|
2008-11-04, 01:52
|
#12 | |
May 2007
Kansas; USA
101·103 Posts |
Quote:
Karsten, Yep, he's correct here. Not only is 12*15^1-1 prime, in analyzing all k up to 100 that have a one's digit of "2", there should also be: 32*15^1-1 is prime 62*15^1-1 is prime 82*15^1-1 is prime k=22 & 92 have a trivial factor of 7 so are not considered, k=42 has a factor of 17, k=52 has a factor of 19, and k=72 has a factor of 13. Gary |
|
|
|
|
2008-11-04, 11:13
|
#13 |
Mar 2006
Germany
55338 Posts |
now i see what you meant.
i run this script with high n values (>2000) and with srsieve sieved files. therefore this error not occured so far (have to check my posted results for Riesel base 15). i mentioned to use another script (with pfgw) for n from 1 to about 1000 or 5000 to create the input file for these scripte described here. i changed both (awk-script and batch) for processing the input file from henryzz. works fine now. download it from the link in post #1 again and test. thanks David and Gary |
|
|
|
2008-11-04, 13:17
|
#14 | |
|
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
16F916 Posts |
Quote:
 it appears to be fixed the new num_seq file is not in the readme file you really have done a good job with this script it will save a lot of manual intervention and time lost |
|
|
|
|
|
2008-11-05, 08:54
|
#15 | |
|
Mar 2006
Germany
1011010110112 Posts |
Quote:
Code:
(...) Put the number of k's in the file 'num_seq' (see example included). So you only need really 4 files to start/continue the whole work! The Log-file ------------- (...) BTW. there's a new edit: i changed the initial script 'do.bat' with a parameter: the base (here base=15) Code:
call do_base 15 so the only change when you do another base-file is editing the base in the 'do.bat'!!! nothing else changed. (downloadable from link in post #1) PS: i used the file-names because cLLR and srsieve will autoatically create these names (when a llr.ini is not present). so no need to create a LLR-result or -ini file at first run! Last fiddled with by kar_bon on 2008-11-05 at 08:57 |
|
|
|
|
|
2008-11-05, 09:03
|
#16 |
|
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
5,881 Posts |
i searched the readme for it before i posted and coundnt find it
i just searched again and found it i dont know what happened before |
|
|
|
|
2008-11-06, 11:35
|
#17 |
|
Mar 2006
Germany
32×17×19 Posts |
i've now uploaded batch-files for low n for testing with pfgw:
see http://www.rieselprime.org/dl/Automated_low_n.zip Example: for Riesel base 15 create a file with: ABC2 $a*15^1-1 a: from 2 to 1000000 step 2 for first input in pfgw. all even k's from 2 to 1M will be tested here for n=1. the scripts automatically save the resultfiles, primes/PRP's found and the input for the next n. there's also a log file written like: 2008/11/05 23:48:52 n=1 : 121370 primes, 0 PRPs, 378630 remain 2008/11/05 23:48:53 k==1 mod 7: 307201 remain 2008/11/05 23:49:41 n=2 : 75007 primes, 0 PRPs, 232194 remain 2008/11/05 23:50:31 n=3 : 49582 primes, 0 PRPs, 182612 remain 2008/11/05 23:51:32 n=4 : 2754 primes, 32030 PRPs, 147828 remain 2008/11/05 23:52:23 n=5 : 95 primes, 25306 PRPs, 122427 remain (...) i also patched the included pfgw.exe to speed up the output by supressing some textmessages. all is described in the ReadMe.txt and examples included. please test and say you opinion. so this is the first step to test a whole base with one script. the described batches in post #1 are the next step after this. i will combine both steps with creating a file for srsieve and automated sieving for higher n say n=100k or so. one step away from testing/proving a whole base with one batch-call!!! Karsten |
|
|
|
|
2008-11-07, 06:58
|
#18 | |
|
May 2007
Kansas; USA
101×103 Posts |
Quote:
Absolutely brilliant! I'm going to get it loaded on my machines and do some testing with it this weekend. After understanding it better, I think I'll use it to continue my Sierp base 31. There are ~2200 k's remaining at n=~8300 and it is an extremely prime base. I have it sieved to n=10K and was going to stop and unreserve it there. If I get your process running correctly, I now think I may sieve it up to n=50K, test it myself to n=~20K-30K, and leave the rest of the sieve file for the group. I want to get it to < 1000 k's remaining, which would be excellent for a conjecture of k>6.3M. My estimate is that it should happen at n=~25K-35K. Very prime bases like this are fun IF the conjecture isn't too high! Bases 3, 7, and 15 just make my head spin. lol Gary Last fiddled with by gd_barnes on 2008-11-07 at 06:59 |
|
|
|
|
|
2008-11-07, 10:54
|
#19 |
|
Mar 2006
Germany
55338 Posts |
i uploaded a slightly changes awk-script: the base was set to '15' and i edited this.
