mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PrimeNet (https://www.mersenneforum.org/forumdisplay.php?f=11)
-   -   W T F(actor)... (https://www.mersenneforum.org/showthread.php?t=23338)

lycorn 2018-05-15 19:42

W T F(actor)...
 
It popped up this afternoon (check the Recent Cleared Report):

Gsv Manual testing 5140403 F 2018-05-15 15:39 0.3 0.0000 [B]Factor: 10280807 / TF: 10-20[/B]

How come it had been missed? :w00t:

GP2 2018-05-15 19:50

It's a genuine factor.

However [URL="http://www.mersenne.ca/exponent/5140403"]Mersenne.ca has date of discovery = blank[/URL], so it must have been known. Strange.

Edit:
This factor shows up in earlier snapshots of the list of factors. So this was simply a failure of the duplication suppression check, not a new discovery.

PS,
10280807 is 24 bits, so how could it show up in TF 10–20 ?

lycorn 2018-05-15 20:37

[B]10280807[/B] * 1028080601 = 10569498239325007, a 53-bit factor that pops up imediately when you run P-1 on M5140403. So may be this was the first time that 10280807 was reported "on his own right", therefore being accepted by the server.
In any case it doesn´t explain why the TF effort was recorded as TF 10-20.

ATH 2018-05-15 21:07

The first PRP CF run from 2017-09-20 was run on 3 known factors, so it was known back in September.

chalsall 2018-05-15 21:13

[QUOTE=ATH;487686]The first PRP CF run from 2017-09-20 was run on 3 known factors, so it was known back in September.[/QUOTE]

Just wondering... Why are we so consternated about this factor?

ATH 2018-05-15 21:28

[QUOTE=chalsall;487688]Just wondering... Why are we so consternated about this factor?[/QUOTE]

Because a factor in 5M range with k=1 was reported found today in 2018, when it should have been found in 1996-1997 ?

chalsall 2018-05-15 21:37

[QUOTE=ATH;487690]Because a factor in 5M range with k=1 was reported found today in 2018, when it should have been found in 1996-1997 ?[/QUOTE]

OK. Perhaps the software way back then wasn't as good is it is now.

A few days of GPU effort will settle this once and for all....

GP2 2018-05-15 23:23

[QUOTE=chalsall;487691]OK. Perhaps the software way back then wasn't as good is it is now.[/QUOTE]

This factor was already known to Primenet in 2016, and no doubt much, much, much earlier.

It can't be a software problem. This factor has k=1, it's literally the very first iteration of TF.

Usually when a known factor is re-reported, it is filtered out. But sometimes for some reason, the filtering fails and the known factor appears again in the History. I recall this sort of thing happening before, but can't think of any concrete examples offhand.


All times are UTC. The time now is 10:34.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.