mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   Factoring (https://www.mersenneforum.org/forumdisplay.php?f=19)
-   -   Pascal's OPN roadblock files (https://www.mersenneforum.org/showthread.php?t=19066)

frmky 2021-05-06 01:28

Yes, I think so. Along with computing resources to complete both the sieving and linear algebra.

wreck 2021-05-06 11:34

Congratulations, it's a pleasure to see that msieve could work without crash when unique relations' count larger than 2000M.

henryzz 2021-05-06 18:16

Congratulations for finally getting this problematic number done.

mathwiz 2021-05-08 15:28

[QUOTE=Pascal Ochem;577754]Ryan, frmky, thank you.
Now I explore the subtrees unlocked by the new factors for lower bounds on many parameters of an OPN.[/QUOTE]

Which file(s) of numbers needed will be regenerated as a result of this exploration?

Pascal Ochem 2021-05-10 20:26

For the factor P220 of the C301, I have factored 1+P220 by finding a P48.
And now the roadblock has completely disappeared for the bound [$]10^{2200}[/$].
I started the full run for proving the bounds [$]10^{2100}[/$] (to see how the weights drop)
and [$]10^{2200}[/$] (to hopefully get a new bound).
They will finish in a few weeks and we will have the new mwrb files.
The tXXXX files are unaffected.

The other parameters where we encountered the C301 are
- The total number of prime factors [$]\Omega[/$]: I try to extend the bound from 111 factors to 115.
- [$]\Omega-2\omega[/$]: trying to go from 51 to 53 seems difficult.
In both cases, C301 was also the worst roadblock, although way less bad than for the bound [$]10^{2100}[/$]
In both cases, there are still a few mild roadblocks in the unlocked subtrees, not worth considering.
There is no available file of composites here.

And my webpage needs an update: C301 was the example to explain how to circumvent roadblocks.

Pascal Ochem 2021-05-24 21:35

The run with bound [$]10^{2100}[/$] just finished and we see the impact of the C301.
The old file has 1589 composites.
[url]http://www.lirmm.fr/~ochem/opn/old_mwrb2100.txt[/url]
The new file has 1192 composites.
[url]http://www.lirmm.fr/~ochem/opn/mwrb2100.txt[/url]

Don't worry if you are working on a composite that has disappeared.
It will re-appear in mwrb2200 once the run with bound [$]10^{2200}[/$] is over, because of the composite [$]\sigma(11^{330})[/$].

mathwiz 2021-05-24 22:09

[QUOTE=Pascal Ochem;578993]The run with bound [$]10^{2100}[/$] just finished and we see the impact of the C301.
The old file has 1589 composites.
[url]http://www.lirmm.fr/~ochem/opn/old_mwrb2100.txt[/url]
The new file has 1192 composites.
[url]http://www.lirmm.fr/~ochem/opn/mwrb2100.txt[/url]

Don't worry if you are working on a composite that has disappeared.
It will re-appear in mwrb2200 once the run with bound [$]10^{2200}[/$] is over, because of the composite [$]\sigma(11^{330})[/$].[/QUOTE]

Does this tell us anything about whether the [$]10^{2200}[/$] run will succeed, or are the two completely independent?

wblipp 2021-05-25 05:01

@Pascal

I think you posted the old mwrb file to both links. They both start with

6115909044841454629 16

--------------
@mathwiz
As described in [URL="http://www.lirmm.fr/~ochem/opn/"]Pascal's summary page[/URL], he circumvents roadblocks using a process similar to Kevin Hare's process. This can be applied recursively. In principle, the proof can be extended to any level. In practice, each roadblock causes a large number of additional tree terms, and the limit on the proof is how much time he is willing to commit to generating the proof. Pascal's intuition is that the factoring of the C301 reduced the roadblock complexity sufficiently to allow him to complete the 2200 proof in tolerable time. Unless it turns out he has grossly underestimated the time, he will almost certainly stick to the process long enough to complete the 2200 proof.

ryanp 2021-05-25 05:21

[QUOTE=wblipp;579019]@Pascal

I think you posted the old mwrb file to both links. They both start with

6115909044841454629 16[/QUOTE]

Seems fixed now at least?

[CODE]$ curl -s http://www.lirmm.fr/~ochem/opn/old_mwrb2100.txt | wc -l
1589
$ curl -s http://www.lirmm.fr/~ochem/opn/mwrb2100.txt | wc -l
1192[/CODE]

ryanp 2021-05-25 16:40

[QUOTE=Pascal Ochem;578993]The run with bound [$]10^{2100}[/$] just finished and we see the impact of the C301.
The old file has 1589 composites.
[url]http://www.lirmm.fr/~ochem/opn/old_mwrb2100.txt[/url]
The new file has 1192 composites.
[url]http://www.lirmm.fr/~ochem/opn/mwrb2100.txt[/url]

Don't worry if you are working on a composite that has disappeared.
It will re-appear in mwrb2200 once the run with bound [$]10^{2200}[/$] is over, because of the composite [$]\sigma(11^{330})[/$].[/QUOTE]

I'm doing some ECM work now. Already found one: [url]http://factordb.com/index.php?id=1100000000012016966[/url] (I assume if I just report to FDB, it'll be picked up by Pascal's scraper as usual)?

wblipp 2021-05-25 21:12

[QUOTE=ryanp;579021]Seems fixed now at least?[/QUOTE]

I needed to clear my cache. Probably worked correctly all along if not cached.


All times are UTC. The time now is 20:59.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.