![]() |
![]() |
#56 | |
May 2007
Kansas; USA
72×211 Posts |
![]() Quote:
|
|
![]() |
![]() |
![]() |
#57 |
Nov 2009
5368 Posts |
![]()
576*172132695-1 is prime
|
![]() |
![]() |
![]() |
#58 | |
May 2007
Kansas; USA
72·211 Posts |
![]() Quote:
On an interesting note, the Riesel side primes are far outpacing the Sierp side primes so far on this drive despite the Sierp side starting with far more k's; 63 to 51. Last fiddled with by gd_barnes on 2011-09-03 at 07:21 |
|
![]() |
![]() |
![]() |
#59 |
Jun 2008
Wollongong, .au
B716 Posts |
![]()
Using prpclient-4.3.5 for Drive #2. I'm wondering if there is a way to maintain a buffer and report each work-unit as it is completed.
My internet can be a bit flaky, and it'd be nice if I could return workunits as soon as they're done, but have a twelve-hour buffer just in case my internet goes out. Is there some combination of options in the .ini file that can achieve this? Last fiddled with by paleseptember on 2011-09-23 at 05:59 |
![]() |
![]() |
![]() |
#60 | |
A Sunny Moo
Aug 2007
USA (GMT-5)
3×2,083 Posts |
![]() Quote:
![]() You may want to suggest the feature over in Mark's PRPnet 4.x announcements/feature-requests/bug-reports thread; he would be able to give a better answer than I as to how tough it would be to implement. If it is to be pursued, I would think it might be better to (at least in the short term) fork off a second version of the client so as to leave a solid, stable single-threaded client while myriad bugs are worked out of the necessarily multi-threaded "buffered" version. In early versions of PRPnet, the idea of buffering workunits was addressed by the ability to run a "slave" server, much like an LLRnet proxy, which would keep a constant buffer of work from the main server and dole it out to clients on the local network. This feature was initially retained from ECMNet, which contributed heavily to PRPnet's initial code; but it was dropped in later PRPnet versions since it would have needed heavy rewriting to work well for PRPnet, and nobody was using it anyway (being that it didn't work ![]() Anyway...in the meantime, all I can suggest is that you keep a large queue of workunits in your clients, to reduce the number of times they have to connect to the server (and consequently, the probability that they'll attempt to connect during a time when the internet is out). For instance, if each test takes a maximum of (say) 3500 seconds to complete--factoring in a little extra wiggle room--each core should be able to safely do 40-45 or so workunits within the 48 hour limit. Depending on how long your internet usually goes out for (just a few minutes? or hours?) you may want to give a little more space to allow extra time in case the internet does happen to be out when the tests are up to be returned, so you don't risk having them expired while your internet is out. It's not a perfect solution, but does help some at least. ![]() Last fiddled with by mdettweiler on 2011-09-23 at 11:20 |
|
![]() |
![]() |
![]() |
#61 | |
"Mark"
Apr 2003
Between here and the
627910 Posts |
![]() Quote:
|
|
![]() |
![]() |
![]() |
#62 |
Jun 2008
Wollongong, .au
2678 Posts |
![]()
Thanks mdettweiler, thanks rogue. Will continue on crunching as best I can.
Have queued up a much larger buffer. Am slightly embarrassed by how much of the 'assigned work' table I'm taking up. |
![]() |
![]() |
![]() |
#63 |
Jan 2006
Hungary
26810 Posts |
![]()
Hello friends,
I've managed to start a PRP-client succesfully. I noticed one thing that makes it harder for me run it: the CMD-screens. I run this on the family PC and all the rest of my family may just kill off the sessions from the taskbar. What's an easy trick to hide the prpclients? Thanks, Willem. |
![]() |
![]() |
![]() |
#64 | |
Account Deleted
"Tim Sorbera"
Aug 2006
San Antonio, TX USA
17×251 Posts |
![]() Quote:
Another more general option is Hide-it. It can hide any window. It's a handy utility for math projects, where many tools can only run in a command window. Last fiddled with by Mini-Geek on 2011-09-25 at 20:19 |
|
![]() |
![]() |
![]() |
#66 |
May 2007
Kansas; USA
101000011000112 Posts |
![]()
Adding S148 with 1k remaining at n=150K to the drive. I will begin sieving on it in 2-3 days.
|
![]() |
![]() |
![]() |
Thread Tools | |
![]() |
||||
Thread | Thread Starter | Forum | Replies | Last Post |
Other Bases? | wblipp | GPU Computing | 50 | 2012-10-11 13:23 |
New PRPnet drive discussion | mdettweiler | Conjectures 'R Us | 89 | 2011-08-10 09:01 |
PRPnet 1st drive-R/S base 2 even-k/even-n/odd-n | mdettweiler | Conjectures 'R Us | 153 | 2011-08-10 06:54 |
Subproject #1: 210k-250k sequences to 100 digits | henryzz | Aliquot Sequences | 204 | 2009-07-30 12:06 |
Different bases = different times? | roger | Information & Answers | 1 | 2007-04-25 14:35 |