Calcudoku puzzle forum
https://www.calcudoku.org/forum/

Calcudoku.org 20 Million Solved Puzzles Analysis
https://www.calcudoku.org/forum/viewtopic.php?f=2&t=1177		 Page 5 of 10
Author:  paulv66  [ Mon Dec 02, 2019 1:31 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
paulv66 wrote:
paulv66 wrote:
At midnight last night, the projected time for the 20 millionth puzzle being solved was 04:26:41 on 4th February 2020. This is very close to the cut off point between beaker and mikas.

As before, this is a straight linear extrapolation and does not allow for any trends that might be emerging. If I think of it, I'll post a further update based on puzzles solved by midnight next Sunday.


At midnight last night the projected time has moved forward by around 18 hours to 10:13:02 on 03/02/2020.


Another weekly update and the projected time has moved back to 00:15:46 on 04/02/2020.
Author:  beaker  [ Mon Dec 02, 2019 1:53 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
I am seeing a lot less 8x8 timed puzzles being done and if it were to continue, I would imagine time would go back a little further ( and I have stopped doing all 13 of the timed puzzles......only reason I was doing them all was to reach 60,000 puzzles plateau for myself)......Am sure other users are cutting back or increasing their timed puzzle solves in order to affect the time for themselves.....but I think either way, it would be too hard to predict just when 20M will be achieved [lol]......It is going to be just shy of Feb 12th (I think) when I will have solve my 60,000the puzzle.....again hard to predict [smile]
Author:  paulv66  [ Mon Dec 02, 2019 2:06 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
I totally agree beaker. The projections are not intended as predictions - they are mere extrapolations of recent results. When we get to mid January, they may become a bit more meaningful, but in the meantime, I find it interesting to watch the ebb and flow.
Author:  skeeter84  [ Mon Dec 02, 2019 7:23 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
Hey guys, I've decided to do another experiment to see when puzzle 20 million may arrive. I took 3 weeks of data this time around using the same method as in my previous 10 day experiment. Without further ado, I present my data.

Image

Image

Image

Image

Image

I apologize in advance if these images are too small. Anyhow, I also did some rough estimates of when we'll reach 500K and 250K puzzles left to go. As of 12/1/2019 at 18:00:00 US EDT, there were 19,468,300 puzzles solved according to the ticker on the main puzzle page. Because we're headed towards 20 million, there were 531,700 puzzles left to solve at that given point in time.

500K left to go predictions:

US EDT: Thursday, December 5th 2019, between 9:46:05 and 12:59:02
CET: Thursday, December 5th 2019, between 15:46:05 and 18:59:02

250K left to go predictions:

US EDT: anytime from Friday, January 3rd, 2020 at 5:54:14 to Saturday, January 4th, 2020 at 10:36:29

CET: anytime from Friday, January 3rd, 2020 at 11:54:14 to Saturday, January 4th, 2020 at 16:36:29

What we all want to know, of course, is when we'll have zero
puzzles left to go. Here are my new estimates:

US EDT: Monday, January 27th, 2020 at 3:37:26 to Monday, February 3rd, 2020 at 13:22:05
CET: Monday, January 27th, 2020 at 5:08:10 to Monday, February 3rd, 2020 at 19:22:05

My predictions are only that - predictions. I may very well be in the ballpark or very much away from it when everything is said and done. As expected, there was a solid 1,500 to 2,500 dip in the number of puzzles typically solved per day on Thanksgiving. First off, I'm sure there were tons of people eating, drinking, and partying due to it being a holiday. Secondly, the next day is Black Friday, on which tons of people go shopping (besides maybe recover from hangovers and indigestion). Anyhow, I hope everybody had a happy holiday and take care.

skeeter84
Author:  paulv66  [ Mon Dec 09, 2019 5:44 pm ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
paulv66 wrote:
paulv66 wrote:
paulv66 wrote:
At midnight last night, the projected time for the 20 millionth puzzle being solved was 04:26:41 on 4th February 2020. This is very close to the cut off point between beaker and mikas.

As before, this is a straight linear extrapolation and does not allow for any trends that might be emerging. If I think of it, I'll post a further update based on puzzles solved by midnight next Sunday.


At midnight last night the projected time has moved forward by around 18 hours to 10:13:02 on 03/02/2020.


Another weekly update and the projected time has moved back to 00:15:46 on 04/02/2020.


Further update using my somewhat primitive deterministic approach. Projected time has now moved out to 05:26:55 on 04/02/2020.

As I said before, the extent to which the number of solvers fluctuates over the holiday period will probably be the key determinant of when the 20 million mark will be reached.
Author:  paulv66  [ Mon Dec 16, 2019 1:10 pm ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
Projected time has moved a bit closer over the past week. Weekly progression over the past five weeks is as follows:

04:26:41 on 04/02/2020

10:13:02 on 03/02/2020

00:15:46 on 04/02/2020

05:26:55 on 04/02/2020

21:41:30 on 03/02/2020
Author:  paulv66  [ Mon Dec 23, 2019 1:01 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
paulv66 wrote:
Projected time has moved a bit closer over the past week. Weekly progression over the past five weeks is as follows:

04:26:41 on 04/02/2020

10:13:02 on 03/02/2020

00:15:46 on 04/02/2020

05:26:55 on 04/02/2020

21:41:30 on 03/02/2020


Now 00:37:33 on 04/02/2020
Author:  paulv66  [ Mon Dec 30, 2019 2:15 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
paulv66 wrote:
paulv66 wrote:
Projected time has moved a bit closer over the past week. Weekly progression over the past five weeks is as follows:

04:26:41 on 04/02/2020

10:13:02 on 03/02/2020

00:15:46 on 04/02/2020

05:26:55 on 04/02/2020

21:41:30 on 03/02/2020


Now 00:37:33 on 04/02/2020


While fewer puzzles than average were solved over the past week, the difference was somewhat less than I expected and the revised projected time is 12:50:06 on 04/02/2020
Author:  paulv66  [ Mon Jan 06, 2020 1:23 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
paulv66 wrote:
Weekly progression is as follows:

04:26:41 on 04/02/2020

10:13:02 on 03/02/2020

00:15:46 on 04/02/2020

05:26:55 on 04/02/2020

21:41:30 on 03/02/2020

00:37:33 on 04/02/2020

12:50:06 on 04/02/2020


Revised projected time is 19:30:08 on 04/02/2020.

As I will be going to Portugal for a week from next Wednesday, I will not be producing a revised projection until the Sunday after I return.
Author:  eclipsegirl  [ Mon Jan 06, 2020 2:05 am ]
Post subject:  Re: Calcudoku.org 20 Million Solved Puzzles Analysis
Paul,
If non-subscribers take up my plea to become a subscriber and do more puzzles, everyone’s math is off.

[scared]
Page 5 of 10 All times are UTC + 1 hour [ DST ]
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/