Strategic Double Clicking - Page 101

Mark Rose · 2016-08-08, 00:20

Quote:

Originally Posted by ATH

I have been keeping a list and updating it with all the exponents from this thread. Once in a while I run a script to check the status of the exponents:

Right now there are 737 exponents from the list that are available that still does not have a successful double check:
available.html

and there are 304 exponents currently assigned:
assigned.html

and there are 1943 exponents we have successfully double checked. That is very nice to see:
verified.html

I was wondering why I kept seeing your name pop up with active assignments throughout the exponents. I guess I just duplicated all your work, except I've divided it between double and triple checks (I've done it for all the exponents, but I haven't posted the lists above 50M).

rudi_m · 2016-08-08, 07:44

Quote:

Originally Posted by ATH

I have been keeping a list and updating it with all the exponents from this thread. Once in a while I run a script to check the status of the exponents:

Right now there are 737 exponents from the list that are available that still does not have a successful double check:
available.html

and there are 304 exponents currently assigned:
assigned.html

and there are 1943 exponents we have successfully double checked. That is very nice to see:
verified.html

In the 1st of your lists you see a lot of exponents that does not have been checked once since they were mentioned in this thread. So I think these ones are still on the original "strategic double check" todo. For example these ones:
50010461
50013547
50035823
50071361
50134391
50183843
50214071
50231387
50251009
50368427
50436173
50440301
50454631
50478761
50484293
50495573
50501257
50522753
50532067
50616283

Madpoo · 2016-08-09, 01:10

The lists I generate tend to vary here and there as more data gets collected from the machines in question. A machine which seemed "iffy" at first may prove to be a little better than expected and don't get included again next time I check.

The really easy pickings were handled early on so over time I've lowered the threshold for what I consider to be "likely bad", and I also run queries over different time periods (how well did that CPU do month-by-month, or annually, or for "all time"?)

Occasionally I'll throw in some where there's just one or two exponents left to check for a machine that merely had more-bad-than-good so we may as well check the few they had left because we're not getting any more useful predictive data anyway.

Well, with all that said, here's a new list... it's a mix of all the different criteria I mentioned above.

