Oops - New Prime! (M49 related)

[NBtarheel_33]

Originally Posted by Madpoo

You would think so, but the residue for it in the database really is zero which is why I did a double-check.

The fact that it's marked C in that report might just indicate that it hasn't been listed as a known prime or something... I forget how that report decides whether to show a C or not, but that's probably what it is.

Isn't there also the infinitesimal chance that the *entire* residue is non-zero, but each of the last 64 bits just happens to be zero?

[Mark Rose]

Originally Posted by CuriousKit

Given that the type of result is marked as "C", not "P", I will hazard a guess that the last 2 digits of the residue are 02, not 00, and is most likely a hardware error - as described in the "Erroneous values of s_n" topic, if a hardware error causes the intermediate residue to become zero at any point, the next value in the sequence will be -2, and then that will become +2, and then +2 will keep reproducing itself from then on.

There are a couple other exponents like that:

M79060367
M66921341

[CuriousKit]

Strange. I stand corrected.

(Now back to programming my own!)

[Jayder]

I would really urge somebody with authority (likely only George) to consider contacting Brady Haran of Numberphile (his email is at the bottom of his website) so he can work on a video on it with some advance notice. This could be a great opportunity for cross promotion. Perhaps in your estimation, though, it would not be as worth doing as I think it is.

I won't bring it up again. I'm very excited that this prime has been found!

[Nick]

Last time, we had Hendrik Lenstra on Dutch national TV to talk about it (in Dutch):
http://dewerelddraaitdoor.vara.nl/ga...endrik-lenstra

[aketilander]

Originally Posted by Batalov

Now, let's all exercise patience because the publication process and its intricacies are entirely out of our hands. People with trusted skills in this matter will take care of it.

Tomorrow is maybe not a very good day to announce the finding since all science journalists will be occupied reporting that WHO finally has declared the Ebola epedimi to be over. So, better then on Friday?

[Mark Rose]

Originally Posted by aketilander

Tomorrow is maybe not a very good day to announce the finding since all science journalists will be occupied reporting that WHO finally has declared the Ebola epedimi to be over. So, better then on Friday?

Nobody pays attention on Fridays. That's usually when you release bad news :)

[Prime95]

Originally Posted by Mark Rose

Nobody pays attention on Fridays. That's usually when you release bad news :)

The announcement will be Tuesday on 9AM EST. Scott feels this has the best chance of getting the news started and building some traction throughout the week.

Having been through this several times, you never know how big the reaction will be. The last prime was a big to-do, I've seen others barely get a reaction. Fingers crossed.

Another benefit of the Tuesday announcement is it will give Chris Caldwell time to get his web pages updated over the weekend. This week was an especially busy one for him with his real job.

[ixfd64]

This is going to be a long wait...

[ewmayer]

Originally Posted by NBtarheel_33

Exactly how I felt last year as I underwent a root canal and several large replacement fillings. "2015 and we're still going through this? Seriously, isn't there some kind of laser or something?!"

Well, there are laser treatments for many spinal problems these days ... and for the prostate thing there is a blood test, as there is for the colorectal thing (fecal occult blood test ... despite the name I am fairly certain it does not involve standing inside a pentagram or sacrificing a goat.)

Originally Posted by NBtarheel_33

Isn't there also the infinitesimal chance that the *entire* residue is non-zero, but each of the last 64 bits just happens to be zero?

Can't speak for Prime95 or CuLu, but Mlucas, while reporting just the bottom 64 bits of the residue to the .stat (run-status) file, checks all the bits in determining the final test result:

isprime = 1;

/* For Fermat numbers, in balanced-digit form, it's prime if the lowest-order digit = -1, all others 0: */
if(MODULUS_TYPE == MODULUS_TYPE_FERMAT)
	final_res_offset = 1;
else
	final_res_offset = 0;

a[0] += final_res_offset;
for(i=0; i < n; i++)
{
	j = i + ( (i >> DAT_BITS) << PAD_BITS );
	if(a[j] != 0.0) { isprime = 0; break; }
}
a[0] -= final_res_offset;

/*...Unbelievable - I must be hallucinating.	*/
if(isprime)...

[retina]

It is curious that curtisc would double check his own exponent. Later to be checked by MadPoo

http://www.mersenne.org/report_expon...2295777&full=1