lowest 10M exponents available - pick your choice

[Batalov]

Originally Posted by cochet

When there is an error code, like that :
33600877,Serge Batalov,homeQ6600,97A2A2D3914C6C__,00000300

do the signification is the necessity of LLtest again for the first time ?

Thanks
Alain

Just in case, for your curiosity -
http://v5www.mersenne.org/report_ll/?exp_lo=33600877
Feel free to triple-check it.

[cochet]

Thank you for your sollicitude. But I was not interessed by this number in particular. I asked you the question in general, in case of error after LLtesting.
Aniway, M45 is for me inferior at 34762857. But it's evident that your will think that it's a numerologist's result, is'nt it ? If you have a little bit time, but I don't believe you will do that, look at the spreadsheet on my thread, it is public !

[cochet]

erratum (j'ai du mal avec la langue anglaise) : M45 is NOT inferior at 34762857 .

[robo_mojo]

cochet do you mean you actually did the LL test but got an error? Sometimes the error is not significant and the result is actually correct.

Did you return the result in to GIMPS?

Anyway, if you can get an error-free LL result on the exponent, and it shows the number composite, no amount of additional work makes the number prime.

[Batalov]

Originally Posted by cochet

... 34762857. But it's evident that your will think that it's a numerologist's result, is'nt it ? If you have a little bit time, but I don't believe you will do that, look at the spreadsheet on my thread, it is public !

34762857 = 3 * 2029 * 5711, i.e. it divides by 3!
Let's have some fun with LaTeX while we are at it...
For all n=3k

(in Cartman's voice) "We've learned something today, we've learned LaTeX"

[robo_mojo]

Originally Posted by Batalov

34762857 = 3 * 2029 * 5711, i.e. it divides by 3!

Oh my. Someone really should have at least made sure they are testing prime exponents, since a mersenne prime necessarily has a prime exponent, which is actually proven. Testing mersenne numbers with composite exponent is a serious waste of time.

[Batalov]

Originally Posted by cochet

Thank you for your sollicitude. But I was not interessed by this number in particular. I asked you the question in general, in case of error after LLtesting.
Aniway, ...

This reminds me of an old anecdote:

There's a long and heated meeting at the city council. People try to choose the new mayor and it's all down to two candidates. Then, the time is after midnight already and people start getting ready to vote, just finishing discussing the candidate A, ...and there's a question from the last row - "Can a child molester be a mayor?" -- "OF COURSE, NOT!" follows the furious answer and the candidate A is promptly dismissed and the meeting is adjourned. People approach that person from the last row who raised the question, and thank him for alerting them of this possible problem. He says -"what problem? I merely asked if a child molester can be a mayor. In general."

Well, anyway, I was just looking for an occasion to tell you an anecdote. As for me, I don't care. I just wanted to check if my own theories about "restarting from an earlier file" etc, and "of course, my system is stable, trust me" were true. So I double-checked on a completely different computer (64-bit, too) and from scratch of course... Now I am content with my own results and with my home computer. What, do you think, _I_ like errors? Nope.

[cochet]

34762857 is a boundary, and not a candidate, obviously. It's an inferior limit.
But when we want to make fun of somebody...

[markr]

I take 33985241 and 34200259. They'll probably take until August 2008.

[JHagerson]

My computer is running well. The number I have checked out is now estimated to complete while I am away for a few days. I took 34200457 and it is currently estimated to be complete in early July.

[markr]

Originally Posted by markr

I take 33985241 and 34200259.

I'll take 34200671, as stage 1 of P-1 found a factor of 34200259.

I did the trial factoring & P-1 since the "1" at the end of the Test= lines didn't seem right with it only being TF'd to 63 bits, and the v5 server shows no P-1 results for this candidate, or the others TF'd to 63 bits.