zoukankan      html  css  js  c++  java
  • ZOJ 2562 More Divisors

    More Divisors

    Time Limit: 2000ms
    Memory Limit: 65536KB
    This problem will be judged on ZJU. Original ID: 2562
    64-bit integer IO format: %lld      Java class name: Main

    Everybody knows that we use decimal notation, i.e. the base of our notation is 10. Historians say that it is so because men have ten fingers. Maybe they are right. However, this is often not very convenient, ten has only four divisors -- 1, 2, 5 and 10. Thus, fractions like 1/3, 1/4 or 1/6 have inconvenient decimal representation. In this sense the notation with base 12, 24, or even 60 would be much more convenient.

    The main reason for it is that the number of divisors of these numbers is much greater -- 6, 8 and 12 respectively. A good quiestion is: what is the number not exceeding n that has the greatest possible number of divisors? This is the question you have to answer.

    Input:

    The input consists of several test cases, each test case contains a integer n (1 <= n <= 1016).

    Output:

    For each test case, output positive integer number that does not exceed n and has the greatest possible number of divisors in a line. If there are several such numbers, output the smallest one.

    Sample Input:
    10
    20
    100
    
    Sample Output:
    6
    12
    60
    

     

    Source

    Author

    Andrew Stankevich
     
    解题:直接贪心地搜,选完2,一定从3开始选,不会直接从2跳到4,没必要,因为即使因子数相同,也会导致目标值变大
     
     1 #include <bits/stdc++.h>
     2 using namespace std;
     3 typedef long long LL;
     4 const LL INF = 0x3f3f3f3f3f3f3f3f;
     5 const int maxn = 1010;
     6 int p[maxn],tot;
     7 bool np[maxn] = {true,true};
     8 void init(){
     9     tot = 0;
    10     for(int i = 2; i < maxn; ++i){
    11         if(!np[i]) p[tot++] = i;
    12         for(int j = 0; j < tot && p[j]*i < maxn; ++j){
    13             np[p[j]*i] = true;
    14             if(i%p[j] == 0) break;
    15         }
    16     }
    17 }
    18 
    19 LL ret,tsum,x;
    20 
    21 void dfs(int cur,LL sum,LL val){
    22     if(sum > tsum || sum == tsum && val < ret){
    23         ret = val;
    24         tsum = sum;
    25     }
    26     for(int i = 1; i < 61; ++i){
    27         val *= p[cur];
    28         if(val > x) break;
    29         dfs(cur + 1,sum*(i + 1),val);
    30     }
    31 }
    32 int main(){
    33     init();
    34     while(~scanf("%lld",&x)){
    35         ret = INF;
    36         tsum = 1;
    37         dfs(0,1,1);
    38         printf("%lld
    ",ret);
    39     }
    40     return 0;
    41 }
    View Code
  • 相关阅读:
    java中的socket编程有关printStream的println方法和write方法
    json在php中的用法
    js的数组处理函数splice
    将博客搬至CSDN
    mapreduce导出MSSQL的数据到HDFS
    基于信息熵的无字典分词算法
    搜索引擎手记(三)之网页的去重
    算法之常用的距离和相似度度量
    搜索引擎手记(二)之爬虫的开发
    搜索引擎手记(一)之引擎工作的开始
  • 原文地址:https://www.cnblogs.com/crackpotisback/p/4860419.html
Copyright © 2011-2022 走看看