mersenneforum.org  

Go Back   mersenneforum.org > Prime Search Projects > Prime Gap Searches

Reply
 
Thread Tools
Old 2019-11-15, 16:22   #111
Bobby Jacobs
 
Bobby Jacobs's Avatar
 
May 2018

193 Posts
Default

Quote:
Originally Posted by rudy235 View Post
That is very good! Congratulations.
As to Dr Nicely he usually doesn't take so long (it's been 81 days) to update the pages. I hope he is well.
Dr. Nicely died.
Bobby Jacobs is offline   Reply With Quote
Old 2019-11-15, 16:56   #112
mart_r
 
mart_r's Avatar
 
Dec 2008
you know...around...

26×32 Posts
Default

Quote:
Originally Posted by robert44444uk View Post
Unbelievably I just found my third "top 10" merit in just ten days. The new find is the new 5th largest merit

37.458 37668 PRP437 = x*y#/46410-z

It is a new member of the "no larger gaps discovered to date with larger merit" list, and displaces my own recent find:

37.08 36760
Way to go Robert!
Hope to see a 42-Merit gap from you soon
mart_r is offline   Reply With Quote
Old 2020-01-02, 10:01   #113
robert44444uk
 
robert44444uk's Avatar
 
Jun 2003
Oxford, UK

35258 Posts
Default

No 42 level gap, but this monster appeared at the tail end of 2019

35.51 merit 40920 gap 501 digits x*y#/46410-z smaller prime

Not a "top 10 merit" or a "best merit at" record, but pretty worthwhile!
robert44444uk is offline   Reply With Quote
Old 2020-01-07, 17:14   #114
robert44444uk
 
robert44444uk's Avatar
 
Jun 2003
Oxford, UK

1,877 Posts
Default

Quote:
Originally Posted by robert44444uk View Post
Unbelievably I just found my third "top 10" merit in just ten days. The new find is the new 5th largest merit

37.458 37668 PRP437 = x*y#/46410-z

It is a new member of the "no larger gaps discovered to date with larger merit" list, and displaces my own recent find:

37.08 36760

The top ten merit list looks like the following now:

Code:
1	8350	 41.938 	Gapcoin  	2017
2	15900	 39.620 	Jacobsen	2017
3	18306	 38.067 	Jacobsen	2017
4	8382	 37.824 	Jacobsen	2018
5	37668	 37.458 	Smith   	2019
6	36760	 37.085 	Smith   	2019
7	26054	 37.005 	Jacobsen	2017
8	7362	 36.929 	Gapcoin  	2017
9	10716	 36.858 	Jacobsen	2016
10	35952	 36.756 	Smith   	2019
Another top 10 merit gap for me

5. 44294 37.29 Smith 2020

It is a new member of the "no larger gaps discovered to date with larger merit" list. First big find of the new decade!!
robert44444uk is offline   Reply With Quote
Old 2020-02-03, 23:05   #115
SethTro
 
SethTro's Avatar
 
"Seth"
Apr 2019

163 Posts
Default

For some work on mpz_prevprime I plotted the highest merit of a prime gap by year discovered.

I started with
Code:
select year, max(merit) from gaps group by year;
Code:
-300|1.44
1877|10.2
1934|10.03
1957|12.46
1961|12.45
1967|16.54
1973|22.8
1980|21.27
1989|25.03
1991|25.02
1995|24.59
1996|27.44
1997|25.28
1998|26.55
1999|32.28
2000|28.6
2001|29.87
2002|31.07
2003|31.34
2004|30.85
2005|34.98
2006|33.77
2007|33.29
2008|32.7
2009|35.31
2010|31.42
2011|34.12
2012|35.42
2013|32.32
2014|34.94
2015|34.48
2016|36.86
2017|41.94
2018|37.82
2019|35.510486
I then plotted this with colab
https://colab.research.google.com/dr...hyN0AVT2DPmasA

The trendline is
Code:
y = 0.4073 * (year - 2000) + 29.96
Attached Thumbnails
Click image for larger version

Name:	Screenshot from 2020-02-03 12-50-19.png
Views:	66
Size:	16.8 KB
ID:	21717  
SethTro is offline   Reply With Quote
Old 2020-02-04, 09:58   #116
robert44444uk
 
robert44444uk's Avatar
 
Jun 2003
Oxford, UK

35258 Posts
Default

Quote:
Originally Posted by SethTro View Post
For some work on mpz_prevprime I plotted the highest merit of a prime gap by year discovered.

