Thread for posting tiny primes

[CRGreathouse]

Originally Posted by 3.14159

I'm practically a comp. illiterate. I can't help it!

That's probably why it's so insulting that you feel the need and ability to write things like this:

Originally Posted by 3.14159

Only executables can be run as apps outside of the Internet. If it's Internet-based, JavaScript rules.

[3.14159]

Originally Posted by Charles

That's probably why it's so insulting that you feel the need and ability to write things like this:

Insulting to you, lack of knowledge to me.

Take your high horse elsewhere.

[CRGreathouse]

Originally Posted by 3.14159

The app time for the p10 was <1 second.

...And? A 3 GHz processor does 3e9 cycles per second. But I was speaking to your 2.5 second time:

Originally Posted by 3.14159

I would hardly call trial division to 10⁶ in about 2.5 seconds slow.

Originally Posted by 3.14159

Caught, red-handed!

I have no idea what you're talking about. You're (apparently) choosing not to believe my claim that I'm a computer programmer. That's fine with me, I'm not about to give you evidence to show that I am. I don't need to defend myself.

It's bizzarre (to me) that you would even question that, given our relative programming abilities.

[CRGreathouse]

Originally Posted by 3.14159

Take your high horse elsewhere.

No thanks.

[3.14159]

Originally Posted by Charles

I have no idea what you're talking about. You're (apparently) choosing not to believe my claim that I'm a computer programmer. That's fine with me, I'm not about to give you evidence to show that I am. I don't need to defend myself.

It's bizzarre (to me) that you would even question that, given our relative programming abilities.

Fine. You win.

And, one does not need to be a programmer to know something about programming. Anyone with a keen interest in programming does not necessarily have an occupation as a programmer.

[CRGreathouse]

Originally Posted by 3.14159

Whatever you say. I base it on the odds. And, one does not need to be a programmer to know something about programming. Anyone with a keen interest in programming does not necessarily have an occupation as a programmer.

Ah yes, because I have so much reason to lie about a thing like that.

How about you? What's your occuparion?

[3.14159]

Originally Posted by Charles

Ah yes, because I have so much reason to lie about a thing like that.

It would have been as simple as showing me your website, and I would let it be.

Originally Posted by Charles

How about you? What's your occuparion?

Yes, because I was accusing you of being jobless.

Anyway:

My comp measures up at 2.4GHz, if I remember correctly.

(It alternates between 2.37 and 2.4, too lazy to get an exact measure.)

[CRGreathouse]

Originally Posted by 3.14159

It would have been as simple as showing me your website, and I would let it be.

How would that prove anything? Anyone can have a website that claims that they are a computer programmer.

Originally Posted by 3.14159

Yes, because I was accusing you of being jobless.

I don't deny the possibility that while I claim to be a programmer, I am in fact a hobo deceiving you (unsuccessfully?) into believing that I'm a computer programmer.

But this doesn't address my question. What do you do? If you're going to pester me about what I do, disbelieve my answer, and accuse me of being jobless,* I think I can at least ask you what you do, even if you're not willing to "prove it".

* Why in the world wouldn't I tell you if I was jobless? It would be to my advantage to say, "I'm an out-of-work programmer" if I was out of work, in hopes that eventually I would say this to someone who might hire me (or who knew someone who would hire me).

Originally Posted by 3.14159

My comp measures up at 2.4GHz, if I remember correctly.

(It alternates between 2.37 and 2.4, too lazy to get an exact measure.)

OK. So if trial-dividing up to a million takes 2.5 seconds, that's 12,000 to 13,000 cycles per prime. This is about 200 times less work per cycle than this computer. But this computer uses an old architechture, meaning that it's able to do less work per cycle than newer computers. If yours is a Core or a Core 2, the program ("applet") is actually 300 times worse than my Pari script.

This is surprising, since this is not really the kind of work that Pari is good at.

[3.14159]

Originally Posted by Charles

I don't deny the possibility that while I claim to be a programmer, I am in fact a hobo deceiving you (unsuccessfully?) into believing that I'm a computer programmer.

Nah, that's out of the question. My skepticism isn't that harsh.

Originally Posted by Charles

OK. So if trial-dividing up to a million takes 2.5 seconds, that's 12,000 to 13,000 cycles per prime. This is about 200 times less work per cycle than this computer.But this computer uses an old architechture, meaning that it's able to do less work per cycle than newer computers. If yours is a Core or a Core 2, the program ("applet") is actually 300 times worse than my Pari script.

Is your PARI script:

trialdivide(n) = {
forprime(p=2,(<insert primelimit here>),
if(n%p==0, return(p))
);
}
??

And, yes, the processor is a Core 2.

Originally Posted by Charles

But this doesn't address my question. What do you do? If you're going to pester me about what I do, disbelieve my answer, and accuse me of being jobless,* I think I can at least ask you what you do, even if you're not willing to "prove it".

I'm the average high schooler.

[3.14159]

I wonder what the largest primes for each category are:

For the designated categories:
1. Proths: 22147 * 2²⁵⁶⁷²⁰ + 1 (3.14159) (77285 digits)
2. Generalized Proths: 4*17¹⁷⁸⁴³⁸ + 1 (Batalov) (219561 digits)
3. Factorial-based Proths: 2634 * 2480!² + 1 (3.14159) (14690 digits)
4. Primorial-based Proths: 703 * p(125)⁶⁶ + 1 (3.14159) (19104 digits)
5. Prime-based Proths: 28380*1999⁸⁵⁶⁰ + 1 (3.14159) (28260 digits)
6. Primorial: 466*7297#+1 (3.14159) (3124 digits)
7. Factorial: 1364 * 4200! + 1 (3.14159) (13399 digits)
8. Generalized Cullen-Woodall: 4034 * 1500¹⁵³⁴ + 1 (3.14159) (12816 digits)
9. Factorial-based Cullen-Woodall: (No submissions)
10. Primorial-based Cullen-Woodall: (No submissions)
11. Prime-based Cullen-Woodall: (No submissions)
12. k-b-b's: (No submissions)
13. Factorial-based k-b-b's: (No submissions)
14. Primorial-based k-b-b's: (No submissions)
15. Prime-based k-b-b's: 240790*4861⁴⁸⁶¹ + 1 (3.14159) (17927 digits)
16. Number, square, and fourth:

1546750398064958524633425292455441988676752527730835687139862311217981986247572619886186890734016949799686352443760408934653066300968034776190705901570570913201114184380563243205276690363368824249952639567750758594330067937015662345505149388718977581056000000000000000000000000000000000000000000000001

(3.14159) (301 digits)
17. Special Cofactor: (No submissions)
18. General Cofactor: (64177156568540609146951155341 * 6⁷⁶⁰⁰ + 1)/500312594495474956967917511 (3.14159) (5917 digits)
19. General arithmetic progressions: 2304*2⁷⁹⁸⁰ + 19 (3.14159) (2406 digits)
20. Obsolete-tech proven primes: 570331 * 2⁹³⁵⁶⁰ + 1 is prime! (3.14159) (28171 digits) (Proth.exe)

[mdettweiler]

A submission for #2 on the -1 side:

2778*211^47085-1 is prime!

Again, found in an effort to prove the Riesel base 211 conjecture as part of the Conjectures 'R Us project. Found and proven by PFGW via PRPnet.