Login [Register]
Don't have an account? Register now to chat, post, use our tools, and much more.
So, I'm finishing up the C/Linux coding and am preparing to start porting to java for android/web. My concerns are finding server space to host the trial division results so multiple people can help solve each big prime a "small" bite at a time.

Any advice?

and I need to make a wikipedia page for my theory/method. No experience there. My theory has increasingly accurate results the higher the primorial used. Haven't figured out the %accurate equation, but I'm sure it would involve diminishing prime frequency (~1/ln(N))

also, i need to get published and validated. seriously!

anyways, any input is welcome. thanks guys. hopefully the EFF and Guiness recognize these efforts someday
Wikipedia is for ideas that are already notable. Assuming you invented this algorithm, I suggest you first ask a question on Math Stack Exchange, something like "Does my method for finding large primes have any advantage over currently existing algorithms?"--describing it in detail and following the site rules of course. Then the community can suggest improvements (and possibly explain why it is not competitive with the state of the art). Much easier than creating a Wikipedia page, and much more likely to meet with success.
It's advantages are obvious, so I'll just work on crafting a Wikipedia page. it isolates 20M+ digit highly probable primes in just minutes each. I'll publish the android app in a few weeks.


thanks for the math stack exchange tip. also, I'll try to register this with oeis.org
What are the advanteges? How are they obvious?

As for wikipedia, someone who isn't the original discoverer has to write the page, and they have to use already-written sources as references. See WP:OR
thanks. I'll focus on oeis.org and the app. the advantages are the huge amount of time saved and the possibility of finding many actual primes in a run.

I'll include a properly written paper in the app.
seklorean wrote:
thanks. I'll focus on oeis.org and the app. the advantages are the huge amount of time saved and the possibility of finding many actual primes in a run.


Can you explain how it works and how exactly it performs better than the "previous best"?
I think this is pretty neat! Like iPhoenix said, we'd love to hear more descriptive details about it!
The package I uploaded does many things, and isn't complete yet. It's a special sieve, working off p#n + pn to pick a suitable range and sieve from there. Completely reliable until pn, after that it's back to trial division. the package generates a list of large probable prime candidates for further testing, if it's set up properly. lots of time is saved compared to other methods.
This is just a suite for finding large primes because you found some numbers that work well in a number sieve. There's no "novel" algorithm involved; it's just using a bunch of pre-made libraries to test large numbers for primeness.

I don't understand the purpose of this. Anyone can call libgmp functions.
That's a fair viewpoint, but the sieve is set up in a way that easily generates increasingly accurate results the larger the numbers involved. Other sieves generally don't seem to work well with BIG numbers - this one is designed specifically for them. Let's look at the sample list of probable prime candidates included:

p5M + 86028121 + 36
p5M + 86028121 + 100
p5M + 86028121 + 102
p5M + 86028121 + 138
p5M + 86028121 + 148
p5M + 86028121 + 156
p5M + 86028121 + 180
p5M + 86028121 + 202
p5M + 86028121 + 208
p5M + 86028121 + 222
p5M + 86028121 + 232
p5M + 86028121 + 268
p5M + 86028121 + 280
p5M + 86028121 + 298
p5M + 86028121 + 312
p5M + 86028121 + 330
p5M + 86028121 + 358
p5M + 86028121 + 390
p5M + 86028121 + 418
p5M + 86028121 + 442
p5M + 86028121 + 460
p5M + 86028121 + 462
p5M + 86028121 + 490


All of those are 37 million digits and change and pass trial division of the first 5 million primes. There are even 2 instances of possible binary primes. I acknowledge that there's still much to do, especially analyzing the accuracy rate for each iteration. It'll be a neat exponential growth value to discover. Many years ago a guy here named Thomas said I didn't have the math background to pull this off, and at the time he was right. It was humbling to face that truth. I've come very far since then
  
Register to Join the Conversation
Have your own thoughts to add to this or any other topic? Want to ask a question, offer a suggestion, share your own programs and projects, upload a file to the file archives, get help with calculator and computer programming, or simply chat with like-minded coders and tech and calculator enthusiasts via the site-wide AJAX SAX widget? Registration for a free Cemetech account only takes a minute.

» Go to Registration page
Page 1 of 1
» All times are GMT - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement