statistics questions

Hi Louie,

When i look into the stats i see some positions set with n/a.

For example for k=28433 there are no lower and upper bounds. Does this mean that this k value will not be worked on by SoB ?

I see that there is an upper bound for k=19249. I think that means that you have an aggreement with the other contributor about how works on which n values ?

Then i have a feature request for the stats. Is it possible to show how many n values are not yet send out but are put into your database. So there would be a better feeling about the open work and one could see how the amount of work is reduced by extra sieving.

My last question concerns the k values where we found a prime. Will the open n values lower then the prime send out to us or will there be an internal calculation to finish the work.

Is it possible to display who many n values are untestet for the primes ??

Best regards,

Lars

The status of the numbers that are not (yet?) on the SoB page can be found at the following locations:

K=4847
K=19249
K=28433

Why does the stat page say that good ol' 27653 have a lower bound of 466124:confused:

27653 7182 tests 49 tests 466124 2999769

That surprised me too. Last night the lower bound was something like 2.2M , and the number of tests pending below 40

There are several changes to the lower bounds. Maybe they have found some missing values - like when you tidy up the loft... ;)

Next question.

How is it possible that the upper bound is higher then the min n value of the current test window ? :confused:

For example k=10223 "n bounds upper"=767020" and test window min n=693641.

And the description for n upper bound says: All !! n values less or equal to n have been tested.

Lars

k=27653 is having 9 tests below n=2.3mill retested. that's why there is a display error in the lowerbound.

-Louie

Quote:

---

Originally posted by ltd
How is it possible that the upper bound is higher then the min n value of the current test window ? :confused:

. . .
Lars [/B]

---

I was wondering this myself and searching the forum I found your question, but no answer.

The only thing I can think of is that n Upper Bound is updated more frequently than Current Test Window Min n.

DP

I guess it's not due to a lag in updates. It looks like the updates are automated.

Some values are retested for double check or some other reason we don't know (and that's ok). Since those values are already tested at least once, it's ok to see n upper bound higher than Current Test Window Min n.