Sieving for CRUS

[gd_barnes]

S465 sieving is complete for n=25K-100K.

Reserving S810 for sieving n=50K-100K.

[gd_barnes]

Originally Posted by MisterBitcoin

Good point!
I use -x an every base with more than 100 k´s remain, because it´s taking some time to build them.

This is a mistake as long as you have reasonable memory. You can build a Legendre table on any base and you should at least try it out before deciding not to do it. I've done it with up to 1000 k's remaining. Even if it takes 30 minutes to an hour, sr2sieve is so much faster than srsieve or sr2sieve with the -x switch. My general rule is that the table has to take less than 2 hours to build and be < 3 GB in size. A month or so ago I did one with over 600 k's remaining. The file took an hour to build and was about 2 GB in size but in the long run it saved a tremendous amount of sieving time. Running it on a machine with 8 GB of RAM I was able to run 3 instances of it.

On all of the n=25K-100K sieve files that you have seen me do recently, all with 180-350 k's remaining, I have used sr2sieve after creating the Legendre table. The largest table was ~1.5 GB but most of them were 50-500 MB. Doing that I can easily sieve to P=1T on a nearly 10 year old quad in ~5-6 days.

[gd_barnes]

Originally Posted by MisterBitcoin

@Gary
Can you zip the remain k´s for all bases tested up to n=10K with less than 1,000 k´s?
Those bases should be R606, R877, R996, S606, S807 and S1005.
I´ll sieve them up to n=25K.

I´m also working on the sieve files done my yoyo@home, maybe I´ll upload them in the middle of the night or tomorrow.

How much longer for the sieve file on these bases? It's been almost 2 months. I've already tested S606 to n=25K and will soon be ready to test R606 for n=10K-25K.

What sieve depth are you at for R606? If you have passed P=250G that should be enough and you can post the file here.

See the last post for why using the -x switch is likely a mistake on many of these bases and is slowing you way down.

[MisterBitcoin]

Originally Posted by gd_barnes

How much longer for the sieve file on these bases? It's been almost 2 months. I've already tested S606 to n=25K and will soon be ready to test R606 for n=10K-25K.

What sieve depth are you at for R606? If you have passed P=250G that should be enough and you can post the file here.

See the last post for why using the -x switch is likely a mistake on many of these bases and is slowing you way down.

I was focusing on R411, which is now free to go.
Factors up to 2.43T attached.
I´m now going to R606, just give me a couple of days.

[gd_barnes]

Originally Posted by MisterBitcoin

I was focusing on R411, which is now free to go.
Factors up to 2.43T attached.
I´m now going to R606, just give me a couple of days.

Great thanks. My machines will need work in ~3-4 days so the timing will work out well.

[gd_barnes]

S810 sieving is complete for n=50K-100K.

Reserving S576 for sieving n=25K-100K.

S576 has 219 k's remaining but 34 of those k's are effectively being searched by S24 and are already at n>100K. So I will be sieving the other 185 k's.

[gd_barnes]

Reserving S646 for sieving n=25K-100K.

[gd_barnes]

S576 sieving is complete for n=25K-100K. This includes the 185 k's not searched by base 24.

[gd_barnes]

Reserving R936 and S936 for sieving n=25K-100K.

Ian is still working on S936 to n=25K and will be done in ~7-10 days. At that point I will begin sieving both bases together.

[gd_barnes]

Pepi has sent me 8 files sieved by Yoyo. Information as follows:

Riesel bases: 281, 338, 370, 380, 410, 447, 452, and 516
All 2, 3, and 4 k's remaining.
n=100K-300K
P=50T

They are now posted on the reservations page.

[pepi37]

Originally Posted by gd_barnes

Pepi has sent me 8 files sieved by Yoyo. Information as follows:

Riesel bases: 281, 338, 370, 380, 410, 447, 452, and 516
All 2, 3, and 4 k's remaining.
n=100K-300K
P=50T

They are now posted on the reservations page.

And I will send rest ( four of them) they are finished today, all WU is returned.
So expect new mail tomorrow :)