mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > Riesel Prime Search

Reply
 
Thread Tools
Old 2008-01-25, 17:17   #155
Xyzzy
 
Xyzzy's Avatar
 
"Mike"
Aug 2002

7·29·37 Posts
Default

Please play nice. This is supposed to be fun.

Xyzzy is offline   Reply With Quote
Old 2008-01-25, 20:16   #156
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

3×2,083 Posts
Default

Quote:
Originally Posted by VBCurtis View Post
Well put, MooooMoo. It seems rather than address this, or wait for me to email him any sieved file, Gary has decided to use an old file from the 50T range when we were first starting to sieve k=5. He helped with the sieve for a couple T, and has deemed himself ready to continue the search as far as he feels like. I view this basically as using a stolen sieve file. I can't imagine someone helping a team-sieve and then deciding to use the file for his own purposes.... that is not the behavior of a team member, period.

In return, I will no longer send Gary sieved files from k=5, and like Kosmaj will no longer recognise his work within RPS. Gary, please leave RPS and k<300 alone. You are a stubborn, deceitful individual who runs over anyone who does not conform to your sensibility. You joined an effort with established methods for dealing with many things, and tried to force your own judgement of right and wrong, acceptable time, etc. Rather than adopt a "when in rome, do as the Romans do", you did all in your power to force us to do it your way. You stole k=243 against the wishes of organizers of the project, badgered members about their reservations as if you ran the show, and disregarded any feedback that did not support your personal goals. I do not want to be associated with such a poor team player.

Anyone who wishes to continue on k=5, please PM me or post here for the next available range. I will continue work on this k on my own, and send files as requested by others.
-Curtis
Why not simply post the sieved files in the first post of this thread like most team drives do? Then, nobody's left waiting for files to be mailed to them (and thus stuck trying to keep their computers busy in the meantime), and they thus don't have to go to an older copy of the file that they've got sitting around in order to start on their range immediately.
mdettweiler is offline   Reply With Quote
Old 2008-01-27, 04:00   #157
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

47·89 Posts
Default

Anon-
It was originally faster, as I was emailing the files as soon as the sieve was ready when we were quickly running from 500k to 1M. Direct email was faster than send the files to an admin to be posted, and reservations were made before files were ready. Also, carlos was reserving large chunks, so it seemed easier to do it via email, particularly because I do not have mod privledges, and no moderator had shown interest in k=5. Rather than ask someone to administer a project they did not start or necessarily want, I've been trying to do the next best thing via email.

Sheep has moved his considerable sieving resources on to other projects, so the sieve for k=5 is now moving very slowly. We will be out of fully-sieved files shortly, and there will be quite a delay before the sieve is ready for another batch, as I am now sieving alone on a single core. It is sieving as part of a many-k sr2sieve, so removing candidates does not make sense when the sieve will run anyway on that exponent region- we give up free sieving, as well as make the sieve itself less efficient, by continuing with less than optimally sieved files.
-Curtis
VBCurtis is offline   Reply With Quote
Old 2008-01-27, 05:07   #158
BlisteringSheep
 
BlisteringSheep's Avatar
 
Oct 2006
On a Suzuki Boulevard C90

2·3·41 Posts
Default

Quote:
Originally Posted by VBCurtis View Post
Anon-
Sheep has moved his considerable sieving resources on to other projects, so the sieve for k=5 is now moving very slowly.
If you send me an updated dat, I've got some sieve resources available that I can donate again for a while.
BlisteringSheep is offline   Reply With Quote
Old 2008-02-02, 19:13   #159
VBCurtis
 
VBCurtis's Avatar
 
"Curtis"
Feb 2005
Riverside, CA

47·89 Posts
Default

Thanks, Sheep. The dat is in the mail, along with our current progress.
-Curtis
VBCurtis is offline   Reply With Quote
Old 2008-02-03, 09:42   #160
BlisteringSheep
 
BlisteringSheep's Avatar
 
Oct 2006
On a Suzuki Boulevard C90

2·3·41 Posts
Default

Great. It'll take a couple of days to get it on the farm (home computer problems), but should be started Monday or Tuesday.
BlisteringSheep is offline   Reply With Quote
Old 2008-02-09, 04:45   #161
gd_barnes
 
gd_barnes's Avatar
 
May 2007
Kansas; USA

24·3·211 Posts
Default

k=5 1240-1260 complete; no primes

Results already mailed to Curtis.

