mersenne.ca

[bhelmes]

A peaceful night for you,

is it possible to add a button for the search with next exponent without factor and untested
under mersenne.ca/exponent/ ?

Would be helpful for me and I think also for other persons.

Thanks for your work.



P.S. The energy prices are falling and the question is, when will the power companies decrease also their prices.

[James Heinrich]

Originally Posted by bhelmes

is it possible to add a button for the search with next exponent without factor and untested
under mersenne.ca/exponent/

There's already buttons for next/previous: (< exponent >), (<< exponent with factor >>), (<! exponent without factor !>), (<<< Mersenne Prime >>>)

[Rubiksmath]

Originally Posted by James Heinrich

Poor P-1 page always calculates probability based on Primenet-default TF level, regardless of what TF has actually been done, or ECM, or previous P-1 effort. This is to give a consistent evaluation of how "good" the existing P-1 bounds are relative to the exponent, without being thrown off by somebody TF'ing the exponent 10 bit levels higher or whatever the case may be.

Ah okay, makes more sense now. Doing it this way would prevent cases where the P-1 bounds are fine but the TF is absurd from showing up on the poorly factored page when in fact they have been factored just fine. The downside is that for exponents where TJAOI limit is above the PrimeNet default they wouldn't show up since their probability is >20% with default TF but this is a minority of cases.

After all, this is meant to be a list of "poorly factored exponents" and not a page that sorts exponents by actual P-1 probability, especially in ranges where there are no poorly factored exponents.

[Andrew Usher]

That minority, though, contains almost all exponents below 1M as the PrimeNet default there falls off a cliff to 2^40, making any good P-1 >30%. The most reasonable solution would be to set a minimum for assumed TF - for example, 2^56, which would affect only the exponents <1M that don't really work at all now - or 2^68, nearest to the TJAOI level, and having the merit of being constant across almost all the work on factored exponents, while not affecting those above the DC limit the tool was originally meant for.

[Mark Rose]

The TF bounds graph for this exponent has a rendering bug. The width of the pink div for 2⁷⁴ is set to 146% for whatever reason.

Perhaps it has to do with the no factor from 2¹ to 2⁷⁴ manual result line? I only mention that because it's unusual.

[James Heinrich]

Originally Posted by Mark Rose

The TF bounds graph for this exponent has a rendering bug.

Not really a bug -- the TF effort is from 1-74, and the TF graph starts at 2⁵⁰, the coloured segment for 1-74 starts (arguably correctly) where 2¹ would be, which is negative-X off the left of the window. That that happens for any exponent with TF < 2⁵⁰.

[Mark Rose]

Originally Posted by James Heinrich

Not really a bug -- the TF effort is from 1-74, and the TF graph starts at 2⁵⁰, the coloured segment for 1-74 starts (arguably correctly) where 2¹ would be, which is negative-X off the left of the window. That that happens for any exponent with TF < 2⁵⁰.

Makes sense. Just looks a little weird.

[kruoli]

In the torrent summary file, there is a link that makes in non-extractable on Windows. The link file has the same name as the file above it. Additionally, some files are by the root user.

[James Heinrich]

Thanks, fixed.

[James Heinrich]

Originally Posted by kruoli

In the torrent summary file

If you want to help seed, you should Download all finalized archives (1997-2022, torrent) to get the files just once, and then load all the various .torrent files from all finalized .torrent files (if you want to seed) into your client and point them to where you downloaded all the finalized archives. If you download all the individual torrents you will end up downloading each file 3 times over (once in Finalized Archives, twice in the group-by-year set, thrice in the group-by-G-Range set).

[kruoli]

That's would I should have done, but since disk space is no problem here I was not bothering to do that. :D

One additional note: The last archive I downloaded had a file known_factors_1997-2021 which is most likely old and has no seeders at all.