20100523, 11:35  #67 
Mar 2010
On front of my laptop
7×17 Posts 
A bug?
When I enter a factor during a RabinMiller PRP test, then the applet will search for perfect power for a long time, and it will say the factorization is complete.
I tried Phi(9816,10)*Phi(24,10). The applet found the factor 9817 easily, and I entered 1059411433 as a factor(the known factor), then the applet searched for perfect power for a long time, and it said the factorization is complete. It says that Phi(9816,10)*Phi(24,10) = 9817*1059411433*(a large prime cofactor). But that cofactor is certainly not a prime. It has a prime factor: Phi(24,10) = 99990001. It is certainly a bug. 
20100820, 16:49  #68 
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
5,857 Posts 
I think I have found a bug in the applet. When I enter x=447213595500007;x=x1;i21;(x*x200000000000034400000000000999) in the box at the bottom it produces some negative numbers as well as positive numbers. The positive numbers are factored correctly but the negative numbers aren't. I realize that factorizing negative numbers wasn't really your plan but it would be helpful sometimes if it was fixed although a factor of 1 would need to be added. When I found this I was searching for a suspected bug in my quadratic sieve program.

20100820, 17:43  #69 
Aug 2002
Buenos Aires, Argentina
2^{2}·337 Posts 
I'm busy with "real life" now but I will change the applet as soon as possible in order to add this request.

20100820, 20:11  #70 
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
5857_{10} Posts 

20100821, 06:45  #71  
Sep 2006
Brussels, Belgium
2^{2}·5·83 Posts 
Quote:
Jacob 

20100821, 08:27  #72  
Bamboozled!
"πΊππ·π·π"
May 2003
Down not across
2×5,323 Posts 
Quote:
Paul 

20100823, 02:25  #73  
Aug 2002
Buenos Aires, Argentina
2^{2}·337 Posts 
Quote:


20100823, 09:02  #74 
Just call me Henry
"David"
Sep 2007
Cambridge (GMT/BST)
5,857 Posts 

20100905, 17:05  #75 
Nov 2005
101 Posts 
I did some timings for different JDK or Java compilers.
Times in sec. jdk 6_21 .588 > 100% jdk 7b108 .591 > 100% jdk 6_16 .61 > 104% jrokit4.0.1 .974 > 166% jet7.2 .455 > 77% Windows XP Intel dual core T9400 2.53 Ghz 32 bit Jet is a Java compiler like gcj. I tried gcj some years ago on windows with cygwin, but had to fight to get an executable (on Windows). http://www.excelsiorusa.com/downloa...tdlevalaw.html http://dlc.sun.com.edgesuite.net/jdk...ies/index.html Last fiddled with by ThiloHarich on 20100905 at 17:05 
20100917, 12:27  #76 
Aug 2002
Buenos Aires, Argentina
2^{2}·337 Posts 
New ECM record when running the applet
Robert J DuChateau was running ECM using the applet on 24+29^81 and found the factorization:
347 x 34503928686970842350938851377044353167566697578436905273 (Curve 210438) x 2377058158103859925326651218033244145946260234462377898431463 So he found a 56digit factor using the applet (a new record). Of course this could have been found faster with SIQS or GNFS (the latter is not implemented in the applet and it will never be done). 
20100917, 13:25  #77 
Dec 2009
89 Posts 
And fastest with SNFS.

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Java applet alternative  a1call  Programming  19  20191108 22:31 
New online applet for factorization  ET_  Lone Mersenne Hunters  69  20140601 17:34 
A strange applet:  3.14159  Miscellaneous Math  7  20100601 01:29 
Faster factorization applet  alpertron  Factoring  14  20060101 04:00 
Binomial Expansion Applet  jinydu  Lounge  2  20040505 08:33 