Achievements

[danaj]

I just found a gap with merit 34.475548, which I believe is the 6th largest merit to date.

[robert44444uk]

Back from my holidays and I have discovered the largest gap with merit >30. Its size is in the range 157,000 to 158,000 so a large improvement over the previous record of 120,664 and almost double the size of the third largest.

[robert44444uk]

I have found another large top 20 #1, sadly it knocks my 214260 record very much into 2nd place for the class of merit >26

379304 C?C RobSmith 2015 26.31 6261 79*14593#/46410-129600

I have made a separate posting to Dr. Nicely this morning.

[henryzz]

Found my first 30 today.
30.048243 10628 PRP154 = 11069*383#/1793490-5400 + prev 24.483594
Need to get around to submitting soon.
I am running with a slightly old merits.txt. Do I need to check all my gaps before submitting to see if they have been beaten?

[danaj]

I sent mine in 2 days ago. 3073 gaps. Largest ones:

371916 14.121453 13*26513#/30-141418
381750 14.517393 11*26479#/30-324724
397530 15.105782 11*26497#/30-220046
402984 14.234797 11*28547#/30-136282
461692 16.296756 11*28559#/30-247134

Largest merits:

52498 30.563313 134327*1753#/2310-18652
32100 30.450170 4636459*1087#/2310-24952
31970 30.252503 5622691*1087#/210-17294
31670 29.983506 3327241*1087#/210-11090
10634 29.858884 115478999*367#/30-4542

Since then:

646 gaps, largest merits: 29.68, 28.27, 27.90, 27.48, 27.45.

For both sets:

- First gap entry: 6 of the smallest 20, including the first blank 71974. 72830 is now the smallest gap with no entry.
- 3 of smallest 20 with merit under 10 are now over.
- 8 of smallest 20 with merit under 15 are now over.
- 4 of smallest 20 with merit under 20 are now over.

With the AWS experiment running (single instance running a fixed divisor, see other gaps thread) the percent of gaps that are beating my own record is going up. The results started out slow but once it hit the 60k or so range it really took off. I'm not sure how much longer I'll let it run. Hard to shut it off when it keeps pushing out a new gap every few minutes.

Rob, another couple weeks and you will be in #2 for overall gaps. Until Hans does his next submission :).

[mart_r]

Originally Posted by mart_r

Then something must be wrong with my computation... thanks for the doublecheck!
I was reluctant to submit my results because I doubted the gap was way >4M, according to my PC.
What bounding primes do you have?

I was already half asleep when I posted this and completely failed to notice danaj's misinterpretation of the WolframAlpha formula primorial(x) where WA computes the product of the first x primes rather than the product of primes up to x...

So, to be precise: is there someone that may check if there is any prime between 230077#/2229464046810-3131794 and 230077#/2229464046810+1548362 ?

[robert44444uk]

This thread is to capture prime gap results that demonstrate a high degree of achievement.

This might include:

• First instance gaps
• Record merit
• Record size gap
• Record merits at the size of gap

[danaj]

Just found: 34.605960 12176 PRP153 = 577924573*359#/30-7508

Fifth largest currently known merit. Still hoping for a 35+ someday.


Also: 32.250212 109776 PRP1479 = 98543*3491#/7410-57272

Nice large find, though not breaking any particular records.


On another note, when submitting my results I also run a program that runs an ECPP proof on both endpoints of gaps. I combine my new results and all the current ones from Dr. Nicely's site, and run them in digit size order. It's been currently running to about 1360 digits. This is why we get "C?C" instead of "C?P" on the web pages for most gaps up to ~60k.

[robert44444uk]

Wow, such a huge find. Jealousy here.

Yesterday I submitted my first few days of my primorial offset results- still only the one 30, but I am only working on 4 cores.

I should really do the C?C thing - I forget sometimes that this is important.

[danaj]

I do have an advantage running on something like 30 cores when I add up all the computers. My searches are scattered all over, similar to Antonio's from what I see. Some 4k, some 25k, some 80k, some 200k.

For proofs, I run everything in the latest allgaps.dat file plus my new data, so there isn't a great need to run it unless you particularly want to or it is very large. I restart every day or two so it runs my newly found gaps up to where it left off, and of course start it every time there is new data on Dr. Nicely's page. I actually don't check the existing C?C status as it's rarely marked. I have a little database that stores the test results so it doesn't redo ones that have already passed.

[Antonio]

Originally Posted by danaj

My searches are scattered all over, similar to Antonio's from what I see.

My searches are not as scattered as you may think.
I have 3 cores, each running a different divisor but with the same range of multiplier, they are looking for starting primes in the range 1200 to 3260 digits, these are looking for the missing gaps below 100k. The other core is looking for large merit for starting primes in the range 200 to 399 digits (was 18 to 200 digits until today).
All cores are running a modified version of your code to allow me to specify the digit range to search over.