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
39110789	2	0	2	0	2	0	DoubleCheck=39110789,71,1
39600287	2	0	5	0	5	0	DoubleCheck=39600287,71,1
39840851	2	0	5	0	5	0	DoubleCheck=39840851,71,1
39923197	2	0	2	0	2	0	DoubleCheck=39923197,71,1
39998071	2	0	5	0	5	0	DoubleCheck=39998071,71,1
39999763	2	0	3	0	3	0	DoubleCheck=39999763,71,1
40059067	3	0	3	0	3	0	DoubleCheck=40059067,72,1
40090997	2	0	2	1	2	1	DoubleCheck=40090997,72,1
40094797	2	0	4	0	4	0	DoubleCheck=40094797,72,1
40094909	2	0	4	0	4	0	DoubleCheck=40094909,72,1
40154327	2	0	5	0	5	0	DoubleCheck=40154327,72,1
40282651	2	0	3	0	3	0	DoubleCheck=40282651,72,1
40357469	2	0	5	0	4	1	DoubleCheck=40357469,72,1
40455187	3	0	1	0	1	0	DoubleCheck=40455187,72,1
40682749	2	0	1	0	1	0	DoubleCheck=40682749,72,1
40693207	2	0	5	0	4	1	DoubleCheck=40693207,72,1
40924393	2	0	4	0	4	0	DoubleCheck=40924393,72,1
40967081	2	0	1	1	1	1	DoubleCheck=40967081,72,1
41000893	3	0	3	0	3	0	DoubleCheck=41000893,72,1
41011319	3	0	3	0	3	0	DoubleCheck=41011319,72,1
41091527	2	0	1	0	1	0	DoubleCheck=41091527,72,1
41115257	2	0	3	0	3	0	DoubleCheck=41115257,72,1
41401181	2	0	3	0	3	0	DoubleCheck=41401181,72,1
41449861	2	0	5	0	4	1	DoubleCheck=41449861,72,1
41486197	2	0	3	0	3	0	DoubleCheck=41486197,72,1
41585671	2	0	1	0	1	0	DoubleCheck=41585671,72,1
41591327	2	0	2	0	2	0	DoubleCheck=41591327,72,1
41730151	2	0	3	0	3	0	DoubleCheck=41730151,72,1
42220699	4	0	1	3	1	3	DoubleCheck=42220699,72,1
42265469	2	0	1	1	1	1	DoubleCheck=42265469,72,1
42295639	2	0	8	0	8	0	DoubleCheck=42295639,72,1
42341239	2	0	8	0	8	0	DoubleCheck=42341239,72,1
42430721	3	1	1	0	1	0	DoubleCheck=42430721,72,1
42693701	2	0	4	0	4	0	DoubleCheck=42693701,72,1
43060453	2	0	5	0	4	1	DoubleCheck=43060453,72,1
43347061	2	0	4	0	4	0	DoubleCheck=43347061,72,1
45531323	2	0	4	0	4	0	DoubleCheck=45531323,72,1
45902009	2	0	4	0	4	0	DoubleCheck=45902009,72,1
46683467	2	0	2	0	2	0	DoubleCheck=46683467,72,1
46686749	2	0	2	1	2	1	DoubleCheck=46686749,72,1
46785359	2	0	2	0	2	0	DoubleCheck=46785359,72,1
47017771	2	0	8	0	8	0	DoubleCheck=47017771,72,1
47869397	2	0	2	0	2	0	DoubleCheck=47869397,72,1
49435933	2	0	5	0	5	0	DoubleCheck=49435933,72,1
50152231	2	0	2	1	2	1	DoubleCheck=50152231,73,1
51728737	2	0	5	0	5	0	DoubleCheck=51728737,73,1
52046131	2	0	2	0	2	0	DoubleCheck=52046131,73,1
52591499	2	0	5	0	5	0	DoubleCheck=52591499,73,1
56874541	2	0	2	0	2	0	DoubleCheck=56874541,73,1
65789629	3	1	5	0	5	0	DoubleCheck=65789629,74,1
65789653	3	1	5	0	5	0	DoubleCheck=65789653,74,1
67173527	3	0	1	0	1	0	DoubleCheck=67173527,74,1
67728877	3	1	5	0	5	0	DoubleCheck=67728877,74,1
67852633	3	1	5	0	5	0	DoubleCheck=67852633,74,1
67852639	3	1	5	0	5	0	DoubleCheck=67852639,74,1
70541467	8	0	2	3	3	2	DoubleCheck=70541467,75,1
70615357	8	0	2	3	3	2	DoubleCheck=70615357,75,1
71992741	2	0	0	1	1	0	DoubleCheck=71992741,75,1
72850889	4	0	1	2	2	1	DoubleCheck=72850889,75,1
73115041	8	0	2	3	3	2	DoubleCheck=73115041,75,1
73138057	3	0	1	0	1	0	DoubleCheck=73138057,75,1
74683481	2	0	3	0	2	1	DoubleCheck=74683481,75,1
75256669	3	0	1	0	1	0	DoubleCheck=75256669,75,1
75818773	2	0	3	0	2	1	DoubleCheck=75818773,75,1
76193773	2	0	1	0	1	0	DoubleCheck=76193773,75,1
77888123	2	0	2	0	2	0	DoubleCheck=77888123,75,1
78401023	25	6	3	0	3	0	DoubleCheck=78401023,75,1
78931819	25	6	3	0	3	0	DoubleCheck=78931819,75,1

Mark Rose · 2016-08-09, 02:13

Quote:

Originally Posted by Madpoo

Well, with all that said, here's a new list... it's a mix of all the different criteria I mentioned above.

I took the ones below 42M.

ric · 2016-08-09, 16:58

Took the 42-43M ones, namely:

Quote:

Originally Posted by Madpoo

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
42220699	4	0	1	3	1	3	DoubleCheck=42220699,72,1
42265469	2	0	1	1	1	1	DoubleCheck=42265469,72,1
42295639	2	0	8	0	8	0	DoubleCheck=42295639,72,1
42341239	2	0	8	0	8	0	DoubleCheck=42341239,72,1
42430721	3	1	1	0	1	0	DoubleCheck=42430721,72,1
42693701	2	0	4	0	4	0	DoubleCheck=42693701,72,1
43060453	2	0	5	0	4	1	DoubleCheck=43060453,72,1
43347061	2	0	4	0	4	0	DoubleCheck=43347061,72,1

ATH · 2016-08-10, 17:51

One of Curtis Cooper's results is bad, though it is 7 years old so that machine probably don't exist anymore: http://mersenne.org/M41425513

It just made me wonder if any of his current machines have mostly bad results? and if yes do you report it to him?

rudi_m · 2016-08-10, 22:31

Quote:

Originally Posted by ATH

One of Curtis Cooper's results is bad, though it is 7 years old so that machine probably don't exist anymore: http://mersenne.org/M41425513

It just made me wonder if any of his current machines have mostly bad results? and if yes do you report it to him?

