20060415, 19:42  #23  
Sep 2005
UGent
2^{2}·3·5 Posts 
Quote:


20060415, 20:29  #24 
Sep 2005
177_{8} Posts 
Yes, I'm serious  I estimate (assuming all goes well  which is the point of the test) about 1 year to find the 20digit factor of (M+2)607. Interested?
About 8000 cpu years to find 40digit factor of similar numbers, about 64000000 cpu years to find 80digit factors, etc... [so I know which algorithm _I_ would prefer for big factors] This is using my (alreadywritten) Java code  if I were to rewrite it in C might be faster?? J 
20060415, 20:38  #25 
"Nancy"
Aug 2002
Alexandria
2,467 Posts 
Run it yourself. I don't mind if you waste your own cpu time, but don't waste others' by luring them into pointless endeavours with baseless claims.
Alex 
20060415, 22:21  #26 
Sep 2005
127 Posts 
It's not a baseless claim (I've already had promising results).
Lay off of my potential helpers! :) J 
20060415, 23:50  #27  
Nov 2003
1D22_{16} Posts 
Quote:
It is strictly an exponential time algorithm and is actually worse than trial division. It is pointless. 

20060416, 02:24  #28 
Sep 2005
127 Posts 
Trial division requires testing values of the 'base' up to the size of the factor. The results I'm already getting show that only a small percentage of the size of the factor is needed from the 'base' to find the factor. What's more this beneficial effect increases with size of factor.
Thinking theoretically, it is clear to me at least, that in the limit, the size of base needed for any size factor will tend to a constant (sic). How can this be worse than trial division!!!. Couple this fact with the polynomial/logarithmic time needed for each step (as I've already said, using Russian Peasant) and you get a polynomial/logarithmic algorithm. If you don't believe me  try it!! There's working programs there... 
20060416, 19:18  #29  
∂^{2}ω=0
Sep 2002
República de California
2^{4}×7×101 Posts 
Quote:
Quote:
Quote:


20060816, 04:57  #30 
Sep 2005
127_{10} Posts 
You may already be aware of the Mersenneplustwo project  but if not I'd like to bring it to your attention, pls...
http://bearnol.isageek.com/Mersenn...neplustwo.html thanks, J Last fiddled with by akruppa on 20060816 at 07:57 Reason: Post moved to "Tester(s) required" thread 
Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Help required  ET_  Programming  19  20170716 11:20 
Triple check required  houding  PrimeNet  14  20151221 09:34 
New hardware  help required  PageFault  Software  2  20140414 15:23 
SBprp  a better PRP tester  Ken_g6  Prime Sierpinski Project  10  20050104 14:36 
Sound Card Tester??  dave_0273  Hardware  2  20040619 15:34 