[QUOTE=science_man_88;252333]because I want to try and make one that works without that so it's not limited by primelimit.[/QUOTE]

There are very few cases when you'd actually want to do that. In almost all cases the better approach would be to calculate more primes and keep them on hand -- and you can already do this with
default(primelimit, 10^9)
changing the number as appropriate.

But if you insist, here's a really, really, really bad way:
[code]primeN(n)={
my(p=1);
while(n,
if(isprime(p++), n--)
);
p
};[/code]

But please don't do this! It takes 10.9 seconds to do primeN(10^6), compared to 1.5 milliseconds for prime(10^6) -- seven thousand times slower. primeN(10^7) takes several minutes while prime(10^7) gives an answer in ~10 ms.

[QUOTE=CRGreathouse;252391]There are very few cases when you'd actually want to do that. In almost all cases the better approach would be to calculate more primes and keep them on hand -- and you can already do this with
default(primelimit, 10^9)
changing the number as appropriate.

But if you insist, here's a really, really, really bad way:
[code]primeN(n)={
my(p=1);
while(n,
if(isprime(p++), n--)
);
p
};[/code]

But please don't do this! It takes 10.9 seconds to do primeN(10^6), compared to 1.5 milliseconds for prime(10^6) -- seven thousand times slower. primeN(10^7) takes several minutes while prime(10^7) gives an answer in ~10 ms.[/QUOTE]

fine what ever. want what my minimumprime (x) does ? or have you figured it out ? all it does is x-2 to find out how many to go into the [TEX]6n\pm1[/TEX] figures which one it's on and does a calucation based on where it lands to figure out a n value than it returns [TEX]6n\pm1[/TEX]

Yes, and it's clear that that won't give the right answer because it's linear and the correct figure is about n log n.

This may be of interest to you:
[url]http://primes.utm.edu/nthprime/[/url]

[QUOTE=CRGreathouse;252414]Yes, and it's clear that that won't give the right answer because it's linear and the correct figure is about n log n.

This may be of interest to you:
[url]http://primes.utm.edu/nthprime/[/url][/QUOTE]

Since it's built to give the lowest possible value it could be I know it's not correct. I'm just trying to figure a easy way to correct for it. The first time it's wrong is x = 10 last I checked.

[QUOTE=science_man_88;252416]Since it's built to give the lowest possible value it could be I know it's not correct. I'm just trying to figure a easy way to correct for it. The first time it's wrong is x = 10 last I checked.[/QUOTE]

I gave pretty much the easiest way to do it without using the built-in prime list. There are decently fast methods that use the prime list but avoid the prime() command itself, if you're interested. For example,
[code]primeN(n)={
my(v=primes(n));
v[#v]
};[/code]
and
[code]primeN(n)={
if(n<6, return([2,3,5,7,11][n]));
my(l=log(n), lower=ceil(n*l+n*(log(l)-1)), upper=floor(n*l+n*log(l)), t);
while(upper - lower > 1,
if(primepi(t = (lower + upper) \ 2) < n, lower = t, upper = t)
);
if(primepi(lower) == n, lower, upper)
};[/code]

These might be a hundred times slower rather than 10,000 times slower.

By the way, I think that re-implementing basic features of a language is a good way to learn. Working through Prime Numbers: A Computational Perspective really helped me to understand how the various algorithms fit together. So I don't think there's anything wrong with doing this -- I just want you to code it, learn, and then use the built-in functions rather than your own. :smile:

[QUOTE=CRGreathouse;252426]By the way, I think that re-implementing basic features of a language is a good way to learn. Working through Prime Numbers: A Computational Perspective really helped me to understand how the various algorithms fit together. So I don't think there's anything wrong with doing this -- I just want you to code it, learn, and then use the built-in functions rather than your own. :smile:[/QUOTE]

yeah bad part is I'd be lucky to program all the things my pocket calculator can do. I have an idea for one thing I haven't done before mainly because it's like the base converter.

[QUOTE=science_man_88;252634]yeah bad part is I'd be lucky to program all the things my pocket calculator can do. I have an idea for one thing I haven't done before mainly because it's like the base converter.[/QUOTE]

Actually that's a good project because PARI doesn't know how to do that on its own. I mean, it inputs and displays numbers in decimal while storing and calculating in binary, and it can return a number in binary (as a vector) if requested, but it doesn't have generic base-conversion routines.

[QUOTE=CRGreathouse;252671]Actually that's a good project because PARI doesn't know how to do that on its own. I mean, it inputs and displays numbers in decimal while storing and calculating in binary, and it can return a number in binary (as a vector) if requested, but it doesn't have generic base-conversion routines.[/QUOTE]

we made the base converter last I checked, a list of what my calculator can do:

[QUOTE]list of scientific constant (not sure what all of them are)
unit conversion(40 in total, but I think I could cut that to 20)
powers (not so hard in PARI)
base 10 and natural logarithms (not so hard)
some SI prefixes ( mega,giga,tera,micro,milli,kilo,fempto,pico,nano)
combinations and permutations
trig functions
xth roots ( pretty much done)
convert answer to fraction if possible.
insert function.
9 variable storage.
Pol( and Rec( (haven't used them but I'm pretty sure I know what they do).
integrals ?
"random" numbers
rounding ( not hard)
factorial
Pi
a+bi ( seen the equation before don't know what it is)
percentage
[TEX]Re\leftrightarrow lm[/TEX]
degrees rad and grad
base conversion( DEC,HEX,BIN,OCT)[/QUOTE]

modes are : [QUOTE]COMP,CMPLX,SD,REG,BASE,EQN,MAT,VCT,Deg,Rad,Gra,Fix,Sci,Norm,Disp[/QUOTE]

Most of those are of course already in GP. If you want to implement some of the others, give it a shot and I'll take a look.

[QUOTE=CRGreathouse;252703]Most of those are of course already in GP. If you want to implement some of the others, give it a shot and I'll take a look.[/QUOTE]

the conversion among units is basically a base conversion with each digit in the number as a different conversion base.

I'm not completely sure which others aren't.