Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.
|
|
|
|
Re: Twin prime quickie
Posted:
Oct 26, 2004 6:05 PM
|
|
richard miller a écrit :
>
> Today I chanced upon (yeah right) a twin prime candidate - it passes the
> Fermat Test to a few bases. I know its not fool proof, even I have a better
> 'strong pseudo-prime' test. But, can anyone rattle off a quick'ish
> verification.
>
> Twin prime = 10^3024 + 119213059, 10^3024 + 119213061
> (Fermat test approx 25s on my AMD1800MHz, single base)
>
> Last time I looked, 65,000 digits was the baby to beat so I'm well short on
> 3025, but that 65,000 digit number was a 2-power case and I avoid them like
> the plague simply because everyone else doesn't. It mean't I have to slum it
> and use a less efficient multiplication method than FFT, but that's life.
>
> Some others...
>
> Single prime candidate = 10^10008 + 6129 (I call that one 'Debs', after my
> girlfriend)
>
> A nice small one for quick verfification...
>
> Smallest 1001 digit single prime candidate = 10^1000 + 453 ('ELP' - you tell
> me)
> (Fermat test approx <0.5s on my AMD1800MHz, single base)
>
> Smaller twin = 10^1004 + 20077, 20079 (hoepfully quick to verify)
>
> Triplet = 10^512 + 888610317, 888610319, 888610323
>
> Quadruplet = 10^156 + 292536541, 292536543, 292536547, 292536549
Except the 10000-digit one, you can certify the primality of your
primes with Primo.
For 3000 digits, it's a matter of days. For 1000 digits, it
shouldn't take you more than 30 mn. For 500 digits, it will do the
job in a 2 mn. And for 150 digits, you click on the Run button and
that's done.
--
mm
<a href="http://www.ellipsa.net/">http://www.ellipsa.net/</a>
mm@ellipsa.no.sp.am.net ( suppress no.sp.am. )
|
|
|
|