I started with
Code:
select year, max(merit) from gaps group by year;
Code:
-300|1.44
1877|10.2
1934|10.03
1957|12.46
1961|12.45
1967|16.54
1973|22.8
1980|21.27
1989|25.03
1991|25.02
1995|24.59
1996|27.44
1997|25.28
1998|26.55
1999|32.28
2000|28.6
2001|29.87
2002|31.07
2003|31.34
2004|30.85
2005|34.98
2006|33.77
2007|33.29
2008|32.7
2009|35.31
2010|31.42
2011|34.12
2012|35.42
2013|32.32
2014|34.94
2015|34.48
2016|36.86
2017|41.94
2018|37.82
2019|35.510486
I then plotted this with colab
https://colab.research.google.com/dr...hyN0AVT2DPmasA

The trendline is
Code:
y = 0.4073 * (year - 2000) + 29.96
37.458 was the highest in 2019.

I know of one gap over 38 in 2020! Well done Seth!
robert44444uk is offline   Reply With Quote
Old 2020-02-04, 11:32   #117
SethTro
 
SethTro's Avatar
 
"Seth"
Apr 2019

163 Posts
Default

I got incredibly lucky testing out a new heuristic and find a prime two of merit 38! It's the 4th largest prime gap by merit found!

35308 38.05 x*y#/z-a

I'm working to record it to the githib list.
SethTro is offline   Reply With Quote
Old 2020-09-09, 07:31   #118
SethTro
 
SethTro's Avatar
 
"Seth"
Apr 2019

163 Posts
Default Death to missing gaps

Yesterday I started a personal project to fill the 25 gaps with no known start prime < 125,000.

I created a new list on the primegap-list-project page that lists the smallest gap without starting prime (without a record? without a known prime pair?).

And I have derived some rather clever formula (to large to discuss here) that I believe make it feasible to find a pair of primes for any gap.

After 24 hours I had my first pseudo success, when I found this gap with no previous record. Hopefully I can continue finding ~1 missing gap a day

Code:
130898 64003 * 12007#/1110 -47954 to +82944
SethTro is offline   Reply With Quote
Old 2020-09-11, 06:39   #119
SethTro
 
SethTro's Avatar
 
"Seth"
Apr 2019

2438 Posts
Default

I validated and slightly improved my approach.

I've found a 2nd missing gap.

Code:
127622  56653*9001#/1110 -69548 to +58074
I'll chug away on these for a couple of days and see if I can find all 25 or a good fraction of them.
SethTro is offline   Reply With Quote
Old 2020-09-12, 08:56   #120
mart_r
 
mart_r's Avatar
 
Dec 2008
you know...around...

26·32 Posts
Default

Can I submit megagaps with primes in the 39000+ digits range via the cloudygo window or is it advisable to rather post them here?
mart_r is offline   Reply With Quote
Old 2020-09-12, 13:38   #121
SethTro
 
SethTro's Avatar
 
"Seth"
Apr 2019

A316 Posts
Default

Quote:
Originally Posted by mart_r View Post
Can I submit megagaps with primes in the 39000+ digits range via the cloudygo window or is it advisable to rather post them here?
Currently I have primegaps.cloudygo.com double checking the submitted records. on records with 30,000+ digits this would take 1hr+ each. I'd prefer not to tie up that computer 24x7 verifying records.

A simple compromise might be to check endpoints but skip full verification for known contributors on gaps with merit less than 30 (on the occasional larger records I'm happy to run the full verification). This seems reasonable because the alternative is that they probably don't get verified at all (and are just copy pasted into the allgaps.sql file)

It would be nice to get a 2nd person (Robert, Jansen, DanaJ) to confirm this sounds reasonable.
SethTro is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
News gd_barnes Conjectures 'R Us 293 2020-09-13 07:42
News gd_barnes No Prime Left Behind 250 2020-06-29 13:23
P!=NP in the news willmore Computer Science & Computational Number Theory 48 2010-09-19 08:30
The news giveth, the news taketh away... NBtarheel_33 Hardware 17 2009-05-04 15:52
Some news about Home Prime ? MoZ Factoring 6 2006-02-28 12:02

All times are UTC. The time now is 07:40.

Fri Sep 25 07:40:23 UTC 2020 up 15 days, 4:51, 0 users, load averages: 1.02, 1.21, 1.39

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

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.