| WIKI | FAQ | Tech FAQ | |
 http://csua.com/feed/ | |
| ||||||
| 5/23 |
| 2003/2/5-6 [Computer/SW/Languages/C_Cplusplus, Computer/SW/Languages/Java] UID:27308 Activity:high |
2/5 In C, the typical way to generate random numbers in the range [0, N)
is (int) ((double) rand() / ((double) RAND_MAX + 1) * N). What
guarantee is there that RAND_MAX + 1 does not overflow?
\_ RAND_MAX is 32767.
\_ RAND_MAX is guaranteed to be at LEAST 32767.
\_ aren't your parentheses misplaced wrt N?
\_ no. You generate a floating value in the range [0, 1), multiply
by N, and cast to int.
\_ right. as written, it's generating [0,1) and dividing by N,
n'est-ce pas?
\_ Note that you're casting RAND_MAX to a double. Standard IEEE754 has
54 bits of significand in a double. That's typically more than
anything a 32-bit machine has to offer. However there is no
guarantee against overflow.
By the way, rand() is a crappy generator. The method above leads to
non-uniform probability distribution (some values will be twice as
likely than the rest), etc.
\_ so can you change the seed each time, based on prev #
and other factors to make it distribute more evenly?
\_ it doesn't matter what the seed is. most implementations
of rand() just suck.
\_ understood. care to suggest a better way of
generating random numbers?
\_ digitize johnson noise and take the least
signifigant bit? i'll bet you could make
a box to do this for about a dollar in electronics.
I think there are chips that do this, or use
shot noise which is also white.
-naive physicist
\_ Actually, the best way would be to
to take some radioactive material
\_ well even just C's "random" function is better than
"rand". look at "man 3 rand". i would try the /dev
random device but then that's not very portable
and measure either the amount of
decay or the interval between two
decay events and use that as a
basis for your random numbers.
\_ read from /dev/(u)random
\_ http://www.boost.org/libs/random/index.html
\_ well even just C's "random" function is better
than "rand". look at "man 3 rand". i would try
the /dev random device but then that's not very
portable
\_ It works on MacOS X, Solaris (8+),
L1NUX, *BSD, how much more portability
do you want?
\_ (int) (drand48() * N); |
| 5/23 |
|
| www.boost.org/libs/random/index.html In order to map the interface of the generators and distribution functions to other concepts, some 33 decorators are available. Rationale The methods for generating and evaluating deterministic and non-deterministic random numbers differ radically. Furthermore, due to the inherent deterministic design of present-day computers, it is often difficult to implement non-deterministic random number generation facilities. Thus, the random number library is split into separate header files, mirroring the two different application domains. History and Acknowledgements In November 1999, Jeet Sukumaran proposed a framework based on virtual functions, and later sketched a template-based approach. Ed Brey pointed out that Microsoft Visual C++ does not support in-class member initializations and suggested the enum workaround. The first public release of this random number library materialized in March 2000 after extensive discussions on the boost mailing list. Many thanks to Beman Dawes for his original min_rand class, portability fixes, documentation suggestions, and general guidance. Harry Erwin sent a header file which provided additional insight into the requirements. Ed Brey and Beman Dawes wanted an iterator-like interface. Beman Dawes managed the formal review, during which Matthias Troyer, Csaba Szepesvari, and Thomas Holenstein gave detailed comments. The reviewed version became an official part of boost on 17 June 2000. Gary Powell contributed suggestions for code cleanliness. Dave Abrahams and Howard Hinnant suggested to move the basic generator templates from namespace boost::detail to boost::random. Ed Brey asked to remove superfluous warnings and helped with uint64_t handling. Matthias Troyer contributed a lagged Fibonacci generator. Michael Stevens found a bug in the copy semantics of normal_distribution and suggested documentation improvements. |