Newer milestone thread - Page 33

Dubslow · 2011-12-02, 05:24

I've been doing that since about halfway through October. As of 18/11/11, the linear regression put completion somewhere in March, so I predicted summer; I didn't get any data between 18/11 and 27/11, but since I've resumed taking data, it's gotten significantly later. (In fact I have a screenshot from three weeks ago somewhere in this forum; I'll upload a new one. Any chance I can get your data that I missed? Or perhaps we can both do it as a Google Doc? I'm currently using LibreOffice.)

(Edit: Link to my previous data.)

Primeinator · 2011-12-02, 06:18

Quote:

Originally Posted by Dubslow

I've been doing that since about halfway through October. As of 18/11/11, the linear regression put completion somewhere in March, so I predicted summer; I didn't get any data between 18/11 and 27/11, but since I've resumed taking data, it's gotten significantly later. (In fact I have a screenshot from three weeks ago somewhere in this forum; I'll upload a new one. Any chance I can get your data that I missed? Or perhaps we can both do it as a Google Doc? I'm currently using LibreOffice.)

(Edit: Link to my previous data.)

Absolutely. I will upload probably tomorrow. My data was taken between the hours of 11pm and 1am CST for those days (GMT -6). Are there any certain file formats you prefer?

I would have done a linear regression already but I thought a dozen data points would have resulted in a very poor prediction. I assume you can't use Excel?

Dubslow · 2011-12-02, 06:22

LibreOffice is capable of reading and writing Microsoft Office formats (including Excel). Would you prefer a Google Doc? (I have 157 data points, ATM.) Actually the more I think about it the better a Google Doc sounds because then (for instance) if I'm not around for a couple of days you can still collect the data. OTOH, as I think about it even more, is Google Docs capable of recording the time as LOSpreadsheet or MOExcel?

Primeinator · 2011-12-02, 15:09

Quote:

Originally Posted by Dubslow

LibreOffice is capable of reading and writing Microsoft Office formats (including Excel). Would you prefer a Google Doc? (I have 157 data points, ATM.) Actually the more I think about it the better a Google Doc sounds because then (for instance) if I'm not around for a couple of days you can still collect the data. OTOH, as I think about it even more, is Google Docs capable of recording the time as LOSpreadsheet or MOExcel?

You're asking the wrong guy. To date, I have never used a Google Document.

Primeinator · 2011-12-02, 23:54

Info for M41 Countdown Starting 11-20

2
2
1
1
1
1
1
1
1
1
0
0

Info for M42 countdown starting 11-20

3,304
3,287
3,266
3,244
3,234
3,213
3,200
3,183
3,164
3,148
3,134
3,114

Info for M43 countdown starting 11-20

64,336
64,237
64,096
63,965
63,955
63,667
63,548
63,404
63,285
63,168
63,014
62,871

Info for M44 Countdown starting 11-20

110,272
110,175
110,034
109,903
109,893
109,605
109,486
109,342
109,223
109,106
108,952
108,808

M45 countdown starting 11-20

203,834
203,737
203,596
203,462
203,452
203,162
203,043
202,899
202,775
202,655
202,499
202,354

M46...

318,075
317,978
317,837
317,703
317,693
317,403
317,283
317,139
317,014
316,894
316,738
316,593

M47...

328,066
327,969
327,828
327,693
327,683
327,393
327,273
327,129
327,004
326,884
326,728
326,583

Would you also like the all exponents below n first time and DC tests as well as all below M46 and M47 tested once?

Dubslow · 2011-12-03, 00:49

Got a Google Doc going. PM me your gmail address; will reply with (short) directions to use (at least) my pile of data. If you'd like, you can upload your other data to other sheets.

Quote:

Originally Posted by Primeinator

My data was taken between the hours of 11pm and 1am CST for those days (GMT -6).

So if your data says 25/11, I should record that as midnight between 25 and 26-11?
So the first row would be midnight between 20 and 21-11?

Primeinator · 2011-12-04, 05:01

Quote:

Originally Posted by Dubslow

Got a Google Doc going. PM me your gmail address; will reply with (short) directions to use (at least) my pile of data. If you'd like, you can upload your other data to other sheets.

So if your data says 25/11, I should record that as midnight between 25 and 26-11?
So the first row would be midnight between 20 and 21-11?

