I didn't try to hold down ^c, maybe that would work for me too. I don't know when I'll make good on my goal of running a sequence, so I don't know when I'll be able to test. Thanks,

- ben.

Could you adjust the "neat factor" feature to only appear on the DOS window when the factor is prime? I have lowered the neat factor limits so the messages actually appear sometimes, but I often get things like:

[code]c71: running P-1 at B1=22e4...
Using B1=220000, B2=658485462, polynomial x^1, x0=4236689816
Step 1 took 359ms
Step 2 took 641ms
********** Factor found in step 2: 226108288088882191380086476075187039
*** Neat 36-digit factor found: 226108288088882191380086476075187039
*** c36 = 226108288088882191380086476075187039[/code]

which are not "neat" at all.

[QUOTE=10metreh;175215]Could you adjust the "neat factor" feature to only appear on the DOS window when the factor is prime? I have lowered the neat factor limits [b]so the messages actually appear sometimes[/b], but I often get things like:[/quote]

Methinks you are a bit impatient. :smile:

Sure, fixed for the next release.

[quote=mklasson;175220]Methinks you are a bit impatient. :smile:[/quote]

Well, I suppose you're probably not going to get a neat factor with the default cut-off until your sequences are in the 120-130 digit range. The highest I've ever gone was 114 digits on 10212 (I'll probably venture a bit further with 130396), and that was before the invention of aliqueit.

As you will probably have seen in the "Feature request: warn on owerwrite" thread in the Msieve forum, I had something overwrite my test.dat file when running "aliqueit 130396 -e" during a GNFS run. This had never wiped out my file before, but this time I was resuming at the "msieve -nc2" stage. I have a strong suspicion that this is a bug in factMsieve.pl not aliqueit, but just in case it is aliqueit, I've posted it here.

[QUOTE=10metreh;175510]As you will probably have seen in the "Feature request: warn on owerwrite" thread in the Msieve forum, I had something overwrite my test.dat file when running "aliqueit 130396 -e" during a GNFS run. This had never wiped out my file before, but this time I was resuming at the "msieve -nc2" stage. I have a strong suspicion that this is a bug in factMsieve.pl not aliqueit, but just in case it is aliqueit, I've posted it here.[/QUOTE]

Yeah, I too think this is probably in factMsieve.pl. Aliqueit never deletes the ggnfs files until after the job completes [b]and[/b] a factor is found.

msieve -np now may return deg-4 and 6 polys
 

Mikael,

here is a new challenge for you.
I see in the code that
[FONT=Arial Narrow] bool convert_poly( string infile, string poly_name )[/FONT]
expects msieve to return a 5-degree poly. Well, this is not true anymore, starting from msieve 1.42. Please scan the msieve.fb even if it doesn't have A5 (or if it has A6). As it is now, if A5 is absent, aliqueit thinks that msieve failed and erases the resulting poly.

[SIZE=1]So I found to my amazement that the process was stuck in pol51opt...[/SIZE]
[SIZE=1]I'll set the path to the old msieve 1.41 for now. Too lazy to fix .cc cde.[/SIZE]
[SIZE=1][/SIZE]
--Serge

[QUOTE=Batalov;176178]Mikael,

here is a new challenge for you.
I see in the code that
[FONT=Arial Narrow] bool convert_poly( string infile, string poly_name )[/FONT]
expects msieve to return a 5-degree poly. Well, this is not true anymore, starting from msieve 1.42. Please scan the msieve.fb even if it doesn't have A5 (or if it has A6). As it is now, if A5 is absent, aliqueit thinks that msieve failed and erases the resulting poly.

[SIZE=1]So I found to my amazement that the process was stuck in pol51opt...[/SIZE]
[SIZE=1]I'll set the path to the old msieve 1.41 for now. Too lazy to fix .cc cde.[/SIZE]
[SIZE=1][/SIZE]
--Serge[/QUOTE]

I've actually already fixed this, so it'll be in the next release (hoping to get it out before too long). Up to sextics should be ok then.

[QUOTE=10metreh;176238]I used msieve 1.42 to search for a poly and put a msieve.fb and a test.poly in my newly created ggnfs directory for a C95 that I am doing, but something (either aliqueit or factMsieve.pl) deleted test.poly. What is happening?[/QUOTE]

Have a look at the post above yours. :smile: I'll try to get a new version out soonish as no doubt a lot of people will be running msieve v1.42 shortly.

[quote=mklasson;176239]Have a look at the post above yours. :smile: I'll try to get a new version out soonish as no doubt a lot of people will be running msieve v1.42 shortly.[/quote]

Sorry, I didn't read the post fully, so I've deleted my post.

Alright then, have a stab at [url=http://mklasson.com/aliquot.php]v1.08[/url].

+ "-q" quits aliqueit after factoring the first number.
+ detects bad factors (e.g. from recent yafu problem).
+ "-s <start>" submits the given sequence's elf file to Syd's DB, starting with iteration <start>. You'll need wget to use this.
+ timestamps in log file. Thanks kar_bon.
+ handles up to sextic polynomials output by msieve.
* only prime factors are considered "neat". Thanks 10metreh.

I was hoping to get some other good stuff included in this release, but time is a harsh mistress. Hopefully next time! This should at least take care of the swanky new polys from msieve 1.42.