BTW: these scripts are for Riesel-side, i have to change some things to make it for Sierpinski's! To make it clearly: the scripts in post #1 are best for sieved files, so n>10k or higher. the scripts in previous post are for the beginning with low n from 1 to 1000 or so! |
|
|
|
|
2008-11-07, 23:08
|
#20 | ||
Jan 2005
1110111112 Posts |
Can you make a simple example for me for it to run on Riesel base 3?
It appears to me that there are too many k's left at my trial run, but that might be because of me bugging up the input file :> I used Quote:
and Quote:
The modular removed 1 (one) candidate... |
||
|
|
|
|
2008-11-08, 22:26
|
#21 | |
|
I quite division it
"Chris"
Feb 2005
England
31×67 Posts |
Quote:
Works fine here and produces times similar to Geoff's builds, i.e. about 30% faster than LLR. (C2Quad, XP 32bit, tested with Riesel base 27 file.) |
|
|
|
|
|
2008-11-10, 12:35
|
#22 | |
|
Mar 2006
Germany
32·17·19 Posts |
Quote:
Code:
@echo off call doranges 3 1 1000 there's no need to delete pairs with the 'modular'-script. see Gary's page http://gbarnes017.googlepages.com/Ri...onjectures.htm and the column for "Trivial k's (factor)": all odd k's are composite for any n, so you can put Code:
ABC2 $a*3^1-1 a: from 250000000 to 251000000 step 2 i'm testing this range and can send you the whole log-file tomorrow. first logs: Code:
2008/11/10 11:24:12 n=1 : 73441 primes, 426560 remain 2008/11/10 11:40:50 n=2 : 60548 primes, 366012 remain 2008/11/10 11:55:08 n=3 : 50537 PRPs, 315475 remain 2008/11/10 12:07:12 n=4 : 41750 PRPs, 273725 remain 2008/11/10 12:17:07 n=5 : 33662 PRPs, 240063 remain 2008/11/10 12:25:50 n=6 : 28350 PRPs, 211713 remain 2008/11/10 12:33:19 n=7 : 23047 PRPs, 188666 remain 2008/11/10 12:39:43 n=8 : 20451 PRPs, 168215 remain 2008/11/10 12:46:03 n=9 : 16941 PRPs, 151274 remain 2008/11/10 12:51:09 n=10 : 14722 PRPs, 136552 remain 2008/11/10 12:56:10 n=11 : 12838 PRPs, 123714 remain 2008/11/10 13:00:12 n=12 : 11386 PRPs, 112328 remain 2008/11/10 13:04:09 n=13 : 9149 PRPs, 103179 remain 2008/11/10 13:07:42 n=14 : 8193 PRPs, 94986 remain 2008/11/10 13:10:40 n=15 : 7643 PRPs, 87343 remain 2008/11/10 13:13:19 n=16 : 6740 PRPs, 80603 remain 2008/11/10 13:15:41 n=17 : 5801 PRPs, 74802 remain 2008/11/10 13:17:56 n=18 : 5336 PRPs, 69466 remain 2008/11/10 13:19:59 n=19 : 4656 PRPs, 64810 remain 2008/11/10 13:21:55 n=20 : 4268 PRPs, 60542 remain 2008/11/10 13:23:40 n=21 : 3724 PRPs, 56818 remain 2008/11/10 13:25:14 n=22 : 3421 PRPs, 53397 remain 2008/11/10 13:26:45 n=23 : 3164 PRPs, 50233 remain 2008/11/10 13:28:17 n=24 : 3113 PRPs, 47120 remain 2008/11/10 13:29:32 n=25 : 2442 PRPs, 44678 remain 2008/11/10 13:30:42 n=26 : 2308 PRPs, 42370 remain 2008/11/10 13:31:49 n=27 : 2265 PRPs, 40105 remain 2008/11/10 13:32:52 n=28 : 2042 PRPs, 38063 remain so i have to check which the best way for this. |
|
|
|
|
 |
| Thread Tools | |
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| Automated LLR testing with LLRnet | mdettweiler | No Prime Left Behind | 24 | 2011-11-04 19:20 |
| Automated PRP using LLRNet | axn | Sierpinski/Riesel Base 5 | 73 | 2008-11-26 03:46 |
| Automated primality testing with LLRnet | mdettweiler | Conjectures 'R Us | 18 | 2008-03-04 00:06 |
| Automated P-1 thoughts. | nucleon | Marin's Mersenne-aries | 3 | 2004-03-25 02:45 |
| Semi-automated P-1 testing | GP2 | Marin's Mersenne-aries | 2 | 2003-09-29 19:01 |
