Better prime finding methods? i.e with a better big-O performance?

charting.R

require("ggplot2")
# data load
n_squared = read.csv("nth.csv", header=T)
n_modulo = read.csv("nth_modulo.csv", header=T)

n_modulo$nth_log <- log(n_modulo$nth)
n_modulo$time_log <- log(n_modulo$time)

n_squared$time_log <- log(n_squared$time)
n_squared$nth_log <- log(n_squared$nth_prime)

png("log_performance.png",800,600)
ggplot(n_squared, aes(nth_log, time_log, colour="methods")) + geom_line(aes(colour="scan")) + geom_line(data=n_modulo, aes(nth_log, time_log,colour="modulo")) + ggtitle("Performance of prime finding algorithms (log)")
dev.off()

png("raw_performance.png",800,600)
ggplot(n_squared, aes( colour="methods")) + geom_line(aes(nth_prime, time, colour="scan")) + geom_line(data=n_modulo, aes(nth_prime, time,colour="modulo")) + ggtitle("Performance of prime finding algorithms")
dev.off()

png("logn_performance.png",800,600)
ggplot(n_squared, aes( colour="methods")) + geom_line(aes(nth_log, time, colour="scan")) + geom_line(data=n_modulo, aes(nth_log, time,colour="modulo")) + ggtitle("Performance of prime finding algorithms (logn)")
dev.off()

prime-hunt.py

import time
import math
from sets import Set

def is_it_prime(n):
    largest_candidate = int(math.ceil(n/2))
    candidates =  range(2, largest_candidate + 1)
    #print "%d => %s" % (n, candidates)
    for i in candidates:
        if not n % i:
            #print "Not prime: %d => %d" % (n, i)
            return False
    return True

def nth_prime(n):
    primes = [2]
    current_number = 3
    while len(primes) < n:
        if is_it_prime(current_number):
            primes.append(current_number)
        current_number += 2
    return primes[n-1]

def nth_prime_modulo(n):
    primes = [2]
    current_number = 3
    while len(primes) < n:
        is_prime = 1
        for prime in primes:
            if not current_number % prime:
                is_prime = 0
                break
        if is_prime:
            primes.append(current_number)
        current_number += 2
    return primes[n-1]

def generate_nth_prime_modulo(n):
    start = time.time()
    nth_prime_modulo(n)
    total_time = time.time() - start
    #print "%f for %d" % (total_time, n)
    return total_time

def generate_nth_prime(n):
    start = time.time()
    nth_prime(n)
    total_time = time.time() - start
    return total_time

n = 3000

fp = open("data/nth.csv", "w")
fp.write("nth_prime,time\n")
for i in range(0, n, 100):
    fp.write("%d,%f\n" % (i, generate_nth_prime(i)))
fp.close()

fp = open("data/nth_modulo.csv", "w")
fp.write("nth_prime,time\n")
for i in range(0, n, 100):
    fp.write("%d,%f\n" % (i, generate_nth_prime_modulo(i)))
fp.close()