Strategic Double Clicking - Page 150

Madpoo · 2018-01-24, 22:19

I haven't done an updated "potentially bad" list in a while, so here goes. Top 100 by their bad/good ratio, then re-sorted by exponent size:

Code:

exponent	Bad	Good	Unk	worktodo
43907971	1	1	1	DoubleCheck=43907971,72,1
45210229	1	0	3	DoubleCheck=45210229,73,1
45270061	1	0	3	DoubleCheck=45270061,73,1
45554557	1	0	1	DoubleCheck=45554557,73,1
45639227	1	0	1	DoubleCheck=45639227,73,1
45771151	1	0	1	DoubleCheck=45771151,73,1
46041311	1	0	1	DoubleCheck=46041311,73,1
46326053	1	0	1	DoubleCheck=46326053,73,1
46480969	2	3	3	DoubleCheck=46480969,73,1
46481797	2	3	3	DoubleCheck=46481797,73,1
46731457	1	1	2	DoubleCheck=46731457,75,1
46880387	2	0	2	DoubleCheck=46880387,73,1
46893641	1	0	1	DoubleCheck=46893641,73,1
47164939	1	1	8	DoubleCheck=47164939,73,1
47186539	1	1	2	DoubleCheck=47186539,73,1
47236151	1	1	8	DoubleCheck=47236151,73,1
47294899	1	1	2	DoubleCheck=47294899,73,1
49481087	3	4	2	DoubleCheck=49481087,73,1
49883291	1	0	1	DoubleCheck=49883291,73,1
50331461	1	1	1	DoubleCheck=50331461,73,1
50391623	1	1	5	DoubleCheck=50391623,73,1
50461591	1	1	3	DoubleCheck=50461591,73,1
50745581	1	0	1	DoubleCheck=50745581,73,1
51774959	1	1	5	DoubleCheck=51774959,73,1
51783289	1	0	1	DoubleCheck=51783289,73,1
51867259	1	1	3	DoubleCheck=51867259,73,1
51968027	1	1	10	DoubleCheck=51968027,73,1
51987197	1	1	10	DoubleCheck=51987197,73,1
52009949	1	1	9	DoubleCheck=52009949,73,1
52012523	1	1	9	DoubleCheck=52012523,73,1
52012553	1	1	9	DoubleCheck=52012553,73,1
52162093	1	1	10	DoubleCheck=52162093,73,1
52242319	1	1	4	DoubleCheck=52242319,73,1
52316051	1	1	10	DoubleCheck=52316051,73,1
52433503	1	1	10	DoubleCheck=52433503,73,1
52451279	1	1	10	DoubleCheck=52451279,73,1
52500229	1	1	10	DoubleCheck=52500229,73,1
52582693	1	1	10	DoubleCheck=52582693,73,1
52582697	1	1	10	DoubleCheck=52582697,73,1
52731869	1	1	10	DoubleCheck=52731869,73,1
52751147	1	1	5	DoubleCheck=52751147,73,1
52852691	1	1	9	DoubleCheck=52852691,73,1
53036561	1	1	5	DoubleCheck=53036561,73,1
53149091	1	1	5	DoubleCheck=53149091,73,1
53155153	1	1	5	DoubleCheck=53155153,73,1
53625107	1	1	10	DoubleCheck=53625107,73,1
53627617	2	3	6	DoubleCheck=53627617,73,1
53669309	2	3	6	DoubleCheck=53669309,73,1
53691199	1	1	6	DoubleCheck=53691199,73,1
53693557	2	3	6	DoubleCheck=53693557,73,1
53926087	1	1	8	DoubleCheck=53926087,73,1
54451681	1	0	3	DoubleCheck=54451681,73,1
54468013	1	1	3	DoubleCheck=54468013,73,1
54469427	1	1	8	DoubleCheck=54469427,73,1
54538031	1	1	5	DoubleCheck=54538031,73,1
54539983	1	1	10	DoubleCheck=54539983,73,1
54586663	1	1	4	DoubleCheck=54586663,73,1
54698923	1	1	6	DoubleCheck=54698923,74,1
54852011	2	2	4	DoubleCheck=54852011,73,1
54852407	2	2	4	DoubleCheck=54852407,73,1
55100491	1	1	4	DoubleCheck=55100491,73,1
55448333	1	1	10	DoubleCheck=55448333,73,1
55571273	2	1	5	DoubleCheck=55571273,73,1
55588213	2	1	5	DoubleCheck=55588213,73,1
55588349	2	1	5	DoubleCheck=55588349,73,1
55607131	2	1	5	DoubleCheck=55607131,73,1
55633783	2	1	5	DoubleCheck=55633783,73,1
55776619	1	1	6	DoubleCheck=55776619,74,1
55797523	1	1	6	DoubleCheck=55797523,74,1
55818923	1	1	6	DoubleCheck=55818923,74,1
55829551	1	1	6	DoubleCheck=55829551,74,1
55962287	1	1	6	DoubleCheck=55962287,74,1
56019877	1	1	6	DoubleCheck=56019877,74,1
56217779	1	1	9	DoubleCheck=56217779,73,1
56221273	1	1	9	DoubleCheck=56221273,73,1
56365151	2	3	1	DoubleCheck=56365151,73,1
56655779	1	1	4	DoubleCheck=56655779,73,1
56691853	1	1	4	DoubleCheck=56691853,73,1
56699003	1	1	4	DoubleCheck=56699003,73,1
56750671	1	1	4	DoubleCheck=56750671,73,1
56829193	1	1	4	DoubleCheck=56829193,73,1
56849263	1	1	6	DoubleCheck=56849263,73,1
56857217	1	1	3	DoubleCheck=56857217,73,1
57239123	1	1	8	DoubleCheck=57239123,74,1
57285229	1	1	8	DoubleCheck=57285229,74,1
57307457	1	1	8	DoubleCheck=57307457,74,1
57321293	1	1	8	DoubleCheck=57321293,74,1
57363377	1	1	8	DoubleCheck=57363377,74,1
57408977	1	1	8	DoubleCheck=57408977,74,1
57427057	1	1	8	DoubleCheck=57427057,74,1
57454099	1	1	8	DoubleCheck=57454099,74,1
58885549	1	1	6	DoubleCheck=58885549,74,1
59153929	1	1	5	DoubleCheck=59153929,74,1
59211871	1	1	5	DoubleCheck=59211871,74,1
59219011	1	1	5	DoubleCheck=59219011,74,1
59429039	1	1	5	DoubleCheck=59429039,74,1
59528437	1	1	5	DoubleCheck=59528437,74,1
59648437	1	1	8	DoubleCheck=59648437,74,1
59856241	1	1	8	DoubleCheck=59856241,74,1
59939849	1	1	8	DoubleCheck=59939849,74,1

