sr2sieve 1.9.3 "problems"

[pepi37]

If I run sr1sieve ( latest version) with command sr1sieve.exe -i xx.txt -pxxx -Pxxx -o xxF.txt in output txt file ( in header) I got Pxxx as number ( sieve end) and output file xxF is clear of all factors that sr1sieve found.

But if I run sr2sieve with command
sr2sieve -i sr_2.abcd -P300000000000 then when process finished I got
-factors.txt ( i need to remove it manually)
-when I remove factors in header of abcd file still exist sieve depth from sr1sieve, not new depth since process is over and new depth is much bigger.

So question is
1. what command line I need to use to get automatically "clear" abcd file ( with removed factors) when process is over?
2. why new sieve depth is not written in header of abcd file when process is over.

Thanks for reply

[rogue]

Originally Posted by pepi37

So question is
1. what command line I need to use to get automatically "clear" abcd file ( with removed factors) when process is over?
2. why new sieve depth is not written in header of abcd file when process is over.

sr2sieve does not update the input file at any time. Use srfile to remove candidates with factors from the input file. Since sr2sieve does not update the input file, you need to use -p and -P to sieve a range with it.

[pepi37]

Thanks for answer.
No I know that is nothing wrong with commands.
It is how sr2sieve works!

[rogue]

I could probably add a switch to sr2sieve so that it can write an updated input file. Most people would probably use it.

There are a couple of other pending changes I have for sr2sieve and sr1sieve, but I've been too busy to work on them.

[pepi37]

Originally Posted by rogue

I could probably add a switch to sr2sieve so that it can write an updated input file. Most people would probably use it.

There are a couple of other pending changes I have for sr2sieve and sr1sieve, but I've been too busy to work on them.

Yes updating input file when range is completed will be very usefull

[pepi37]

In readme.txt of sr2sieve says

Only factors larger than k and b will be found, so the sieve must be started and sieved until p >= max(k,b) with another program (NewPGen, Srsieve).

I made small sieve from n 3200000 to 3400000 and K was 139,167,271,353,773,797,and 901.

So tell me how you calculate initial sieve depth ( in my case) so after that I can start use sr2sieve.
I was make sieve for all those factors until 1500000000. Is that ok depth?

Thanks

[Batalov]

You should have sieved only to 3400000 with srsieve, then srfile -a, and then sr2sieve to ~1e10. srsieve is much slower than sr2sieve.

[pepi37]

I made sieve with sr1sieve for every K individually :)
So it is ok depth!
Thanks for reply!

[Batalov]

1500000000 << 1e10