|
|
马上注册，结交更多好友，享用更多功能，让你轻松玩转社区。
您需要 登录 才可以下载或查看，没有账号？注册
x
The random number generators in standard function libraries are not always of the best
quality. With today’s fast computers it is possible to make larger computer simulations than
2
what has been common previously. Larger "Monte Carlo" simulations require better random
number generators. Furthermore, today’s multi-kernel microprocessors require random
number generators with support for multithreading. The present random number generator
libraries are designed for the purpose of meeting these high demands. The advantages of
these packages are:
• Very good randomness
• Very long cycle length
• High resolution
• Support for multiple threads and multiple streams
• Very fast and efficient
• Allow seeds of any length
• Includes Mersenne Twister, Mother-Of-All generator, SFMT generator and
combinations of these
• Discrete uniform distribution over arbitrary interval is exact where other
implementations have rounding errors
• Continuous distributions supported: Uniform and normal
• Discrete distributions supported: Uniform, Poisson, binomial, hypergeometric and
various noncentral hypergeometric distributions
• Open source
• Support for Windows, Linux, BSD, Mac, etc.
Source code:
randomc.rar
(240.32 KB, 下载次数: 60 )
|
|
/1 
eetop公众号
创芯大讲堂
创芯人才网
发表于 2012-12-13 11:19:11
