[QUOTE=gd_barnes;410246]Please review this submission. There are multiple problems. It could be that srbsieve was run incorrectly or that the primes or k's remaining files were posted incorrectly in some fashion. The # of k's remaining is off by several hundred, there are hundreds of duplicate primes in the prime file, and the conjecture is included in the k's remaining file. (The Maxk should always be set to the conjecture minus one).

This base is in the recommended bases thread. Please take a look in that thread. I had already searched it for n<=2500 (using the new bases script) and there were 831 k's remaining. In that thread is a file with the 831 k's remaining at n=2500.

Some analysis:

1. In removing all of your duplicate primes, I found that your primes for n<=2500 exaclty match mine. So...there is NO problem at n=2500.

2. All k's remaining are for k>19500. (??)

3. You have only posted 315 primes for n=2500-10K. 831 minus 315 leaves 516 k's remaining at n=10K (vs. 257).

Please balance the primes and k's remaining before reposting. Thank you.[/QUOTE]

I don't know why there would be duplicates. I will have to take a look.

I think I know why might happened with the remaining k. My computer was restarted in the middle of processing the range and I tried to piece together the results. I most likely edited the wrong pl_remain.txt file.

[QUOTE=rogue;410262]I don't know why there would be duplicates. I will have to take a look.

I think I know why might happened with the remaining k. My computer was restarted in the middle of processing the range and I tried to piece together the results. I most likely edited the wrong pl_remain.txt file.[/QUOTE]

In reviewing the duplicates, all are for n<=5. You may have copy-and-pasted the same primes twice. When I removed them, everything for n<=2500 matched up with my previous run.

The big issue seems to be the k's remaining file.

I think it would help both of us if you would sort the entire prime file by either n-value or k-value. That's what I get from most of our searchers on new bases. If you do that, I would get a sorted file, which would save me time since I have to do it anyway, and you could easily see duplicates, which would help catch problems like this before posting.

I have an Excel spreadsheet that very quickly parses out the k and n-value for each prime. It makes sorting numerically (instead of alphanumerically) either by k-value or n-value quick and easy. Let me know if you want me to post it.

[QUOTE=gd_barnes;410292]In reviewing the duplicates, all are for n<=5. You may have copy-and-pasted the same primes twice. When I removed them, everything for n<=2500 matched up with my previous run.

The big issue seems to be the k's remaining file.

I think it would help both of us if you would sort the entire prime file by either n-value or k-value. That's what I get from most of our searchers on new bases. If you do that, I would get a sorted file, which would save me time since I have to do it anyway, and you could easily see duplicates, which would help catch problems like this before posting.

I have an Excel spreadsheet that very quickly parses out the k and n-value for each prime. It makes sorting numerically (instead of alphanumerically) either by k-value or n-value quick and easy. Let me know if you want me to post it.[/QUOTE]

I have attached the missing 291 k for which there are no primes (yet). In looking at the pl_prime.txt file I suspect that I used ^C early in the run, then started srbsieve again without deleting the pl_*.txt files it created. Based upon what I saw in the file, that makes the most sense.

I assume that you don't need me to resend the pl_prime.txt file.

Let me know if you need anything else from me. Everything is backed up at least once a day (thanks to Time Machine).

[QUOTE=rogue;410296]I have attached the missing 291 k for which there are no primes (yet). In looking at the pl_prime.txt file I suspect that I used ^C early in the run, then started srbsieve again without deleting the pl_*.txt files it created. Based upon what I saw in the file, that makes the most sense.

I assume that you don't need me to resend the pl_prime.txt file.

Let me know if you need anything else from me. Everything is backed up at least once a day (thanks to Time Machine).[/QUOTE]

Did you balance everything?

257 k's remaining that you previously posted for k>19500
291 k's remaining here for k<19500

548 total k's remaining at n=10K

Is that what you are sticking with?

It does not balance with what I have. (See my first response for the # of k's remaining to balance to.) Hint: Here are a few primes shown in your primes file that are also in your remaining file:
480*940^7442-1
1418*940^8834-1
1743*940^8043-1
(plus many more)

As I asked previously, please take a few minutes and balance your files before posting again. If I can't get something that I can trust, I'll have to take the base back into not started status and leave it in the recommended thread. Thank you.

[QUOTE=gd_barnes;410302]It does not balance with what I have. (See my first response for the # of k's remaining to balance to.) Hint: Here are a few primes shown in your primes file that are also in your remaining file:
480*940^7442-1
1418*940^8834-1
1743*940^8043-1
(plus many more)[/QUOTE]

Very odd. I'm not certain how that could have happened. I'll take another look

There were 20 k with primes for which were in pl_remain.txt. I cannot find results for 13035 < k < 19952 and 7000 < n < 10000. I will rerun those k. When done I'll send you updated files.

[QUOTE=rogue;410384]There were 20 k with primes for which were in pl_remain.txt. I cannot find results for 13035 < k < 19952 and 7000 < n < 10000. I will rerun those k. When done I'll send you updated files.[/QUOTE]

No. There were a lot more than 20 k with primes that were in pl_remain.txt. My suggestion is to write a script to find them.

I don't see how that I can accept this submission at this point. Perhaps you can convince me otherwise if you will balance the primes and k's remaining.

Since you already tested this range to n=2500, I could just provide a list of primes up to n=10000. I think that would save us both some time.

Now Iam also in the club. I had a power outtage and my srbsieve was running around 500h. How can I manually start the program to get further on the last position?

[QUOTE=rebirther;410476]Now Iam also in the club. I had a power outtage and my srbsieve was running around 500h. How can I manually start the program to get further on the last position?[/QUOTE]

Unfortunately that is easier said than done. I suggest that you complete manually starting at the beginning of the last phase that was fully completed.

[QUOTE=rogue;410478]Unfortunately that is easier said than done. I suggest that you complete manually starting at the beginning of the last phase that was fully completed.[/QUOTE]

I have 15 phases and was at 22% in phase 14, the ckpt looks like:

phase=14
completedStep=4
completedMaxK=26349
completedKCount=500
inProgressKCount=250

What can I do now?