mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   NFS@Home (https://www.mersenneforum.org/forumdisplay.php?f=98)
-   -   BOINC NFS sieving - NFS@Home (https://www.mersenneforum.org/showthread.php?t=12388)

wombatman 2013-10-25 17:59

Absolutely. You can PM as well if you'd like.

kracker 2013-10-26 01:06

Many boards(I think most) have the option to turn off HT in BIOS.

wombatman 2013-10-26 01:41

Most of them do. This is a laptop, though, so it may be a bit more locked down. I went through all the BIOS options, and I couldn't even find things like the option to change RAM timings.

wombatman 2013-10-26 03:50

W_2_737 factors as:

[CODE]prp55 factor: 3639948209785681640330982423344863288901552433435622717
prp121 factor: 9667131945953633669529391194502837192293510216854828896126077765685195914196332441160274502127300016683619846239468417937[/CODE]

Moving on to GC_3_465!

VictordeHolland 2013-10-26 19:47

GC_8_246 is ready:
[code]
prp64 factor: 7450391778732693727105503441634852941715710901826974899928715291
prp132 factor: 778676426970473145106077221395200338755770932585955845450149945184597427041012733530986228927830026349766204465941888300244625682623
[/code]
For the statistic junkies (including myself) some highlights of the log:
[code]
Sat Oct 26 00:55:04 2013 Msieve v. 1.51 (SVN Official Release)
Sat Oct 26 00:55:04 2013 factoring 5801444449793761902554310998722384964736331243283895289517387025395045627296796552343499559511025101275657028584214475441669022037825124493702315062835511260606214091477648754354144392100293088293 (196 digits)
Sat Oct 26 00:55:05 2013 commencing relation filtering
Sat Oct 26 00:55:05 2013 estimated available RAM is 8145.3 MB
Sat Oct 26 00:55:05 2013 commencing duplicate removal, pass 1
Sat Oct 26 01:40:29 2013 found 19165304 hash collisions in 128889877 relations
Sat Oct 26 01:41:02 2013 added 1217912 free relations
Sat Oct 26 01:41:02 2013 commencing duplicate removal, pass 2
Sat Oct 26 01:46:31 2013 found 15702994 duplicates and 114404795 unique relations
Sat Oct 26 02:31:35 2013 memory use: 3083.1 MB
Sat Oct 26 02:45:40 2013 RelProcTime: 6635

Sat Oct 26 02:45:40 2013 commencing linear algebra
Sat Oct 26 02:55:57 2013 matrix is 4651773 x 4651951 (1385.7 MB) with weight 429481219 (92.32/col)
Sat Oct 26 02:55:57 2013 sparse part has weight 312094378 (67.09/col)
Sat Oct 26 02:55:57 2013 saving the first 48 matrix rows for later
Sat Oct 26 02:55:58 2013 matrix includes 64 packed rows
Sat Oct 26 02:55:59 2013 matrix is 4651725 x 4651951 (1343.9 MB) with weight 340905364 (73.28/col)
Sat Oct 26 02:55:59 2013 sparse part has weight 305764606 (65.73/col)
Sat Oct 26 02:56:16 2013 commencing Lanczos iteration (3 threads)
Sat Oct 26 20:20:26 2013 BLanczosTime: 63286

Sat Oct 26 20:20:26 2013 commencing square root phase
Sat Oct 26 21:03:48 2013 reading relations for dependency 3
Sat Oct 26 21:03:49 2013 read 2327042 cycles
Sat Oct 26 21:03:52 2013 cycles contain 6230122 unique relations
Sat Oct 26 21:09:35 2013 read 6230122 relations
Sat Oct 26 21:10:09 2013 multiplying 6230122 relations
Sat Oct 26 21:17:03 2013 multiply complete, coefficients have about 196.70 million bits
Sat Oct 26 21:17:05 2013 initial square root is modulo 11453653
Sat Oct 26 21:25:48 2013 sqrtTime: 3922

Sat Oct 26 21:25:48 2013 prp64 factor: 7450391778732693727105503441634852941715710901826974899928715291
Sat Oct 26 21:25:48 2013 prp132 factor: 778676426970473145106077221395200338755770932585955845450149945184597427041012733530986228927830026349766204465941888300244625682623

Sat Oct 26 21:25:48 2013 elapsed time 20:30:44
[/code]Taking [B]GC_12_206[/B] next.

wombatman 2013-10-27 22:47

GC_3_465 factors as:
[CODE]prp84 factor: 997449132784140554089559851778056222463774021735548392543653838242
777809872509410633
prp108 factor: 105147348859843817402740035407465879164708293212673130156267228488506925302251314718802799385933747532241997
elapsed time 27:19:55[/CODE]

I'll finish off GC_7_263.

VictordeHolland 2013-10-28 01:32

I'm having issues post-processing GC_12_206. First of all, I can't download GC_12_206.poly from NFS@home. I tried running it without the poly and this happened:
[CODE]
<1000+ reading relation errors>
error -11 reading relation 75446975
error -5 reading relation 75446976
error -1 reading relation 75446977
error -9 reading relation 75446978
error -1 reading relation 75446979
error -5 reading relation 75446980
error -1 reading relation 75446981
error -5 reading relation 75446982
error -5 reading relation 75446983
modsqrt_1 failed[/CODE]After that MSIEVE either shut down by itself, or windows states the program stops responding. Happens on both the official 1.51 version and MSIEVE 1.52 SVN939 (win64_i7). How should I proceed?

wombatman 2013-10-28 03:29

If you're using the factmsieve.py script, it will make the .poly file for you. I'm also getting the modsqrt_1 failure for GC_7_263. I'm going to try redownloading the relations file tomorrow and see if I get the same error.

frmky 2013-10-28 06:21

You can first use remdups on the file. If you are using linux, a binary is in the download directory.

zcat rels.dat.gz | remdups4 800 -v > msieve.dat

I don't have a Windows binary unfortunately...

wombatman 2013-10-28 16:09

As it turns out, if you have MinGW (or MinGW-64), you don't need a binary! I copied the source from here: [url]http://dubslow.tk/random/remdups4.c.txt[/url]

Compiled with [CODE]gcc -O3 -o remdups remdups.c[/CODE] and it worked without issue. Currently starting on GC_7_263!

Edit: Seems I spoke too soon. Now I'm getting the "wants 1000000 more relations" message. I guess it's suspicious that the 12+GB uncompressed relations file went to ~3GB after processing through remdups.

wombatman 2013-10-28 17:23

Here's what I get running your command (limited to the end part, of course), Frmky:

[CODE]Mon Oct 28 12:20:16 2013 31.5M unique relns 1.59M duplicate relns (+0.05M, avg D/U ratio in block was 9.4%)
Found 31809992 unique, 1617420 duplicate (4.8% of total), and 181942 bad relations.
Largest dimension used: 147 of 800
Average dimension used: 97.1 of 800
Terminating program at Mon Oct 28 12:20:18 2013[/CODE]

Does that seem right to you?


All times are UTC. The time now is 23:11.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.