[QUOTE]42632066-42639970 compared to the earlier 42635862-42647058 leaves 42635862-42639970. 42643801 is outside of this range! Either Tony changed the % slightly, the program he's using round the percentages differently than Prime95 (from the last time I checked), or M47 != M42643801. I wonder which it is.[/QUOTE]

I suppose if we wanted to find out before Tony completes his double check we could always hack [URL="http://en.wikipedia.org/wiki/IBM_Roadrunner"]this[/URL]. I am curious to know myself!!

June 10 - 10:32 CST - 25M
 

[QUOTE][Jun 10 17:32:31] Iter. 25000000 ( 58.63%), CPU (0.0175 sec/iter).
[Jun 10 17:32:31] Saved Interim file at iteration 25000000. Res64: 82F7CE03A2DD99D4.[/QUOTE]Jusqu'ici, tout va bien.
(up to now, everything's fine)
Tony

[QUOTE]Jusqu'ici, tout va bien.
(up to now, everything's fine)
Tony [/QUOTE]

Notice how Tony cleverly avoids commenting on how he might have changed the percentage :grin:

There is no discrepancy if you consider that 58.63% is 25000000/42643801 and round up to 2 decimal places. All other reports match this also.

I know, I'm just trying to keep up the conspiracy theory as there is not much else we can do until the double check comes in.

[quote=Kevin;176812]Nope, the urban legend that sewers have overflowed from so many people watching the Super Bowl on TV simultaneously using the bathroom at half-time.[/quote]
There is a similar large peak of demand of telephone calls during the break or at the end of popular TV programmes.

A few years ago I was with the night shift at Deutsche Telekom's International Network Control Centre in Frankfurt for 1 night. They have a glorious 14 metre by 20 metre display wall on which they were watching TV (and reading emails in a small 2 metre by 2 metre window). They said that this was "to monitor for peaks in demand". ;)

At another Network Control Centre they had TVs, which were for "monitoring the quality of the video lines", especially during major sporting events.

I'm pretty sure that Prime95 does a rounding, not a truncation. If Glucas does the same, Tony's results are still consistent with our suspicions.

[quote=Primeinator;176936]George seems to be fairly [URL="http://www.mersenne.org"]confident[/URL].[/quote]

Mersenne.org seems to be [URL="http://downforeveryoneorjustme.com/www.mersenne.org"]down[/URL].

Edit: it's back.

[QUOTE=CRGreathouse;176785]Has anyone crunched the numbers on just how likely this cluster is, assuming our models are right? If we view the region searched so far as a (scaled) Poisson test, could we reject it as nonrandom at the 5% level?[/QUOTE]

If we plot the natural logs of the Mersenne prime exponents versus the (presumed) Mersenne prime order (i.e., 1 through 47) according to theory, we should get a slope e^(-gamma)*ln(2), about 0.389174, which is a pretty good fit, was in fact quite good just before the discovery of the last prime. Now if we just fit the last 8 data points to a line, we get a slope of about 0.107767, a discrepancy of a factor of 3.61. So I conclude that we should have expected to find about 2.215 primes in this range (about 20 to 48 million) where we actually found 8. How unlikely is that?

[QUOTE]If we plot the natural logs of the Mersenne prime exponents versus the (presumed) Mersenne prime order (i.e., 1 through 47) according to theory, we should get a slope e^(-gamma)*ln(2), about 0.389174, which is a pretty good fit, was in fact quite good just before the discovery of the last prime. Now if we just fit the last 8 data points to a line, we get a slope of about 0.107767, a discrepancy of a factor of 3.61. So I conclude that we should have expected to find about 2.215 primes in this range (about 20 to 48 million) where we actually found 8. How unlikely is that?[/QUOTE]

It does not seem likely but then we can't expect the distribution in any given range to match our "average" distribution. However, perhaps we should be open to the possibility that the previous range (where Mersenne primes seemed to be fairly rare) is the anomaly and not the range we are currently in.

Using the formula I found here:
[url]http://en.wikipedia.org/wiki/Poisson_distribution[/url]
the probability works out to be about 0.00157, quite small, but on the other hand, since there are 40 8-element sequences so far known, I'm guessing that the probability of at least one 8-element clustering this close is about 40 times this probability, or about 6.3%. Please correct any errors, but to me, this seems to support the idea that this clustering is unusual.