mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   LMH > 100M (https://www.mersenneforum.org/forumdisplay.php?f=46)
-   -   332.2M - 333.9M (aka 100M digit range) (https://www.mersenneforum.org/showthread.php?t=10693)

Uncwilly 2013-02-22 01:46

1 Attachment(s)
[QUOTE=LaurV;330305]Can you please post the actual version of the graphic attached to post #421? (right when I got used with the format and I start to like it, you change it! :razz:)[/QUOTE]Well the graphic in 421 and in 428 are not the same data. The first is the status of remaining exponents, the second is the bit level of the factors that have been found.
Here is this month's version of the file in 421.

Uncwilly 2013-02-22 02:54

[QUOTE=Uncwilly;309800]Yesterday Sergey Nosov turned in results on 332250257. He took it from 80 to 82. Our first 82 bit number!:party:[/QUOTE]

[QUOTE=Uncwilly;316111]Oops, he did it again. From 82 to 83. That is a 2,948 GHz-Days credit!![/QUOTE]

[QUOTE=Uncwilly;323266][url]http://www.mersenne.org/report_exponent/?exp_lo=332250257&exp_hi=&B1=Get+status[/url]
:shock:
16 GHz-years!!![/QUOTE]
That sound that you heard was my jaw hitting the floor. Can you say 85 bits?
:chalsall::anurag::curtisc::george:
Would love to have that firepower working on numbers from 76 to 79 or 80.

Uncwilly 2013-03-22 01:05

1 Attachment(s)
Here is the status report for the range from 332192831 to 332399999 (plus a bit more):

[code]Date of data 2013/03/20
Average bit depth for first 100 expos 78.44
Average bit depth for first 1000 expos 77.26
Average bit depth for first 5000 expos 74.39
Average bit depth for first 10000 expos 73.31
100th active expo (no factor found) 332198423
1000th active expo (no factor found) 332246153
5000th active expo (no factor found) 332442137
10000th active expo (no factor found) 332680741
Unitless total effort number 205,225,984
Number of first 100 expos to 2^77 90
Number of first 1000 expos to 2^76 924
Number of first 5000 expos to 2^77 815
Number of first 10000 expos to 2^76 1520
Number left in range 4156

Estimated expos in range to be removed 225
(by taking all expos to 2^79)
Estimated expos in range to be removed 369
(by taking all expos to 2^82)
[/code]
22,638,558 effort equivalent of P-1 done.
4,216,779 effort that has been applied to numbers that have had factors found.[code]Bit # at bit level
73 1962
74 395
75 370
76 632
77 323
78 38
79 406
80 24
81 5
82 0
83 0
84 0
85 1
P-1 262[/code]The number of exponents assigned to L-L's in the 332 range by PrimeNet is currently: 2517.
In the last month 126 exponents have been eliminated between 33219283 and 33299999.
Fancy picture below shows the change in time of average bit level for various numbers of exponents.

Aramis Wyler 2013-04-04 01:15

So I've read the last year or so of posts for the LMH stuff because it seems interesting - I had no idea you guys used to work in the lower ranges. What is the current goal? I see the graphs, but not having spent a lot of time studying them I see progress but not what the goal is.

Uncwilly 2013-04-05 04:02

[QUOTE=Aramis Wyler;336031]So I've read the last year or so of posts for the LMH stuff because it seems interesting - I had no idea you guys used to work in the lower ranges. What is the current goal? I see the graphs, but not having spent a lot of time studying them I see progress but not what the goal is.[/QUOTE]Ok, this answer may run on a bit.

Just to make sure that something is clear:
LMH refers to all of the TF work done way ahead of the LL wave-front. It also referred to TF'ing beyond the 79.3 classical PrimeNet upper limit. the 332M range is just a subset.

The current goal is to get enough exponents TF'ed (and P-1'ed) to a reasonable level before people pick them up for LL. There are a lot people that have pick up expos trying to win the prize for the first 100M digit prime. But they don't tend to take them to 77 bits (the default level for Prime95) or do P-1, much less take them to 79 or 80, like a GPU might. So they risk wasting 4,900+ GHz-days on an expo that has a smallish factor.

Doing a lot of lower level factoring is important, it is needed to eliminate exponents with the lowest amount of work invested.

The numbers and pictures are mainly cheer-leading and progress tracking.

