[QUOTE=gd_barnes;356318]You can't just eliminate [I]all[/I] k's where k is a multiple of the base. You should only eliminate k's (that is write the k to a MOB file) where [I]both[/I] of the following conditions are true:
1. k is a multiple of the base
2. k - 1 is composite (Riesel) or k + 1 is composite (Sierp).

[/QUOTE]

Gary, I don't understand MOB completely-I think I might be missing your point.
k=12000030864= 4000010288*3

When testing from 4-4.1 G, this k will be tested.
When testing from 12-12.1 G, you want to avoid testing this k- as it has already been tested in the 4-4.1 range

[QUOTE=Citrix;356334]Gary, I don't understand MOB completely-I think I might be missing your point.
k=12000030864= 4000010288*3

When testing from 4-4.1 G, this k will be tested.
When testing from 12-12.1 G, you want to avoid testing this k- as it has already been tested in the 4-4.1 range[/QUOTE]

My understanding is that there is a prime at n=1 on the 4 G k and that same prime will be n=0 on the 12 G k. The trouble is we start testing at n=1 so the prime for the 4 G doesn't eliminate the 12 G k.

[QUOTE=Citrix;356334]Gary, I don't understand MOB completely-I think I might be missing your point.
k=12000030864= 4000010288*3

When testing from 4-4.1 G, this k will be tested.
When testing from 12-12.1 G, you want to avoid testing this k- as it has already been tested in the 4-4.1 range[/QUOTE]

You are correct, you are not understanding. I'll ask for a 3rd time: [I]Please [/I]look at the CRUS script and attempt to recreate the code in it. k=12000030864 must be tested. It cannot be eliminated because 12000030864 - 1 = 12000030863 is prime.

Example:

4000010288*3^1-1 is prime. (i.e. 12000030863 is prime)

k=12000030864 does not have a prime up to n=2500 as shown by Peter's run of the CRUS PFGW script.

The key: Yes, 12000030864*3^0-1 is prime [I]but [/I]we do not allow n=0.

Thoughts:

This is why we have two even k's remaining for both Riesel and Sierp base 2 up to n=2M.

By the time that you code for all of the conditions that the CRUS script takes into account and allow it to be tested to n=2500 base 3, you may find that it is not any faster than the CRUS script.

[QUOTE=gd_barnes;356356]You are correct, you are not understanding. I'll ask for a 3rd time: [I]Please [/I]look at the CRUS script and attempt to recreate the code in it. k=12000030864 must be tested. It cannot be eliminated because 12000030864 - 1 = 12000030863 is prime.

Example:

4000010288*3^1-1 is prime. (i.e. 12000030863 is prime)

k=12000030864 does not have a prime up to n=2500 as shown by Peter's run of the CRUS PFGW script.

The key: Yes, 12000030864*3^0-1 is prime [I]but [/I]we do not allow n=0.

Thoughts:

This is why we have two even k's remaining for both Riesel and Sierp base 2 up to n=2M.

[/QUOTE]

Thank you for the explanation. It makes sense now.

Reserving 12G to 13G up to n=25k

[QUOTE=Puzzle-Peter;356833]Reserving 12G to 13G up to n=25k[/QUOTE]

Gary, did you get the results. I just received an email telling me they were deleted from the WeTransfer server. Let me know if I need to upload them again.

[QUOTE=Puzzle-Peter;362476]Gary, did you get the results. I just received an email telling me they were deleted from the WeTransfer server. Let me know if I need to upload them again.[/QUOTE]

I saw the Email but could not download them from my slow internet connection in the hotel room while out of town. I'm back home now. Please upload them again. Sorry about that.

Peter completed k=12G-13G to n=25K a couple of weeks ago. 5984 k's remain for the range.

After a PM exchange with VBCurtis, I am taking over his k=2.05G-2.5G reservation. I will test it to n=25K.

k=2.05G-2.5G is complete to n=25K. 2419 k's remain. The range is released.

There are too many subranges in R3. Let's start consolidating. I'll take k=10M-20M to n=500k.