher" Mar 2005 3·7·167 Posts	snewpgen improvement very much appreciated :) I recently downloaded a new copy of the Linux newpgen(static version). When the new version is sieving(the Linux version is totally text-based) it only adds lines if it finds a factor. If all it needs to do is update how high p is, it just overwrites the line. Originally, it just kept adding lines, and if there was more than about 5 minutes between the time you looked and the last factor, you had no idea how efficient the sieving was at that point. This improvement is very much appreciated. So appreciated in fact that I want him to know how thrilled I am. If I could write him a short note, preferably a PM rather than an email(email is more personal, but he might feel like his privacy is being invaded) I could tell him how happy I am that it's fixed. (I'm not stalker happy, it's just the way things were bugged the crap out of me)

2007-11-23, 02:12	#2
paulunderwood Sep 2002 Database er0rr 2·11·191 Posts	I just got the latest newpgen, did: Code: ./newpgen -ho and saw: Code: -ol : Append the removed k's or n's in a file. So you could run newpgen up to say p=10M, then stop it and restart it with logging. Use: Code: tail -f your_log_file HTH Last fiddled with by paulunderwood on 2007-11-23 at 02:14

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Trial Divison Improvement	carpetpool	Information & Answers	1	2016-11-30 03:48
A new Fermat Factorization improvement?	siegert81	Factoring	5	2011-06-13 02:58
performance improvement with assembly	bsquared	Software	15	2010-09-28 19:00
Possible improvement of quadratic sieve	Random Poster	Factoring	4	2010-02-12 03:09
Possible idea for improvement	Kevin	Software	1	2002-08-26 07:57

2007-11-23, 03:10	#3
jasong "Jason Goatcher" Mar 2005 DB3₁₆ Posts	lol, Paul, I think you may have been in a turkey coma when you wrote that. The whole point of my post was that that particular problem has been SOLVED by the version I downloaded and I wanted to thank the programmer.

2007-11-23, 04:06	#4
paulunderwood Sep 2002 Database er0rr 2×11×191 Posts