Uncwilly 2013-04-22 23:19

1 Attachment(s)
Here is the status report for the range from 332192831 to 332399999 (plus other data out to 332999999):

[code]Date of data 2013/04/20
Average bit depth for first 100 expos 78.44
Average bit depth for first 1000 expos 77.26
Average bit depth for first 5000 expos 74.45
Average bit depth for first 10000 expos 73.34
100th active expo (no factor found) 332198423
1000th active expo (no factor found) 332246153
5000th active expo (no factor found) 332442211
10000th active expo (no factor found) 332680921
Unitless total effort number 208,347,136
Number of first 100 expos to 2^77 90
Number of first 1000 expos to 2^76 924
Number of first 5000 expos to 2^77 830
Number of first 10000 expos to 2^76 1563
Number left in range 4154

Estimated expos in range to be removed 223
(by taking all expos to 2^79)
Estimated expos in range to be removed 366
(by taking all expos to 2^82)
[/code]
23,712,566 effort equivalent of P-1 done.
4,216,779 effort that has been applied to numbers that have had factors found.[code]Bit # at bit level
73 1898
74 410
75 375
76 662
77 329
78 40
79 410
80 25
81 5
82 0
83 0
84 0
85 1
P-1 275[/code]The number of exponents assigned to L-L's in the 332 range by PrimeNet is currently: 2408 (down about 200 from the peak).
In the last month all exponents between 33219283 and 33299999 have been taken to at least 72 bits.
This months picture shows the exponent count by bit level over time for the range up to 332599999

Uncwilly 2013-06-21 04:13

1 Attachment(s)
Here is the status report for the range from 332192831 to 332399999 (plus other data out to 332999999):

[code]Date of data 2013/06/20
Average bit depth for first 100 expos 78.48
Average bit depth for first 1000 expos 77.36
Average bit depth for first 5000 expos 74.61
Average bit depth for first 10000 expos 73.46
100th active expo (no factor found) 332198423
1000th active expo (no factor found) 332246281
5000th active expo (no factor found) 332442673
10000th active expo (no factor found) 332681873
Unitless total effort number 219,713,536
Number of first 100 expos to 2^77 90
Number of first 1000 expos to 2^76 928
Number of first 5000 expos to 2^77 873
Number of first 10000 expos to 2^76 1609
Number left in range 4150

Estimated expos in range to be removed 218
(by taking all expos to 2^79)
Estimated expos in range to be removed 361
(by taking all expos to 2^82)
[/code]
[code]Bit # at bit level
73 1757
74 440
75 441
76 667
77 338
78 43
79 428
80 26
81 9
82 0
83 0
84 0
85 1
P-1 285[/code]The number of exponents assigned to L-L's in the 332 range by PrimeNet is currently: 2255 (down about 400 from the peak).

Aramis Wyler 2013-09-23 16:00

Update please! :)

Uncwilly 2013-09-24 00:41

1 Attachment(s)
Here is the overdue status report for the range from 332192831 to 332399999 (plus other data out to 332999999):

[code]Date of data 2013/09/20
Average bit depth for first 100 expos 78.55
Average bit depth for first 1000 expos 77.61
Average bit depth for first 5000 expos 74.76
Average bit depth for first 10000 expos 73.76
100th active expo (no factor found) 332198423
1000th active expo (no factor found) 332246489
5000th active expo (no factor found) 332442961
10000th active expo (no factor found) 332683957
Unitless total effort number 246,185,984
Number of first 100 expos to 2^77 90
Number of first 1000 expos to 2^76 944
Number of first 5000 expos to 2^77 946
Number of first 10000 expos to 2^76 1667
Number left in range 4141

Estimated expos in range to be removed 208
(by taking all expos to 2^79)
Estimated expos in range to be removed 351
(by taking all expos to 2^82)
[/code]
[code]Bit # at bit level
73 1590
74 357
75 626
76 654
77 301
78 54
79 522
80 25 (had been all the way up to 27)
81 11
82 0
83 0
84 0
85 1
P-1 297[/code]The number of exponents assigned to L-L's in the 332 range by PrimeNet is currently: 2300
Attached image is for the total effort by bit level for the classic range.

Uncwilly 2013-09-24 00:43

[QUOTE=Uncwilly;326772]The 332M range is now at 59% factored.
:busy:[/QUOTE]As of today it is 59.65%

Aramis Wyler 2013-10-12 01:14

That's rough, .65% in 9 months. I get more and more tempted every day to start poaching assigned numbers in the first 500 so as to get the avg on them up to 79.


All times are UTC. The time now is 22:40.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.