theory on Mersenne primes ?

[CRGreathouse]

I don't know what you're saying. I know what sopfr is, and I know that you're giving me p, the exponent of a Mersenne prime. But what am I taking sopfr of and which even numbers am I looking at?

[science_man_88]

Originally Posted by CRGreathouse

I don't know what you're saying. I know what sopfr is, and I know that you're giving me p, the exponent of a Mersenne prime. But what am I taking sopfr of and which even numbers am I looking at?

we know the difference between exponents would be even so according to my idea these even difference would follow sopfr(n) = number such that repeating sopfr on it eventually leads to a prime.

[CRGreathouse]

So... what difference are you taking? SPIT IT OUT, man! Don't make me guess!

[science_man_88]

Originally Posted by CRGreathouse

So... what difference are you taking? SPIT IT OUT, man! Don't make me guess!

mersenne prime exponents! only the fourth time saying it I believe.

[CRGreathouse]

Originally Posted by science_man_88

mersenne prime exponents! only the fourth time saying it I believe.

I think there are more than two of those. Which?

[science_man_88]

Originally Posted by CRGreathouse

I think there are more than two of those. Which?

if you do this with exponent x and x-1 for 40>x>6 as far as i can tell my method always works out. but now I'm not confident it will teach anything.

[CRGreathouse]

Well, since you still won't say (or rather, say clearly) after half a dozen posts I'm through. Have fun with whatever exactly your method is. General tip: whatever prediction you make about Mersenne exponents, show what it predicts about random primes around the size that GIMPS is testing. If it shows that most of them have this property, it's probably not very useful. If they don't, maybe you can investigate further.

[xilman]

Originally Posted by CRGreathouse

Well, since you still won't say (or rather, say clearly) after half a dozen posts I'm through. Have fun with whatever exactly your method is. General tip: whatever prediction you make about Mersenne exponents, show what it predicts about random primes around the size that GIMPS is testing. If it shows that most of them have this property, it's probably not very useful. If they don't, maybe you can investigate further.

I believe he means the separations between successive exponents of Mersenne primes.

At least, that's how I interpret this:

Originally Posted by science_man_88

all of the prime factors including all muliples like in pari I've done this for all the differences between the exponents 39>x>7 as Me(x)-Me(x-1) and continued the resaon I didn't go lower is because 4 ( the difference I would have got next) doesn't work because 2+2 = 4 =2*2=4=2+2 etc. so I know 4 doesn't work but all the differences until x=39 from x=7 work out that the repetitive sum and factor process eventually seems to lead to a prime ( if the difference is 2 then it works just as a sum of all of the prime factors of 2).

(my emphasis added).

It's possible that he may mean all values x-y for x>y where M(x) and M(y) are both prime but that interpretation seems harder to support from the quoted material.

Paul

[science_man_88]

Originally Posted by CRGreathouse

Well, since you still won't say (or rather, say clearly) after half a dozen posts I'm through. Have fun with whatever exactly your method is. General tip: whatever prediction you make about Mersenne exponents, show what it predicts about random primes around the size that GIMPS is testing. If it shows that most of them have this property, it's probably not very useful. If they don't, maybe you can investigate further.

considering you wrote the Me function I'm using and I gave you my full attempt in pari it should be child's play to understand what i mean.

[CRGreathouse]

Originally Posted by science_man_88

considering you wrote the Me function I'm using and I gave you my full attempt in pari it should be child's play to understand what i mean.

Considering you're not telling me the arguments to use...

[science_man_88]

Originally Posted by CRGreathouse

Considering you're not telling me the arguments to use...

The x values 7 to 39 in Me(x)-Me(x-1) is what I showed. continue sopfr from these values and eventually you hit a prime.