Times are 11pm ish GMT -6 for when my data was collected. For example, the first line would be 11-20-11 at 2300 hours.

I will PM you my address in a few days. I am extremely busy with school at the moment in regards to a couple 20-page papers and my science finals.

Dubslow · 2011-12-04, 05:53

Lol good luck with that. I have two finals on Friday and two in 8-9 days. I haven't been taking data for the other milestones, so ATM that's still up to you. I'll add the M42 data to my document though.

Primeinator · 2011-12-12, 21:12

Just out of curiosity, are we going to make mention on the homepage that we successfully have completed the double check process for M41?

davar55 · 2011-12-31, 11:53

Some numbers from the milestones page:

Code:

All exponents below 24,052,939 have been tested and double-checked.
Countdown to proving M(25964951) is the 42nd Mersenne Prime: 2,650
Countdown to proving M(30402457) is the 43rd Mersenne Prime: 58,961

December 1, 2011: Double-checking proves M(24036583) is the 41st Mersenne prime. 
July 11, 2010: Double-checking proves M(20996011) is the 40th Mersenne prime.

Congratulations on the M41 corroboration. It was worth the weight.

By my calculatons:
24036583 - 20996011 = 3040572
2011/12/1 - 2010/7/11 = 508 days

So the range was completed at a non-adjusted rate
of 5985.4 values per day (ignoring previously tested
values, non-prime even numbers, etc.)

Also: 25964951 - 24036583 = 1928368

Using the same rate yields
322.2 days from 2011/12/1, or 2012/10/19
(I couldn't calculate the exact TOD)
as one estimate for proving M42.

Since as of today there were 2650 LL tests of primes
needed to prove M42,
that calculates to 8.2 LL tests per day in this range.

Dubslow · 2011-12-31, 19:25

I've been recording that number since, I think, sometime in October, and in fact have a month-old linear regression posted somewhere in this forum. I'll update it. (A month ago, the regression hit 0 in 3/12, so I estimated summer.)

Edit: Linear Regression: http://filesmelt.com/dl/CountdownToM42.png
And the exponential regression: http://filesmelt.com/dl/CountdownToM42Exp.png

The exponential regression is pretty severe. Solving for 100 tests left marks summer 2013. Solving for 500 tests left makes 29/09/2012.

The linear regression is a bit nicer. Solving for zero makes for 01/05/2012. (May 1st.)

Edit2: Oh wait, let me fill in that gap there. Primeinator has also been taking data. Okay, modified data:

Linear Regression: http://filesmelt.com/dl/CountdownToM421.png
And the exponential regression: http://filesmelt.com/dl/CountdownToM42Exp1.png

Not much difference regression wise. Linear regression shows 30/04/2012.

Exponential solved for 500 tests remaining is 28/09/2012. Solved for 100 tests remaining is 20/06/2013.

2011-12-02, 05:24	#353
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3×29×83 Posts	I've been doing that since about halfway through October. As of 18/11/11, the linear regression put completion somewhere in March, so I predicted summer; I didn't get any data between 18/11 and 27/11, but since I've resumed taking data, it's gotten significantly later. (In fact I have a screenshot from three weeks ago somewhere in this forum; I'll upload a new one. Any chance I can get your data that I missed? Or perhaps we can both do it as a Google Doc? I'm currently using LibreOffice.) (Edit: Link to my previous data.) Last fiddled with by Dubslow on 2011-12-02 at 05:25

2011-12-04, 05:53	#360
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3×29×83 Posts	Lol good luck with that. I have two finals on Friday and two in 8-9 days. I haven't been taking data for the other milestones, so ATM that's still up to you. I'll add the M42 data to my document though. Last fiddled with by Dubslow on 2011-12-04 at 05:57

