mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Conjectures 'R Us (https://www.mersenneforum.org/forumdisplay.php?f=81)
-   -   Algebraic factors in sieve files (https://www.mersenneforum.org/showthread.php?t=21916)

pepi37 2017-01-09 21:06

Algebraic factors in sieve files
 
1 Attachment(s)
I was little bored so using Batalov script I found many algebraic factors that can be remove before starting LLR

LIST

k8 -s86
k248832 -s145
k8 -s182
S205
k4 -s230
k27 -s252
k64 -s259
k27 -s328
k4 -s335
k4 -s875
k8 -s911
k8 -s930
k8 -s953
k8 -s993

I targeted "big ones" but also look at few "small" ranges
All factors are in zip file attached to this thread
And as you can see this time all factors are from S bases.

I hope Batalov is not angry that I used his script

Batalov 2017-01-10 18:40

It would be interesting if it was not mentioned quite a few times already.

What is indeed interesting is that every time this is mentioned CRUS answers: "there is no problem! srsieve removes these, you just don't know it!" Well, either it doesn't or the [URL="http://www.noprimeleftbehind.net/crus/Sierp-conjecture-reserves.htm"]hosted sieved files[/URL] were done incorrectly.

Checking sieve-sierp-base252-500K-1M.txt ... yep, the n%3==0 are still there (243 of them).
Checking sieve-sierp-base328-900K-1M.txt ... yeah. The same.
Checking sieve-sierp-base875-600K-1M.txt ... 1025 useless candidates are still there.
...and so on.

P.S. Why would I "be angry" at the clear signs that someone is steadily learning something. It is almost always a good thing. :rolleyes:

rogue 2017-01-10 18:58

Serge, I'm sure there are algebraic factors the srsieve is not finding. They come in many forms and the program is only searching for one form. If you have some code idea to implement into srsieve, please share with me.

MisterBitcoin 2017-01-10 19:42

I have a suggestion.
Maybe add the option to delete these factors by using Batalov´s script after sieving is done. It´s one more sieving step, but it´s worth it.

[CODE]Sieving->Search for Algebraic factors->Delete all factors using srfile[/CODE]

This might be easier before somehow is going to build up an new srsieve.

pepi37 2017-01-10 20:27

[QUOTE=MisterBitcoin;450764]I have a suggestion.
Maybe add the option to delete these factors by using Batalov´s script after sieving is done. It´s one more sieving step, but it´s worth it.

[CODE]Sieving->Search for Algebraic factors->Delete all factors using srfile[/CODE]This might be easier before somehow is going to build up an new srsieve.[/QUOTE]

Delete it before even start sieving :)
If you remember, when I ask you to do two of my sequence, I prepare data for you, and remove it before I send it to you :)
And it is not irrelevant when you remove it: just remove it before send here :)

pepi37 2017-01-10 20:29

[QUOTE=Batalov;450760]

P.S. Why would I "be angry" at the clear signs that someone is steadily learning something. It is almost always a good thing. :rolleyes:[/QUOTE]

I am honored with this words :smile:

gd_barnes 2017-01-11 05:50

It's possible that some of the sieve files were posted to the pages before more recent versions of srsieve.

pepi37 2017-01-11 06:24

[QUOTE=gd_barnes;450791]It's possible that some of the sieve files were posted to the pages before more recent versions of srsieve.[/QUOTE

Even latest srsieve doesnot remove algebraic factors. Remove just small part but doesnot remove all.

pepi37 2017-01-11 10:08

It is easy to test: using all latest version of srxsieve if you make 4*20^n+1 from n=1 to 1 M you will get 57250, and after using Batalov script it finds 5087 algebraic factors in that sieve file. So srsieve doesnot remove them, and also sr1sieve or sr2sieve dont remove them

rogue 2017-01-11 13:23

As I stated before algebraic factors come in many forms. srsieve only removes one of those forms. If someone can provide to me the other forms then I can modify srsieve to eliminate them.

MisterBitcoin 2017-01-11 19:10

Is someone able to post the script in this thread. I searched in the forums for it, but I didn´t found it. :/
I´ve some spare time atm.


EDIT:
Lucky me. [URL]http://www.mersenneforum.org/showpost.php?p=208827&postcount=105[/URL]

EDIT2:
Not so lucky me, seems to be somethink different...


All times are UTC. The time now is 00:10.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.