[QUOTE=pxp;559075]As per my examples, when p = n has been reached, I expect the output file to be sieved to n (and you have said that my expectation is correct). Instead, the program is saying p ~ 200000 but the output file is [I]not[/I] sieved to ~200000 but, rather, to 17. An hour later, p (supposedly, if the program value is to be believed) has advanced by thousands but the new output sieved-to number is still 17. There is no correspondence, so yes, either the sieved-to-number is wrong or the "p=" number in the console is wrong. I can't fathom either possibility which is why I had rather expected you to say that my understanding of the correspondence was [I]incorrect[/I].[/QUOTE]

I see what you are saying. The issue is with the build that you have. There was an issue where it was not choosing the correct value for display on the screen if running multiple threads. The file is to be believed.

The range for x < 17000 has now been double-checked. No missing primes/PRPs.

The range for x < 18000 has now been double-checked. No missing primes/PRPs.

So the range from 17000 to 17999 is an outlier with only 50 PRPs. Interesting.

I started sieving a month ago for all x where 30000 <= x < 40000 (and y < x). Sieving is very slow for such a large range. I will not continue sieving this range or test it at this time. If someone else is interested, please let met know. I will post the file of terms if there is interest. It is only sieved to about 2.6e9 and still has 5.6 million terms in the range.

I am still sieving for 20000 <= x < 30000 (and all y < x). This is also going very slowly. I doubt it will be sieved deeply enough before I finish for x < 20000, but I can split off small y (since those tests are faster) and continue sieving. It is really hard to estimate how much time it will take to test that range because test times vary considerably. It will have around 4 million terms in it. I don't know how far I will test it, but I want to test to x < 25000.

The range for x < 19000 has now been double-checked. No missing primes/PRPs. There are 71 primes/PRPs in that range. I had miscounted in a table in a previous post.

Actually there are 78 primes for 19000 <= x < 20000. I miscounted above.

All x < 20000 has now been double-checked. No missing primes/PRPs.

My tactic for testing 20000 <= x < 30000 is a little different due to speed of sieving. I am testing all x in that range at once, but subdividing by y as PRP tests for smaller y are faster. I am testing for y <= 1000 while sieving for y > 1000 continues.

I have confirmed all known PRPs for x < 30000 where y < 5000. Nothing was missed by previous searchers. Since nothing was missed I am going to discontinue the double-check.

For those who are still searching other ranges for x < 30000, the attached file, sieved to 13896059123, has all x/y pairs that I have not tested for x < 30000. You can use this file in one of two ways. First, see if you tested the x/y pairs that are in this file for the ranges you have tested. Assuming that you haven't sieved as deeply as I have, then this file could have x/y pairs that you didn't test. Second, if you are searching x < 30000 and haven't sieved as deeply as I have, you can use this file to start testing those ranges sooner.