2011-12-31, 11:53	#362
davar55 May 2004 New York City 4235₁₀ Posts	Some numbers from the milestones page: Code: All exponents below 24,052,939 have been tested and double-checked. Countdown to proving M(25964951) is the 42nd Mersenne Prime: 2,650 Countdown to proving M(30402457) is the 43rd Mersenne Prime: 58,961 December 1, 2011: Double-checking proves M(24036583) is the 41st Mersenne prime. July 11, 2010: Double-checking proves M(20996011) is the 40th Mersenne prime. Congratulations on the M41 corroboration. It was worth the weight. By my calculatons: 24036583 - 20996011 = 3040572 2011/12/1 - 2010/7/11 = 508 days So the range was completed at a non-adjusted rate of 5985.4 values per day (ignoring previously tested values, non-prime even numbers, etc.) Also: 25964951 - 24036583 = 1928368 Using the same rate yields 322.2 days from 2011/12/1, or 2012/10/19 (I couldn't calculate the exact TOD) as one estimate for proving M42. Since as of today there were 2650 LL tests of primes needed to prove M42, that calculates to 8.2 LL tests per day in this range. Last fiddled with by davar55 on 2011-12-31 at 11:57 Reason: numeric typ0

2011-12-31, 19:25	#363
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3·29·83 Posts	I've been recording that number since, I think, sometime in October, and in fact have a month-old linear regression posted somewhere in this forum. I'll update it. (A month ago, the regression hit 0 in 3/12, so I estimated summer.) Edit: Linear Regression: http://filesmelt.com/dl/CountdownToM42.png And the exponential regression: http://filesmelt.com/dl/CountdownToM42Exp.png The exponential regression is pretty severe. Solving for 100 tests left marks summer 2013. Solving for 500 tests left makes 29/09/2012. The linear regression is a bit nicer. Solving for zero makes for 01/05/2012. (May 1st.) Edit2: Oh wait, let me fill in that gap there. Primeinator has also been taking data. Okay, modified data: Linear Regression: http://filesmelt.com/dl/CountdownToM421.png And the exponential regression: http://filesmelt.com/dl/CountdownToM42Exp1.png Not much difference regression wise. Linear regression shows 30/04/2012. Exponential solved for 500 tests remaining is 28/09/2012. Solved for 100 tests remaining is 20/06/2013. Last fiddled with by Dubslow on 2011-12-31 at 20:18

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Newer X64 build needed	Googulator	Msieve	73	2020-08-30 07:47
Performance of cuda-ecm on newer hardware?	fivemack	GMP-ECM	14	2015-02-12 20:10
Cause this don't belong in the milestone thread	bcp19	Data	30	2012-09-08 15:09
Newer msieves are slow on Core i7	mklasson	Msieve	9	2009-02-18 12:58
Use of large memory pages possible with newer linux kernels	Dresdenboy	Software	3	2003-12-08 14:47

2011-12-02, 06:22	#355
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3·29·83 Posts	LibreOffice is capable of reading and writing Microsoft Office formats (including Excel). Would you prefer a Google Doc? (I have 157 data points, ATM.) Actually the more I think about it the better a Google Doc sounds because then (for instance) if I'm not around for a couple of days you can still collect the data. OTOH, as I think about it even more, is Google Docs capable of recording the time as LOSpreadsheet or MOExcel?

2011-12-02, 23:54	#357
Primeinator "Kyle" Feb 2005 Somewhere near M52.. 3×5×61 Posts	Info for M41 Countdown Starting 11-20 2 2 1 1 1 1 1 1 1 1 0 0 Info for M42 countdown starting 11-20 3,304 3,287 3,266 3,244 3,234 3,213 3,200 3,183 3,164 3,148 3,134 3,114 Info for M43 countdown starting 11-20 64,336 64,237 64,096 63,965 63,955 63,667 63,548 63,404 63,285 63,168 63,014 62,871 Info for M44 Countdown starting 11-20 110,272 110,175 110,034 109,903 109,893 109,605 109,486 109,342 109,223 109,106 108,952 108,808 M45 countdown starting 11-20 203,834 203,737 203,596 203,462 203,452 203,162 203,043 202,899 202,775 202,655 202,499 202,354 M46... 318,075 317,978 317,837 317,703 317,693 317,403 317,283 317,139 317,014 316,894 316,738 316,593 M47... 328,066 327,969 327,828 327,693 327,683 327,393 327,273 327,129 327,004 326,884 326,728 326,583 Would you also like the all exponents below n first time and DC tests as well as all below M46 and M47 tested once?

2011-12-12, 21:12	#361
Primeinator "Kyle" Feb 2005 Somewhere near M52.. 3·5·61 Posts	Just out of curiosity, are we going to make mention on the homepage that we successfully have completed the double check process for M41?