cuBerBruce · 2018-01-24, 22:52

Quote:

Originally Posted by Madpoo

I haven't done an updated "potentially bad" list in a while, so here goes. Top 100 by their bad/good ratio, then re-sorted by exponent size:

Code:

exponent	Bad	Good	Unk	worktodo
43907971	1	1	1	DoubleCheck=43907971,72,1
45210229	1	0	3	DoubleCheck=45210229,73,1
45270061	1	0	3	DoubleCheck=45270061,73,1
45554557	1	0	1	DoubleCheck=45554557,73,1

I took these first four.

Prime95 · 2018-01-25, 02:03

Excellent timing. I took the next 12:

Code:

45639227	1	0	1	DoubleCheck=45639227,73,1
45771151	1	0	1	DoubleCheck=45771151,73,1
46041311	1	0	1	DoubleCheck=46041311,73,1
46326053	1	0	1	DoubleCheck=46326053,73,1
46480969	2	3	3	DoubleCheck=46480969,73,1
46481797	2	3	3	DoubleCheck=46481797,73,1
46731457	1	1	2	DoubleCheck=46731457,75,1
46880387	2	0	2	DoubleCheck=46880387,73,1
46893641	1	0	1	DoubleCheck=46893641,73,1
47164939	1	1	8	DoubleCheck=47164939,73,1
47186539	1	1	2	DoubleCheck=47186539,73,1
47236151	1	1	8	DoubleCheck=47236151,73,1

ixfd64 · 2018-01-27, 07:24

Silly question: do these lists include exponents tested by computers with bad P-1 results?

If a computer has missed a P-1 factor, then I imagine it's likely to return bad LL results as well.

kruoli · 2018-01-27, 10:30

Some of them are at least missing stage 2 of P-1.

Madpoo, can you generate a list of those which are missing stage 2? Then I'll do all of them. I'll start a single run of 4948107.

