[QUOTE=paulunderwood;532409]It seems to have vanished.[/QUOTE]

Because, an N-1 test has been run since your post, which will remove any traces of the ECPP test.

Well it wasn't done by me, since when I opened the link it was already labeled P.
If the purge was done in November 2018, then it was not done since Paul's post yesterday.

No twin primes for M9689. for k <= 607638

I will rewrite my code to reduce the read/writes for the kCounter by some multiple to get back at multi-threading when I get a chance.:smile:

Well, my rewrite is done so I can get back to my hobby.

[B]3901012.6.M9689 +/- 1[/B] are minimal Twin-Primes for [B]M9689 [/B]with 2925 dd each.:smile:
Found using [B]Pari-GP[/B] & [B]PFGW[/B].


ETA:
[url]http://factordb.com/index.php?query=3901012*6*%282%5E9689-1%29%2B1+[/url]
[url]http://factordb.com/index.php?query=3901012*6*%282%5E9689-1%29-1+[/url]

[B]718187.6.M9941 +/- 1[/B] are minimal Twin-Primes for [B]M9941[/B] with 3000 dd each.:smile:
Found using Pari-GP & PFGW.

ETA:
[url]http://factordb.com/index.php?query=718187*6*%282%5E9941-1%29%2B1+[/url]
[url]http://factordb.com/index.php?query=718187*6*%282%5E9941-1%29-1+[/url]

[B]3527063.6.M11213 +/- 1 [/B]are minimal Twin-Primes for [B]M11213[/B] with 3383 dd each.:smile:
Found using [B]Pari-GP[/B] & [B]PFGW[/B].

ETA:
[url]http://factordb.com/index.php?query=3527063*6*%282%5E11213-1%29%2B1+[/url]
[url]http://factordb.com/index.php?query=3527063*6*%282%5E11213-1%29-1+[/url]

[QUOTE=Batalov;532333]1. Just write a simple sieve using libgmp. ...or use Polysieve? it will be way faster.
[/QUOTE]
a1call, just in case you weren't listening, I will repeat. Use Polysieve. It makes this whole search entirely trivial and very fast.

Illustration:
Warming up: ... [URL="http://factordb.com/index.php?id=1100000001415037066"]120729852*(2^23209-1)-1[/URL] and [URL="http://factordb.com/index.php?id=1100000001415037017"]120729852*(2^23209-1)+1[/URL]
... 292133940*(2^23209-1)±1
... 771902880*(2^23209-1)±1
... 904094700*(2^23209-1)±1
This takes just a couple hours to find.

Next, k*(2^44497-1)+-1 ...

120M would definitely take me more than a day on 16 threads even with my fancy algo. Assuming you got that on 1 core that would be astonishing.
Thank you I will definitely research polySieve with a higher priority now.
Thanks again.
ETA: BTW if anyone would like to pursue any of the remaining Mersenne prime minimal twins, please be my guest.
It works probably help avoid wasting resources if anyone doing so announces it here 1st.
I am still at about k=10M on the next Mersenne prime on the queue.A solution for 60k dd primes of any type would be a world top 10.

I found a solution for M19937.
I will post here when I get a chance.

[B]9522163.6.M19937 +/- 1[/B] are minimal Twin-Primes for[B] M19937[/B] with 6010 dd each.:smile:
Found using [B]Pari-GP[/B] & [B]PFGW[/B].


[B]10073020.6.M19937 +/- 1[/B] are secondary minimal Twin-Primes for[B] M19937[/B] with 6010 dd each.:smile:
Found using [B]Pari-GP[/B] & [B]PFGW[/B].


ETA:
[URL="http://factordb.com/index.php?query=9522163*6*%282%5E19937-1%29%2B1+"]9522163*6*(2^19937-1)+1[/URL]
[URL="http://factordb.com/index.php?query=9522163*6*%282%5E19937-1%29-1+"]9522163*6*(2^19937-1)-1[/URL]
[URL="http://factordb.com/index.php?query=10073020*6*%282%5E19937-1%29%2B1+"]10073020*6*(2^19937-1)+1[/URL]
[URL="http://factordb.com/index.php?query=10073020*6*%282%5E19937-1%29-1+"]10073020*6*(2^19937-1)-1[/URL]

I will next run [B]M21701[/B] with the setup that I have as-is, until I get a chance to improve on it.
I also have a few cores running for [B]M216091[/B] as an effort in wishful thinking.
:smile:

Twin based on M44497 is relatively quite hard to find. The first one is ~3x higher than ME of its location.

Anyway, there, 2024053740*(2^44497-1)±1 are prime