The good news is we now have the largest factor on this page:

[url]http://escatter11.fullerton.edu/nfs/crunching.php[/url]

:razz:

And I'd say you are spoiled by access to 24-CPU computers.

"Let's agree to disagree."
[ATTACH]9067[/ATTACH]

I'll rephrase - maybe some other people won't have anything to do with their computer for three weeks, but I definitely will. I'll find a Fermat factor or a megabit prime or something, instead of running a 17M matrix on a very small number. :rolleyes: Running a 10M matrix on a very small number is sort of ok. Running a 17M matrix on a c261 is fine, too.

p55 * p187 is lackluster for a number which had received t55 ECM work... not an ECM miss, but bad luck :smile:

If it wasn't for bad luck we'd have no luck at all!

:mike:

Since the download takes 8.5 hours and the final relations are trickling in very slowly, I've started to grab L2605B as of now and will start the processing going tonight. Probably it will be the first Lucas factor of 2013 rather than the last of 2012.

For the slightly longer term, who's doing 3617523089023m19 ?

lasievee
 

[QUOTE=fivemack;323011]

For the slightly longer term, who's doing 3617523089023m19 ?[/QUOTE]

Usually for lasievee post-processing or Greg or Mike but this one is still free.

Carlos

[QUOTE=Speedy51;322980]I'll claim 997_67_minus1 please[/QUOTE]
[code] prp61 factor: 1161453536782841917108642302077561510706740359135036934282243
prp95 factor: 53522412511211470387393560525278880705070038175781719883349525261442537403808699242042872805549[/code]

I totally agree with Sergey on the sieve region matter. NFS@Home users only care about the points not if the integer is or not undersieve. I think most of the integers should be more sieved, we need to use the power we have available.
I also don't like to run a huge matrix when resources are available to reduce it by a few %. Sometimes one more week of sieve makes the difference on the post-processing phase.

Carlos

Carlos saw just right what I left between the lines, but didn't write explicitely. The boincified client machines feel no pain, they convert their electricity and idle computer time into warmth for the home and a warm feeling for their owner of having more points than their neighbor ("in Project Mayhem, no one has a name"). The matrix volunteers are a bit less faceless: they type some keystrokes, they tinker, and frankly, sometimes they have not-so-shiny machines; they deal with restarts; a machine that could choke on a 17M matrix, or a machine on which a 17M matrix will only run without largeblocks (i.e. slower still). Running a matrix is more demanding on a machine than sieving; I won't generalize, but on my machines while the matrix is in swing, there are not so many things that can be done; typing is fine; but even while typing you can get a momentary OS freeze and then suddenly the last letter is repeated twenty times. It is preferable to have [I]that[/I] limited to a week instead of three weeks.

What I am trying to say is matrix hadlers deserve a bit more of a break than sievers. And I am not saying they should get 240M or 260M unique relations. But not just 170M. 190M-200M u.r., maybe?

Hey, wait, and before I forgot, Lionel had implemented server side remdups before the sun had set on his server. Wouldn't it be a good thing to restore? It will save on traffic and disk, I think. And it will give a good indication if a project is ready for transfer or not. (Some projects may look good with, say, 225M raw, but after transfer and remdups or msieve filtering when the count shrinks to 167M, the matrix handler will with a good probability get [I]pschik[/I]* for cycles and will write back to managers anyway? Server side remdups would obviate that scenario.)

In the meantime, that number is plodding on (from 196M u.r.). ETA is now nominally 100hrs but I will have to take some breaks (kills/restarts for the daytime), so let's say, a week. (I also will be in Hawaii when it will be close to finishing but hopefully I'll find a connection to monitor the ending.)

__________
[I][SIZE=1]*There was this [/SIZE][/I][URL="http://www.skazki.com/skazki/russkie_narodnye_skazki/b/barin_kuznets/"][I][SIZE=1]Russian tale[/SIZE][/I][/URL][I][SIZE=1]. There was this merchant, and he envied the blacksmith - so little work and so much pay. So, he decided to become a blacksmith himself. Bought all the equipment, the whole deal. Started the furnace, all looked well, and here comes the first customer. The customer ordered an axe. So the new blacksmith forged and forged but the piece was getting smaller and smaller (it's a long tale... he adjusts his expectations gradually to make a scythe, then, - a hammer, maybe... nah...), finally he thinks, well there's only enough left for a nail. So he finishes the nail, puts it in water and the only things he gets is ПШИК (the sound of boiling water).[/SIZE][/I]

This tale is not so different from filtering for a matrix, is it? ;-)

Oh, man, look at the time. ...and who wrote all that fluff?!!
(I must have been sleeping while typing. It was all Tyler Durden!)

[QUOTE=Batalov;322999]... I'll find a Fermat factor or a megabit prime or something, instead of running a 17M matrix on a very small number. :rolleyes: .[/QUOTE]
I [URL="http://primes.utm.edu/primes/page.php?id=110507"]wasn't[/URL] kidding, btw...

[QUOTE]Running a matrix is more demanding on a machine than sieving; I won't generalize, but on my machines while the matrix is in swing, there are not so many things that can be done; typing is fine; but even while typing you can get a momentary OS freeze and then suddenly the last letter is repeated twenty times. It is preferable to have [I]that[/I] limited to a week instead of three weeks.[/QUOTE]We are unable to tell when we are doing work. It affects our system in no way. But even the slightest lag would end our involvement, since we have but one computer now that is used for everything. Maybe we have been lucky so far?