[QUOTE=wblipp;114027]I suggest that you should decide now how much testing is sufficient. I've already established my criterion - I'm not spending any time on this method until if has found at least one prime, and not even then unless it has happened in few enough guesses that random guessing would have less than 10% chance of succeeding.

[b]You could pick a criteria like "I want to be 90% confident that the probability a candidate is prime is less than 2%." That's going to be a lot of tests, but we have people here that can help you figure out how many.[/b]

William[/QUOTE]
Cochet, since in your correspondence, you indicate to me that you're not confident in the idea that M45 could be above 40,000,000, and manual reservations would probably only be able to be made above about 43,000,000, I would advise you to accept the sentence I've placed in bold above as a personal challenge.

Come up with a list of numbers you believe might be prime and post them. There is a huge amount of numbers which haven't had a first-pass test completed yet in the range you indicated. Even if you came up with 500 candidates, if one happened to be M45, that would be statistically significant.

Cochet wrote to me with some numbers to try, although the numbers he strongly prefers are not available. Also unfortunate, the numbers he supplied that I told him were still available were picked up before he wrote back to me.

If someone could please supply me, or direct me to, instructions on how to set up sieving lines in the worktodo file, I'd appreciate it. I'm going to do a search right now, but I'm not sure how readily available the information is. I'll come back and post if I find the info.

Edit: As it turns out the info was dead easy to find.

[code]test=[exponent],[how far factored],[how much further to factor]

Off-topic: Should it be 'how much [b]further[/b] to factor' or 'how much [b]farther[/b] to factor'? Further is conceptual movement, and farther is physical movement, but if you're thinking in terms of a number line, couldn't it be both?

Edit2:Cochet sent me some numbers. Most had factors found, 3 were only factored to 62 bits, and 2 were nowhere to be found. I don't know what my rights are in this situation, but...

First, I don't want to post the numbers to the forum at large, and I know George's on vacation, but is he the one that takes care of manual reservations? Anyway, I have five exponents, and there's a good chance that some of them are already reserved. If someone factors them and sends them back, is there any way they can be "nabbed" at that point so I'll have rights to them? Obviously, if they crunch the test and never give them back, even if I prove the number prime, they have the rights to the money. But if the test gets returned, I'd like to claim it if possible.

What are my rights? (And, yes, I know there's only a tiny chance these numbers hide a Mersenne prime. What can I say? I'm obsessive-compulsive)

[QUOTE=jasong;114222]Obviously, if they crunch the test and never give them back, even if I prove the number prime, they have the rights to the money. But if the test gets returned, I'd like to claim it if possible. What are my rights? (And, yes, I know there's only a tiny chance these numbers hide a Mersenne prime. [/QUOTE]AFAIK, reserving an exponent does not guarantee no-one can pick this exponent too, test it, find it is a prime and claim the price, before you have finished to test it. First to claim it is a prime (based on result of prime95, GLucas, MLucas, ...), first to win the price (one the prime has been independently verified), I think. Does everyone agree ?
Tony

[QUOTE=jasong;114222]test=[exponent],[how far factored],[how much further to factor][/Quote]This is not correct : the last parameter if 0 if P-1 factorisation has still to be done and 1 it it has already be done. [QUOTE=jasong;114222]Edit2:Cochet sent me some numbers. Most had factors found, 3 were only factored to 62 bits, and 2 were nowhere to be found. I don't know what my rights are in this situation, but...[/quote]You should check if the two last proposed exponents are prime : if they are nowhere to be found I bet they are not prime :-)

You can allways put those numbers in a worktodo.ini file. If PrimeNet does not protest during communication, they might be assigned to you (it happened for me with 28 000 087 which I just wanted to P-1 factorise.) If they are refused you can check to whom they are assigned in the [url=http://mersenne.org/primenet/status.txt]status.txt[/url] file or if they have already been returned in the [url=http://mersenne.org/primenet/cleared.txt]cleared.txt[/url] file.

Jacob

[QUOTE=jasong;114222]Cochet wrote to me with some numbers to try, although the numbers he strongly prefers are not available. Also unfortunate, the numbers he supplied that I told him were still available were picked up before he wrote back to me.[/QUOTE]

So? Just keep track of the progress on them and post the results as they come in. Why don't you go ahead and post this "master list" of trial candidates, then start checking them off [or on, as the case may be] as the first-time tests finish. Maybe someone could supply a script to auto-parse the relevant primenet logfiles for the candidates, to reduce the manual labor involved.

[QUOTE=ewmayer;114258]So? Just keep track of the progress on them and post the results as they come in. Why don't you go ahead and post this "master list" of trial candidates, then start checking them off [or on, as the case may be] as the first-time tests finish. Maybe someone could supply a script to auto-parse the relevant primenet logfiles for the candidates, to reduce the manual labor involved.[/QUOTE]
Here are the numbers. All prime. If they're not prime, I accidentally deleted a digit.(They should all be between 43,000,000 and 44,000,000

43567373, 43806017, 43583329, 43583333, 43791581

[quote=jasong;114277]Here are the numbers. All prime. If they're not prime, I accidentally deleted a digit.(They should all be between 43,000,000 and 44,000,000

43567373, 43806017, 43583329, 43583333, 43791581[/quote]

A good year or two before Primenet starts allocating these:sad:

[QUOTE=davieddy;114314]A good year or two before Primenet starts allocating these:sad:[/QUOTE]

There's nothing to stop someone testing them right now.

At least 3 of them are being factored right now. Should I assume that they're going to be sent back if no factor is found?

I'm starting mprime up again until I can figure this out.

[quote=Mr. P-1;114323]There's nothing to stop someone testing them right now.[/quote]
And precious little to start them either:smile:

Well. I know that some of you are working on the bracket 37020163-37271995.
Now, it seems to me that it's possible to reduce this interval at :
37245883-37254759.
According to the top of the columns (ref. my Excel sheet), the ideal candidate is then : [B]37250401[/B]
This reduction of interval become possible with the intoduction of two new rules (règle III and règle IV), witch you find the description at his address :
[URL]http://home.tele2.fr/lacanmaths/reglesnew.html[/URL]

I hope you can return an encouraging answer. Who know ?
Thanks
Alain