Good luck everyone!
gd_barnes is offline   Reply With Quote
Old 2008-02-11, 19:20   #162
BlisteringSheep
 
BlisteringSheep's Avatar
 
Oct 2006
On a Suzuki Boulevard C90

2·3·41 Posts
Default

Quote:
Originally Posted by BlisteringSheep View Post
Great. It'll take a couple of days to get it on the farm (home computer problems), but should be started Monday or Tuesday.
The sieve has been running for about a week. I will have 262-325T completed by the beginning of next week. It should have been completed at the end of this week, but there was a power outage yesterday & I didn't discover it until this morning.

Once I get complete this range, I won't be able to sieve for a while. The farm will be doing OGR full time as that project approaches the end. I will also be having at least two surgeries this Spring (first is 18-March, I hope to have the second ASAP after that), and won't be able to manage/watch them as closely as I like for sieving; my d.net setup is much more automated.
BlisteringSheep is offline   Reply With Quote
Old 2008-02-11, 19:46   #163
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

186916 Posts
Default

Quote:
Originally Posted by BlisteringSheep View Post
The sieve has been running for about a week. I will have 262-325T completed by the beginning of next week. It should have been completed at the end of this week, but there was a power outage yesterday & I didn't discover it until this morning.

Once I get complete this range, I won't be able to sieve for a while. The farm will be doing OGR full time as that project approaches the end. I will also be having at least two surgeries this Spring (first is 18-March, I hope to have the second ASAP after that), and won't be able to manage/watch them as closely as I like for sieving; my d.net setup is much more automated.
Why don't you try building LLRnet for Linux/PPC? Then you could participate in NPLB or CRUS (or any other LLRnet-enabled project, for that matter), since LLRnet is just as automated as d.net.
mdettweiler is offline   Reply With Quote
Old 2008-02-11, 20:34   #164
BlisteringSheep
 
BlisteringSheep's Avatar
 
Oct 2006
On a Suzuki Boulevard C90

111101102 Posts
Default

Quote:
Originally Posted by Anonymous View Post
Why don't you try building LLRnet for Linux/PPC? Then you could participate in NPLB or CRUS (or any other LLRnet-enabled project, for that matter), since LLRnet is just as automated as d.net.
Some reasons:
  1. The LLR & gwnum source uses x86 assembly.
  2. I run d.net in a sneaker-net/NFS configuration, where I can just download a bunch of buffers to one machine and the clients eat from that.
  3. If I need to get more d.net work, there are d.net proxy servers running on port 80. I've yet to see a llrnet server running on port 80.
  4. I'd have to run a llrnet server to distribute work internally as none of the actual compute farm has non-proxied outside access (and the only outside access is generally port 80 - some also have port 22).
The d.net client includes other features, particularly "pause when running" which lets me run without any impact to anything on development machines. In fact, when sieving I always keep one core running the d.net client so that I can have a script shut down the sievers by seeing what d.net is doing (ie. if it's paused then sieving has to stop, too) just by watching the log files.
BlisteringSheep is offline   Reply With Quote
Old 2008-02-11, 20:41   #165
mdettweiler
A Sunny Moo
 
mdettweiler's Avatar
 
Aug 2007
USA (GMT-5)

141518 Posts
Default

Quote:
Originally Posted by BlisteringSheep View Post
Some reasons:
  1. The LLR & gwnum source uses x86 assembly.
  2. I run d.net in a sneaker-net/NFS configuration, where I can just download a bunch of buffers to one machine and the clients eat from that.
  3. If I need to get more d.net work, there are d.net proxy servers running on port 80. I've yet to see a llrnet server running on port 80.
  4. I'd have to run a llrnet server to distribute work internally as none of the actual compute farm has non-proxied outside access (and the only outside access is generally port 80 - some also have port 22).
The d.net client includes other features, particularly "pause when running" which lets me run without any impact to anything on development machines. In fact, when sieving I always keep one core running the d.net client so that I can have a script shut down the sievers by seeing what d.net is doing (ie. if it's paused then sieving has to stop, too) just by watching the log files.
Ah, I see now.
mdettweiler is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
15*2^n-1, n>1M Reservation Thread Kosmaj Riesel Prime Search 702 2018-04-11 07:36
LLR reservation thread (n=480K-500K) Oddball Twin Prime Search 33 2012-01-20 05:37
Octoproth Reservation Thread Greenbank Octoproth Search 2 2007-12-26 09:58
Dodecaproth Reservation Thread Greenbank Octoproth Search 30 2006-02-09 00:33
Hexadecaproth Reservation Thread (n=76) Greenbank Octoproth Search 0 2006-01-25 13:41

All times are UTC. The time now is 04:54.

Sun May 31 04:54:55 UTC 2020 up 67 days, 2:27, 1 user, load averages: 2.49, 2.21, 2.01

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.