[QUOTE=Antonio;460311]I....
with parameters:
-gap 1250 -delta 200 -sb 24 -bs 16 -mem 12 -t 4[/QUOTE]

I've updated my parameters to yours and got nice speed up from 8e9 n/s to 12e9 n/s on only three cores.

[QUOTE=pinhodecarlos;460354]I've updated my parameters to yours and got nice speed up from 8e9 n/s to 12e9 n/s on only three cores.[/QUOTE]
Glad to be of service - that's quite an improvement.

4148-4160e15 complete:
[CODE]4148-4149e15 105 kgaps, largest 1218 @ 4148499167567383901
4149-4150e15 89 kgaps, largest 1192 @ 4149225469690253989
4150-4151e15 81 kgaps, largest 1158 @ 4150394748491183369
4151-4152e15 95 kgaps, largest 1218 @ 4151313332950925429
4152-4153e15 83 kgaps, largest 1290 @ 4152046681574768953
4153-4154e15 89 kgaps, largest 1216 @ 4153951701516973591
4154-4155e15 101 kgaps, largest 1230 @ 4154425229925611567
4155-4156e15 83 kgaps, largest 1176 @ 4155730081231180301
4156-4157e15 105 kgaps, largest 1266 @ 4156949040055701041
4157-4158e15 76 kgaps, largest 1158 @ 4157420831749175101
4158-4159e15 100 kgaps, largest 1140 @ 4158695464181782219
4159-4160e15 81 kgaps, largest 1210 @ 4159212921946842871[/CODE]

Reserving 4160-4180e15.

Some results while I was travelling:

[CODE]
1170 4045630509511133827
1128 4046481288649180253
1128 4046849195504965799
1260 4047841407534753811
1188 4048490181444860051
1148 4049021357890961831
1172 4050220172716052627
1274 4051526038603182527
1260 4052059085828160227
1292 4053867352647841601
1254 4054194003778353503
1174 4055488197330854047

[/CODE]

Reserving 4060e15 to 4070e15

Today Antonio (thanks!) found a minor bug that has been in all posted codes. The good news is that it hasn't missed any solution, just printed some false prime gaps, where (p1,p2) contained a prime number. As I can see this bug was only in the case where delta is divisible by 30 (that is why we haven't found this bug earlier). See the attachment for the new code! Basically even if you won't use it, you need to check only the printed gaps: so p1,p2 are primes and there is no prime in the (p1,p2) interval.

There is no other change in the code (and the speed of the program should be the same).

ps. gap5.c is the direct modification of the already posted gap3_4.c code.

[QUOTE=R. Gerbicz;460511]Today Antonio (thanks!) found a minor bug that has been in all posted codes. The good news is that it hasn't missed any solution,[U] just printed some false prime gaps, where (p1,p2) contained a prime number.[/U] As I can see this bug was only in the case where delta is divisible by 30 (that is why we haven't found this bug earlier). See the attachment for the new code! Basically even if you won't use it, you need to check [/QUOTE]

Checking the results from post #3 I see that in 23 ranges in which the results are divisible by 30 (in 2 separate instances they are double high results –1140– )

The highest gap I found which fits the divisibility by 30 is

4203 4204 antonio completed no 1320 4203690608946500057

The smallest is

4306 4307 danaj completed no 1110 4306985468075926151

[QUOTE=rudy235;460522]Checking the results from post #3 I see that in 23 ranges in which the results are divisible by 30 (in 2 separate instances they are double high results –1140– )
[/QUOTE]

The problem was with 30|delta and not with 30|gap.
Btw it was a quick fix, essentially only 2 lines changed in gap5.c .

[QUOTE=R. Gerbicz;460525]The problem was with 30|delta and not with 30|gap.
Btw it was a quick fix, essentially only 2 lines changed in gap5.c .[/QUOTE]
Ok. My bad, as they say.

I would like to mention a new gap (first known occurrence –C?C–) of 1430 by Leif LeonHardy as reported to Dr. Nicely's web page [URL="http://trnicely.net/gaps/gaplist.html"]http://trnicely.net/gaps/gaplist.html[/URL]
[FONT="Arial Black"]1430 C?C LLnhardy 2017 33.28 19 4606937813294064947[/FONT]

[QUOTE=rudy235;460534]I would like to mention a new gap (first known occurrence –C?C–) of 1430 by Leif LeonHardy as reported to Dr. Nicely's web page [URL]http://trnicely.net/gaps/gaplist.html[/URL]
[FONT=Arial Black]1430 C?C LLnhardy 2017 33.28 19 4606937813294064947[/FONT][/QUOTE]

Note that it is not marked as a first occurrence, so Dr. Nicely doesn't seem to think this was a comprehensive search. The max CFC is still 3.937..e15.

Assuming this forum's effort will be accepted, it won't be long before his records are changed to CFCs. That's his 6th record in the range 4.000e15 to 4.647e15. It looks like this result is out of order from his last one two years ago, so either it got delayed or he isn't doing an exhaustive search. It also means there is a chance we will find a record before we get into that range.

[QUOTE=danaj;460542] It looks like this result is out of order from his last one two years ago, so either it got delayed or he isn't doing an exhaustive search. It also means there is a chance we will find a record before we get into that range.[/QUOTE]

Yes, I did notice that a) it was out of order –the other on 2015 was 4.646– and b) that two years (give or take) have passed since the last C?C in that region (4.0-5.0 e18). I believe some sort of communication with Dr. Nicely and/or Mr. Leonhardy would be mutually beneficial and would go a long way towards avoiding double efforts.