[QUOTE=debrouxl;284774]That's what I used for 1049_67_minus1 (marginally easier than 1061_67_minus1), and it took less than 15h to produce ~2.43M cycles and manipulate them :wink:[/QUOTE]

But I am using a slower processor, probably it will take 48 hours.

[QUOTE=pinhodecarlos;284771]I'm going for 1061_67_minus1. Downloading the files right now. It will be run on a dual-core laptop...lol[/QUOTE]

First run and

[code]commencing linear algebra
matrix starts at (0, 0)
matrix is 3018149 x 3018326 (848.1 MB) with weight 262691566 (87.03/col)
sparse part has weight 204224339 (67.66/col)
saving the first 48 matrix rows for later
matrix includes 64 packed rows
matrix is 3018101 x 3018326 (814.2 MB) with weight 210063161 (69.60/col)
sparse part has weight 195336257 (64.72/col)
using block size 65536 for processor cache size 2048 kB
commencing Lanczos iteration (2 threads)
memory use: 721.3 MB
error: cannot open matrix checkpoint file[/code]

Did you (or a script) run with -ncr instead of -nc2?

[QUOTE=jasonp;284787]Did you (or a script) run with -ncr instead of -nc2?[/QUOTE]

Nope, -nc. Now it is running. LA ETA of 62 hours!

[code]Wed Jan 04 18:33:04 2012 commencing linear algebra
Wed Jan 04 18:33:09 2012 read 3019978 cycles
Wed Jan 04 18:33:33 2012 cycles contain 9114608 unique relations
Wed Jan 04 18:39:35 2012 read 9114608 relations
Wed Jan 04 18:40:37 2012 using 20 quadratic characters above 268434714
Wed Jan 04 18:43:59 2012 building initial matrix
Wed Jan 04 18:52:47 2012 memory use: 1034.5 MB
Wed Jan 04 18:54:07 2012 read 3019978 cycles
Wed Jan 04 18:54:10 2012 matrix is 3019801 x 3019978 (848.2 MB) with weight 262736336 (87.00/col)
Wed Jan 04 18:54:10 2012 sparse part has weight 204236059 (67.63/col)
Wed Jan 04 18:56:32 2012 filtering completed in 2 passes
Wed Jan 04 18:56:35 2012 matrix is 3018135 x 3018312 (848.1 MB) with weight 262690250 (87.03/col)
Wed Jan 04 18:56:35 2012 sparse part has weight 204223027 (67.66/col)
Wed Jan 04 18:58:49 2012 matrix starts at (0, 0)
Wed Jan 04 18:58:53 2012 matrix is 3018135 x 3018312 (848.1 MB) with weight 262690250 (87.03/col)
Wed Jan 04 18:58:53 2012 sparse part has weight 204223027 (67.66/col)
Wed Jan 04 18:58:53 2012 saving the first 48 matrix rows for later
Wed Jan 04 18:58:58 2012 matrix includes 64 packed rows
Wed Jan 04 18:59:00 2012 matrix is 3018087 x 3018312 (814.2 MB) with weight 210061592 (69.60/col)
Wed Jan 04 18:59:00 2012 sparse part has weight 195335017 (64.72/col)
Wed Jan 04 18:59:00 2012 using block size 65536 for processor cache size 2048 kB
Wed Jan 04 18:59:57 2012 commencing Lanczos iteration (2 threads)
Wed Jan 04 18:59:57 2012 memory use: 721.3 MB
Wed Jan 04 19:02:06 2012 linear algebra at 0.1%, ETA 68h19m
Wed Jan 04 19:02:47 2012 checkpointing every 50000 dimensions[/code]

[QUOTE=Mathew;284449]I would like to reserve 431_83_minus_1[/QUOTE]

Complete
[CODE]prp57 factor: 515090667570374099656200578489749950086519577424982516573
prp155 factor: 12440838630604861614186197662352098341051650240869753465953851868829374960570616360441981243989950629822512303096129139864643185605863965699226722217396599[/CODE]

Mathew, 698419_37_minus1 and 1051_71_minus1 are available to post-process as 479_79_minus1 (28 bits task) but I think the latter still needs more relations.

1061_67_minus1 can be moved to to the "postprocessing" state
LA ETA for 1061_67_minus1 is now 21 hours but for 1021_79_minus1 is 69 hours.

Taking 479_79_minus1.

[QUOTE=pinhodecarlos;285247]Taking 479_79_minus1.[/QUOTE]

Still needs more relations. I think relations should be increased up to 40M.

[QUOTE=frmky;284461]Worse than that, the gz file is corrupt. I only got 98.5 million valid unique relations from it before the error. gzrecover unfortunately didn't help.[/QUOTE]

I finally had time to take another look at this. Modifying gzrecover to reject extended ascii output and with judicious use of split and tail to avoid zlib segfaults on malformed data, I extracted 157M unique relations from the file. I've started filtering now.

[QUOTE=frmky;285319]I finally had time to take another look at this. Modifying gzrecover to reject extended ascii output and with judicious use of split and tail to avoid zlib segfaults on malformed data, I extracted 157M unique relations from the file. I've started filtering now.[/QUOTE]
Yay, that was enough.
linear algebra completed 16186 of 12269808 dimensions (0.1%, ETA 77h10m)
:smile: