biwema,

I'm now at p=20T. I'll start another sieving of [5e9; 25e9] and when it reaches 20T, I will merge them. I will start just now, so there is not much work to catch up.

I have merged [150e6; 5e9] and [5e9; 25e9] this morning. Now the whole range [150e6; 25e9] is at p=21.4T, 10,947,000 k's left, rate is about 1 k/sec.

[QUOTE=gribozavr]I have merged [150e6; 5e9] and [5e9; 25e9] this morning. Now the whole range [150e6; 25e9] is at p=21.4T, 10,947,000 k's left, rate is about 1 k/sec.[/QUOTE]
Excellent progress!

I have released [150e6; 200e6] for LLR at p=30.6T. Now the range [200e6; 25e9] is at p=30.6T, 10,667,000 k's left, rate is ~1.4-1.6 k/sec.

MooooMoo's edit: The ranges are now in the "pre sieved range reservation thread".

Now the range [250e6; 25e9] is at p=51.7T, 10,334,553 k's left, rate is ~2.5-2.7 k/sec.

Now the range [250e6; 25e9] is at p=61.0T, 10,226,786 k's left, rate is ~2.5-2.7 k/sec.

[QUOTE=MooooMoo]
Probability of finding a twin in a range:
my calculations of n=195000 give...
range
k=5 G chance of finding a twin: 31%
k=10 G chance of finding a twin: 52%
k=20 G chance of finding a twin: 77%
k=25 G chance of finding a twin: 84%
k=30 G chance of finding a twin: 89%
k=40 G chance of finding a twin: 95%
k=50 G chance of finding a twin: 97.5%
[/QUOTE]

I can confirm that your calculation is correct. Here it is a small ( PARI ) program to compute the probabilities from 5G to 50G step by 5G:

[CODE]
T=4.0;forprime(p=3,10^5,T*=(p-2)/p/(1-1/p)^2);p=T*(195000*log(2))^-2;\
forstep(i=5,50,5,print("k=",i," G ","chance of finding a twin: ",1-(1-p)^(i*10^9/2)))
k=5 G chance of finding a twin: 0.3032663771235724704662848430
k=10 G chance of finding a twin: 0.5145622587534880610868001205
k=15 G chance of finding a twin: 0.6617792038603679416316561856
k=20 G chance of finding a twin: 0.7643501993734845214281243748
k=25 G chance of finding a twin: 0.8358148606793800287773442252
k=30 G chance of finding a twin: 0.8856066930586734521235834760
k=35 G chance of finding a twin: 0.9202983368219543694155290908
k=40 G chance of finding a twin: 0.9444691714646835051627448017
k=45 G chance of finding a twin: 0.9613098046532592367695271155
k=50 G chance of finding a twin: 0.9730432400262686098144673574
[/CODE]

About the first 5 digits of the probabilities is correct.

Nice.

My Approach was slightly different: I was a bit afraid on how much the thiw is dependant on each other. Therefor I sieved a range of 10G to K=1T, and did the probability calculation with the number of remaining candidates to minimize the dependency. Ofcourse, this number has also a statistical distribution.

The range [300e6; 25e9] is at p=83.0T, 10,005,865 k's left, rate is ~5.0-8.0 sec/k.

Please excuse me, in posts #15, #16 and #17 I made a mistake: I wrote "k/sec", while it is "sec/k". :redface:

hi gribozavr,

Could it be, that you are quite unlucky?


[QUOTE=gribozavr]I have released [150e6; 200e6] for LLR at p=30.6T. Now the range [200e6; 25e9] is at p=30.6T, 10,667,000 k's left, rate is ~1.4-1.6 k/sec.
[/QUOTE]
[QUOTE=gribozavr]
The range [300e6; 25e9] is at p=83.0T, 10,005,865 k's left
[/QUOTE]

My Calculation gives:

10667000 * (24.7G / 24.8G) * (Log2(30.6T) / Log2(83T))^2 = 9972751 Candidates, so there are 28500 Factors too few.

Candidates * New Range * Newfactoring Depth ratio^twin

Is there a flaw in my calculations?

Edit:
More detailed calculations with excel give:
3.06E+13 24.8 10667000
5.17E+13 24.75 10334553 10294806 39747
6.10E+13 24.75 10226768 10227121 -353
8.30E+13 24.7 10005867 10010909 -5042

from 30.6T to 51.7T 39747 too few factors;
from 51.7T to 61T approx. correct
from 61T to 83T 5000 too many

The range [400e6; 25e9] is at p=101.1T, 9,849,791 k's left, rate is ~5.0-10.0 sec/k.

Sieving has passed 100T:banana: :showoff:

biwema, I'm intrested in this math, but I don't understand the "Newfactoring Depth ratio" part. Can you explain it to me, please? (I undersatnd that log2(x) is the number of bits in x)