How do we want to organize this?
 

OK, it looks like a bunch of us are starting to recreate the missing sequences in the 200-250k range, and we should probably get this organized before we get too much duplication of effort going on.

There are about 200 sequences that have a listing in Wolfgang's tables that we do not the data for. Some have a status as of 100 digits, but most do not. All the files that were available to us are posted on my space and some have been imported into Syd's DB.

I have started working on the low end of the range. mataje has started in the range above 230k. kar_bon is starting on the 270k range (Wieb Bosma has done this, but does not have any files posted.

How do we want to organize this? My "First holes" thread was an effort to get this rolling in an organized way, but it looks like we're getting enough people that we need a little more organization.....so what do you all think?

More thoughts
 

I'm finding that most sequences take 1-2 dats on my dastest system to get to 100 digits. My preference to keep things easy would be to take blocks of 10-15 sequences, run them until 100 digits (greater if warranted) and then posting a completion message.....

I would suggest something slightly different. Post all the OE sequences that need work (in 10k or 20k blocks). Let people reserve however many sequences they want, with the proviso that they must run it up to at least 100 digits.

[quote=schickel;176304]but it looks like we're getting enough people that we need a little more organization.....so what do you all think?[/quote]

The obvious way to get more organization is to appoint a new moderator. :smile:

i'm displaying all ranges 200000 < n < 250000 on the summary pages.

make a thread, so everyone can reserve a seq there, summarize these info in the first post (like the other reservation-thread) and delete the res-post (after few days). i will update the summary, too. same procedure for unreserving!

the best would be such reservations done in the Fatoring-Database with contributor-name output in the sequence page! the summary-page the same done by the database, but i think Syd got to do enough before implementing such workflow.

[QUOTE=axn;176308]I would suggest something slightly different. Post all the OE sequences that need work (in 10k or 20k blocks). Let people reserve however many sequences they want, with the proviso that they must run it up to at least 100 digits.[/QUOTE]Well, the thing is that the only range so far that needs recreating is 200-250k, so that's only 2 or 3 blocks of 10 or 20k. I'd also like to leave open the possibility of having some "starter work" left over so that someone just coming in to this isn't in the position of being told, "you'd like to help? Sure, just go factor this 110-digit number for us so we can see what happens next."

Wieb Bosma is actively working on the range from 250-400k. I emailed him to try and bring the forum to his attention (twice), but never got any reply. (Maybe somebody with academic contacts could bring the forum to his attention....) I would like to find out if he would be receptive to opening his research up to a larger audience. We could find plenty of work in that large a range to occupy more people than we have currently participating.

I think it misses a thread on info on how to get started on Aliquot Sequences. There's a lot of sticky threads but there's none explaining step by step how and which programs to run.
Is it possible to create one for lazy people like me?

Worker.exe program should be very visible and not in the middle of a huge thread.

Carlos

My suggestion is to get the aliqueit.exe program. Find the sticky thread on it. I posted a thread on ggnfs failing which contains a summary of what I had to do (or read full thread). Note before this I downloaded the ecm, yafu, msieve, ggnfs programs as suggested on the aliqueit web page. I copied the executables (including aliqueit) to c:/aliquot/bin (windows) and added this to my path.

Also you can visit a db that you type a number and it gives you the sequence:
[URL="http://factorization.ath.cx/search.php?se=1"]db location[/URL]

and read this overview on what it is about if you don't know aliquot seqs.
[URL="http://www.aliquot.de/aliquote.htm"]overview on sequences[/URL]

The summary (from problems with ggnfs thread):

I had to do one final thing and that was copy over the more recent msieve.exe into the ggnfs directory - the one that came with ggnfs was
old.

Once I did that it successfully factored using the general number field seive. [see whole thread if this is confusing]

So to summarize, in order to get ggnfs working I needed to:

1) Make sure I had a recent (past 5.2) version of perl
2) Change the aliqueit.ini file entry for ggnfs_cmd to point to ggnfs's factMSieve.
3) Update factMSieve.pl ggnfs_bin_path (top)
4) Make sure I had cat.exe.
5) Update factMSieve.pl sys_bin_path (further down) [for cat.exe]
6) For the 340 version of ggfns (not sure about 360), I needed to copy over a later version of mSeive into the ggnfs_bin_path directory.

Once I did that it successfully factored using the general number field seive.

------------------
More recently there is a way to submit sequences to the db automatically with wget. I think this is explained in the aliqueit.exe thread.
had to do one final thing and that was copy over the more recent msieve.exe into the ggnfs directory - the one that came with ggnfs was
old.

Agree it would be nice to have an intro sticky that was clearly marked as such.

[quote=Greebley;180425]Agree it would be nice to have an intro sticky that was clearly marked as such.[/quote]

OK, but I'm a bit busy ATM (and I encountered some spam), wait till later in the day and I'll add a "Getting Started" sticky.

Ok, cool.
Might be worth mentioning how long things take:
under C60 is less than a second, C80 is seconds, C100 is minutes, C120 2-3 days where C<num> is the cofactor size that needs factoring - something like that.
Also our current primary goals (the 210-250k project and 100k project).

[quote=Greebley;180465]Ok, cool.
Might be worth mentioning how long things take:
under C60 is less than a second, C80 is seconds, C100 is minutes, C120 2-3 days where C<num> is the cofactor size that needs factoring - something like that.
Also our current primary goals (the 210-250k project and 100k project).[/quote]

Remember that speeds are different for different processors. BTW, how many minutes is "minutes" for a C100? I'd think a C100 would only take under 1 hour if you used 4 cores simultaneously (or 2 cores simultaneously on a 64bit system (thanks henryzz)). Generally I take "minutes" to mean "a few minutes". If that's how long a C100 takes for you, how many nodes does your cluster have?