Riesel base 3 reservations/statuses/primes

[appeldorff]

Originally Posted by KEP

So can anyone tell me how to make the latest version of srsieve write a factor file?

If you run srsieve with -h it will tell you this:

-f Append reported factors to the file srfactors.txt

Sadly it's not possible to define the filename yourself. Worked for me at least

[KEP]

Originally Posted by appeldorff

If you run srsieve with -h it will tell you this:

-f Append reported factors to the file srfactors.txt

Sadly it's not possible to define the filename yourself. Worked for me at least

Well I'm sorry to tell you and Max, that using just the -f or the --factors didn't work either, however no mistakes were reported, so I have really no idea, why it doesn't write the factors found. Only reason I can come in mind is if p has to be above a given level before factors is appended, else I have as mentioned no idea why srsieve doesn't write the factors found to srfactors.txt, when using either the -f or --factors in the line. So anyone who knows if srsieve is like NewPGen, and doesn't append factors before p is at a certain level? Also it doesn't matter, I'm currently working on k=3677878 for n>1M to n<=1G, and at least there, I can merge the NewPGen sieve format files, once sr1sieve completes the range I've asked it to sieve

Last but not least, thanks for your ideas and help

Regards

KEP

[Mini-Geek]

Originally Posted by KEP

Only reason I can come in mind is if p has to be above a given level before factors is appended, else I have as mentioned no idea why srsieve doesn't write the factors found to srfactors.txt, when using either the -f or --factors in the line.

Also from srsieve -h:

-m --minimum X    Do not report factors smaller than X (default 100000).

[KEP]

Originally Posted by Mini-Geek

Also from srsieve -h:

-m --minimum X    Do not report factors smaller than X (default 100000).

OK thanks guys, the problem was that I was using the -m 4e9 code too. Now it has been replaced with the -q flag, and all is working fine now

Thanks for all your help :)

On a side note, I'm now reserving the 2355 candidates k<=700M currently remaining at n=25K, I'm going to take them from n=25K-100K. Now that the srfactors problem has been resolved, it will be no real big hassel to reach optimal sievedepth.

Regards

KEP

[gd_barnes]

It would be good if someone did k=580M-600M to n=25K before starting on k=200M-700M for n=25K-100K.

...just a thought anyway.

Kenneth,

I can only reserve you for k=200M-580M and 600M-700M for n=25K-100K right now until someone completes k=580M-600M up to n=25K.

How many candidates total are there in the n=1M-1G range for k=3677878? That's a lot of sieving! Just to confirm: You are using sr1sieve, correct? It's way faster than anything else for one k.


Gary

[KEP]

Originally Posted by gd_barnes

It would be good if someone did k=580M-600M to n=25K before starting on k=200M-700M for n=25K-100K.

...just a thought anyway.

Kenneth,

I can only reserve you for k=200M-580M and 600M-700M for n=25K-100K right now until someone completes k=580M-600M up to n=25K.

How many candidates total are there in the n=1M-1G range for k=3677878? That's a lot of sieving! Just to confirm: You are using sr1sieve, correct? It's way faster than anything else for one k.


Gary

Regarding the 2355 k's there is at p=10G, ~4.4M candidates remaining, and optimal sievedepth for n<=50K is at around or less than p=25G.

Regarding the sieving for k=3677878 from n=1M to n=1G, is at p=1T, ~20M candidates. I'll stay focused on the 2355 k's and once the complete to n=100K, I'll focus on completing the remaining 110K n's for k=3677878 for n<=1M, before continuing sieving.

Hope I got it all answered, else feel free to ask again. Will btw. start crunching the n<=50K range this upcoming friday, as soon as sieving completes to p=25G.

KEP!

Ps. I'll test the new script and take the 580-600M range to n=100K aswell. So you can have me reserved for the entire 200M-700M range towards n=100K. Btw, a n=100K test with k=~700M, takes 49.7 seconds on a Q6600

[KEP]

Forgot to say, that for k=3677878 for n>1M, I'm of course using sr1sieve

Regards

KEP

[Mini-Geek]

Originally Posted by KEP

Ps. I'll test the new script and take the 580-600M range to n=100K aswell. So you can have me reserved for the entire 200M-700M range towards n=100K.

Did you miss this part, Gary?

I'll take k=800M-805M to 25K.

[gd_barnes]

No. I haven't missed anything. I just haven't updated the pages in a few days. It's going to take a while.

[kar_bon]

k=700M-800M at n=10k: 1806 k's left. continue with sieving to n=25k.

[Mini-Geek]

Originally Posted by Mini-Geek

I'll take k=800M-805M to 25K.

Done. Primes, MOBs, and remaining k's emailed to Gary. 25 k's remaining.

800078932
800171464
800532448
800749816
800925284
800983478
801024038
801041998
801253748
801380464
801812696
802047572
802051688
802054384
802647262
802799518
802826098
802966594
803184164
803955168
803983118
804227206
804655372
804655736
804696218