I wonder if mersenne server would send an email in this case?

Nowadays we have this setting on the "CPU page":
"Send an email if computer returns a suspicious LL result"
Would be nice that we also get this email if someone else proves later that our first check was bad.

GP2 · 2016-08-11, 00:24

Quote:

Originally Posted by ATH

One of Curtis Cooper's results is bad, though it is 7 years old so that machine probably don't exist anymore: http://mersenne.org/M41425513

Dozens if not hundreds of Curtis Cooper's results are bad, or suspected bad. I think he's just using whatever machines are available on campus and inevitably some of them turn out to be bad, given the sheer volume of his production.

Here are some verified bad, there are many more mismatches pending triple checks:

36850763
38577367
39006397
39553141
39637811
39933337
40325119
40405793
40538203
40575707
40673887
40775263
41144893
41299637
42240043
43681081
45921637
46328993
47492827
49510871

Mark Rose · 2016-08-11, 00:36

I just confirmed M77900497. The initial result was turned in way back in 2003.

It took me less than six days. I can only imagine how many months it took the initial user.

endless mike · 2016-08-11, 07:20

All done.

Red is mismatch
Blue is a mismatch, already triple checked and verified

Quote:

Originally Posted by endless mike

From the list, I've claimed some from the top and a few from the bottom

Code:

exponent	Bad	Good	Unk	Sus	Solo	Mis	worktodo
39652001	9	10	1	3	1	3	DoubleCheck=39652001,71,1
40035407	2	2	1	1	1	1	DoubleCheck=40035407,72,1
40058047	15	8	1	0	1	0	DoubleCheck=40058047,72,1
40140053	26	20	1	0	1	0	DoubleCheck=40140053,72,1
40156199	21	17	1	0	1	0	DoubleCheck=40156199,72,1
40340723	2	2	1	1	1	1	DoubleCheck=40340723,72,1
40437613	2	2	1	1	1	1	DoubleCheck=40437613,72,1
40580131	1	0	2	0	1	1	DoubleCheck=40580131,72,1
40606603	11	8	1	0	1	0	DoubleCheck=40606603,72,1
40630201	38	22	1	0	1	0	DoubleCheck=40630201,72,1
40692689	3	2	1	0	1	0	DoubleCheck=40692689,72,1
40788833	2	1	1	0	1	0	DoubleCheck=40788833,72,1
40878647	1	0	1	0	1	0	DoubleCheck=40878647,72,1
40938013	2	1	1	0	1	0	DoubleCheck=40938013,72,1
41228987	5	4	1	0	1	0	DoubleCheck=41228987,72,1
41236483	6	2	1	4	1	4	DoubleCheck=41236483,72,1

75620593	2	1	1	0	1	0	DoubleCheck=75620593,75,1
75737747	3	0	1	1	1	1	DoubleCheck=75737747,75,1
76937989	0	0	0	1	1	0	DoubleCheck=76937989,75,1

GP2 · 2016-08-11, 08:59

Quote:

Originally Posted by Mark Rose

I just confirmed M77900497. The initial result was turned in way back in 2003.

From the datamines:

M77900461
M77909869
M77909939

Thirteen year old unfinished business:

http://www.mersenneforum.org/showthread.php?t=1447
http://www.mersenneforum.org/showthread.php?t=1203

Stay tuned...

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Double-Double Arithmetic	Mysticial	Software	52	2021-04-23 06:51
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

2016-08-10, 17:51	#1106
ATH Einyen Dec 2003 Denmark 2×1,579 Posts	One of Curtis Cooper's results is bad, though it is 7 years old so that machine probably don't exist anymore: http://mersenne.org/M41425513 It just made me wonder if any of his current machines have mostly bad results? and if yes do you report it to him?

2016-08-11, 00:36	#1109
Mark Rose "/X\(‘-‘)/X\" Jan 2013 5562₈ Posts	I just confirmed M77900497. The initial result was turned in way back in 2003. It took me less than six days. I can only imagine how many months it took the initial user.