Anyone factoring <5M? - Page 8

markr · 2010-10-12, 11:37

Quote:

Originally Posted by markr

There's still enough [factors]. A quick bit of counting text strings in current & old results files gives 113 factored out of 10181 attempts, for TF from 61 to 62 'bits' between about 4500000 & 5000000. That's 1.1%.

I'm only doing this TF because of two machines I have that are still in use, old Athlon XP's which are good below 2^64 and really shine below 2^62. Doing more P-1 on exponents in this region with relatively little already done is far more productive, even though a P-1 result with the parameters I use is less effort than TF, as measured by the credit.

It took a little mucking about to separate out the P-1-small, but I finally worked out some stats for my currently-main P-1-small machine. Its success rate to date is 4.7% (163/3445) in the 3M & 4M ranges. About 85% of the factors came from stage 2.

It took one core of a core2 quad in the region of 7-10 weeks, but I think it was worth it for 163 factors.

lorgix · 2010-10-12, 11:58

Quote:

Originally Posted by markr

It took a little mucking about to separate out the P-1-small, but I finally worked out some stats for my currently-main P-1-small machine. Its success rate to date is 4.7% (163/3445) in the 3M & 4M ranges. About 85% of the factors came from stage 2.

It took one core of a core2 quad in the region of 7-10 weeks, but I think it was worth it for 163 factors.

Nice job!

I'm back below 5M again btw...

Mostly doing P-1 in the 2~2.5M area (112-128K FFT), but also a little in 4.5~5M.

[Speaking of factors.... anyone who hasn't checked out http://factorization.ath.cx/ yet should do it now.]

petrw1 · 2010-10-18, 21:18

Update in 3M range.

7 Factors out of 816 tests: 0.86%

petrw1 · 2010-10-19, 21:08

Quote:

Originally Posted by petrw1

Update in 3M range.

7 Factors out of 816 tests: 0.86%

How quickly things changer:

8 out of 830 = .96%

gjmccrac · 2010-10-20, 12:54

In the 4.0 to 4.1M range (61 to 62)

4 factors out of 541 = 0.74%

markr · 2010-10-20, 20:54

Quote:

Originally Posted by gjmccrac

In the 4.0 to 4.1M range (61 to 62)

4 factors out of 541 = 0.74%

Cool! That's consistent with a success rate about 1%.

chessmc · 2010-10-24, 02:04

Have a gpu working in the 1.0-1.01M range (up to 64 bits).

alpertron · 2010-10-25, 11:44

I'm running ECM in the range 400000-401000 (100 curves with B1=250000, B2=25000000). The computer found 3 factors for the 15 exponents it tested. There are still 3 exponents to finish this range. So the success rate in this case was 20%.

petrw1 · 2010-10-25, 22:23

Quote:

Originally Posted by petrw1

How quickly things changer:

8 out of 830 = .96%

10 / 911 = 1.1%

lorgix · 2010-11-13, 10:33

Hi everybody,

I figured this thread might be a good place to ask;

Does anyone know what the status is on missed small factors?
Factors that should have been found by TF already.

I know there are threads about this, but I haven't seen one that appears to be up to date.

garo · 2010-11-14, 22:25

In what range? It is generally agreed that it is not worth the effort to refactor all ranges to catch the few factors that may have been missed due to hardware error. A couple of years ago I did a stats analysis and found some ranges that had fewer factors than expected and these were subsequently refactored.

2010-10-18, 21:18	#80
petrw1 1976 Toyota Corona years forever! "Wayne" Nov 2006 Saskatchewan, Canada 3×5²×71 Posts	Update in 3M range. 7 Factors out of 816 tests: 0.86%

2010-10-20, 12:54	#82
gjmccrac Aug 2009 Ontario, Canada 2×67 Posts	In the 4.0 to 4.1M range (61 to 62) 4 factors out of 541 = 0.74%

2010-10-24, 02:04	#84
chessmc Sep 2010 3₁₀ Posts	Have a gpu working in the 1.0-1.01M range (up to 64 bits).

2010-10-25, 11:44	#85
alpertron Aug 2002 Buenos Aires, Argentina 5F3₁₆ Posts	I'm running ECM in the range 400000-401000 (100 curves with B1=250000, B2=25000000). The computer found 3 factors for the 15 exponents it tested. There are still 3 exponents to finish this range. So the success rate in this case was 20%.

2010-11-13, 10:33	#87
lorgix Sep 2010 Scandinavia 3×5×41 Posts	Hi everybody, I figured this thread might be a good place to ask; Does anyone know what the status is on missed small factors? Factors that should have been found by TF already. I know there are threads about this, but I haven't seen one that appears to be up to date.

2010-11-14, 22:25	#88
garo Aug 2002 Termonfeckin, IE 5320₈ Posts	In what range? It is generally agreed that it is not worth the effort to refactor all ranges to catch the few factors that may have been missed due to hardware error. A couple of years ago I did a stats analysis and found some ranges that had fewer factors than expected and these were subsequently refactored.