Madpoo · 2018-01-28, 18:50

Quote:

Originally Posted by ixfd64

Silly question: do these lists include exponents tested by computers with bad P-1 results?

If a computer has missed a P-1 factor, then I imagine it's likely to return bad LL results as well.

There's really no easy way for me to look at a P-1 result and say with any certainty "this guy missed finding a factor".

Then again, not many people are doing P-1 work so it might not be that useful anyway.

chalsall · 2018-01-28, 19:46

Quote:

Originally Posted by Madpoo

Then again, not many people are doing P-1 work so it might not be that useful anyway.

I would disagree with this statement.

Some people are doing a *serious* amount of P-1 work. There is a reason no LL assignments have been handed out which need P-1 work done for several years now.

chalsall · 2018-01-28, 20:45

Quote:

Originally Posted by kruoli

Some of them are at least missing stage 2 of P-1. Madpoo, can you generate a list of those which are missing stage 2? Then I'll do all of them. I'll start a single run of 4948107.

Just so you know, James' mersenne.ca site proves such information. Please see this query for example.

Some of the more anally retentive of us are currently doing P-1 work in the lower ranges with better parameters, so please try to avoid duplication of work.

ixfd64 · 2018-01-28, 23:53

Quote:

Originally Posted by Madpoo

There's really no easy way for me to look at a P-1 result and say with any certainty "this guy missed finding a factor".

James has a page that lists factors missed by P-1: http://mersenne.ca/p1missed.php

LaurV · 2018-01-29, 07:02

Quote:

Originally Posted by Madpoo

There's really no easy way for me to look at a P-1 result and say with any certainty "this guy missed finding a factor".

People, people, wait a moment, there is a confusion here.

kruoli asked for a list of exponents with missing stage 2, not missing factors (this nobody knows, until a "missed" factor is found, by a subsequent run, and not many people re-run P-1 -- I guess that is what Madpoo wanted t say). ixfd64 pointed to a list of missing factors, which is irrelevant for the discussion (it was not asked for, and even if it should have been requested, it is useless for the goal, the man just wants a list with missing stage 2 thingies, most probably to try running stage 2 by himself, and get some nice factors, with cudaPM1 or so).

The list is easy to generate, put into it all exponents where B1==B2, or (in an extended sense) where B2/B1<20 (arbitrary number 20, usually, with a proper stage 2 done, B2 will be 30, 50 or 100 or 500 times larger than B1, depending what you use for the two stages).

The list already exists on James' site, just do the proper query (for example, I used to re-work with CudaPM1 exponents where less than 5% chance to find a factor was done).

So, Madpoo, can you drop a list where (say) B2/B1<xxx ? It may be useful for some of us (or not).

Dubslow · 2018-01-29, 07:09

Quote:

Originally Posted by LaurV

People, people, wait a moment, there is a confusion here.

kruoli asked for a list of exponents with missing stage 2, not missing factors (this nobody knows, until a "missed" factor is found, by a subsequent run, and not many people re-run P-1 -- I guess that is what Madpoo wanted t say). ixfd64 pointed to a list of missing factors, which is irrelevant for the discussion (it was not asked for,

But it was asked for, by ixfd64 himself (1643), before kruoli made his own comment (1644).

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Double-Double Arithmetic	Mysticial	Software	50	2017-10-30 19:16
Clicking an exponent leads to 404 page	marigonzes	Information & Answers	2	2017-02-14 16:56
x.265 half the size, double the computation; so if you double again? 1/4th?	jasong	jasong	7	2015-08-17 10:56
What about double-checking TF/P-1?	137ben	PrimeNet	6	2012-03-13 04:01
Double the area, Double the volume.	Uncwilly	Puzzles	8	2006-07-03 16:02

2018-01-27, 07:24	#1643
ixfd64 Bemusing Prompter "Danny" Dec 2002 California 2,351 Posts	Silly question: do these lists include exponents tested by computers with bad P-1 results? If a computer has missed a P-1 factor, then I imagine it's likely to return bad LL results as well.

2018-01-27, 10:30	#1644
kruoli "Oliver" Sep 2017 Porta Westfalica, DE 11·37 Posts	Some of them are at least missing stage 2 of P-1. Madpoo, can you generate a list of those which are missing stage 2? Then I'll do all of them. I'll start a single run of 4948107.