This question already has an answer here:
这个问题在这里已有答案：
- Arc4random modulo biased 1 answer
- Arc4random modulo偏向于1个答案
So I've seen old posts about the difference between random and arc4random in Obj. C, and I've seen answers to this online but I didn't really understand the answers so I was hoping someone here could explain it in an easier to understand manner.
所以我看过关于Obj中随机和arc4random之间差异的旧帖子。 C，我在网上看到过这个问题的答案，但我并不真正理解答案，所以我希望有人能以更容易理解的方式解释它。
What is the difference between using arc4random and arc4random_uniform to generate random numbers?
使用arc4random和arc4random_uniform生成随机数有什么区别？
1 个解决方案
#1
72
arc4random returns an integer between 0 and (2^32)-1 while arc4random_uniform returns an integer between 0 and the upper bound you pass it.
arc4random返回0到（2 ^ 32）-1之间的整数，而arc4random_uniform返回0和你传递的上限之间的整数。
From man 3 arc4random:
来自man 3 arc4random：
arc4random_uniform() will return a uniformly distributed random number less than upper_bound. arc4random_uniform() is recommended over constructions like ``arc4random() % upper_bound'' as it avoids "modulo bias" when the upper bound is not a power of two.
arc4random_uniform（）将返回小于upper_bound的均匀分布的随机数。 arc4random_uniform（）建议使用像``arc4random（）％upper_bound''这样的结构，因为当上限不是2的幂时，它避免了“模偏差”。
For example if you want an integer between 0 and 4 you could use
例如，如果你想要一个0到4之间的整数，你可以使用
arc4random() % 5
or
要么
arc4random_uniform(5)
Using the modulus operator in this case introduces modulo bias, so it's better to use arc4random_uniform.
在这种情况下使用模数运算符会引入模偏差，因此最好使用arc4random_uniform。
To understand modulo bias assume that arc4random had a much smaller range. Instead of 0 to (2^32) -1, it was 0 to (2^4) -1. If you perform % 5 on each number in that range you will get 0 four times, and 1, 2, 3 and 4 three times each making 0 more likely to occur. This difference becomes less significant when the range is much larger, but it's still better to avoid using modulus.
为了理解模偏差，假设arc4random具有更小的范围。而不是0到（2 ^ 32）-1，它是0到（2 ^ 4）-1。如果你对该范围内的每个数字执行％5，你将得到0四次，并且每次得到1,2,3和4次，使得0更有可能发生。当范围大得多时，这种差异变得不那么显着，但是避免使用模数仍然更好。
#1
72
arc4random returns an integer between 0 and (2^32)-1 while arc4random_uniform returns an integer between 0 and the upper bound you pass it.
arc4random返回0到（2 ^ 32）-1之间的整数，而arc4random_uniform返回0和你传递的上限之间的整数。
From man 3 arc4random:
来自man 3 arc4random：
arc4random_uniform() will return a uniformly distributed random number less than upper_bound. arc4random_uniform() is recommended over constructions like ``arc4random() % upper_bound'' as it avoids "modulo bias" when the upper bound is not a power of two.
arc4random_uniform（）将返回小于upper_bound的均匀分布的随机数。 arc4random_uniform（）建议使用像``arc4random（）％upper_bound''这样的结构，因为当上限不是2的幂时，它避免了“模偏差”。
For example if you want an integer between 0 and 4 you could use
例如，如果你想要一个0到4之间的整数，你可以使用
arc4random() % 5
or
要么
arc4random_uniform(5)
Using the modulus operator in this case introduces modulo bias, so it's better to use arc4random_uniform.
在这种情况下使用模数运算符会引入模偏差，因此最好使用arc4random_uniform。
To understand modulo bias assume that arc4random had a much smaller range. Instead of 0 to (2^32) -1, it was 0 to (2^4) -1. If you perform % 5 on each number in that range you will get 0 four times, and 1, 2, 3 and 4 three times each making 0 more likely to occur. This difference becomes less significant when the range is much larger, but it's still better to avoid using modulus.
为了理解模偏差，假设arc4random具有更小的范围。而不是0到（2 ^ 32）-1，它是0到（2 ^ 4）-1。如果你对该范围内的每个数字执行％5，你将得到0四次，并且每次得到1,2,3和4次，使得0更有可能发生。当范围大得多时，这种差异变得不那么显着，但是避免使用模数仍然更好。