Re: [boost] [random] seeding with arbitrary integers

Subject: Re: [boost] [random] seeding with arbitrary integers
From: Robert Ramey (ramey_at_[hidden])
Date: 2009-10-22 14:47:16

Steven Watanabe wrote:
> AMDG
>
> Robert Ramey wrote:
>> Also, I found the documentation not good enough in general.  I did
>> make it work but it wasn't as easy to use as I would have hoped.
>> Of course I could make the same criticism for many or most
>> libraries.
>>
>
> Do you have any specific issues?

Since you asked.  I think my problem is in the very first section
under the title "Boost Random Number Library".

I expected to find and example like:

normal_distribution nd<int>(mean, standard_deviation)
...
int random_number = nd();
...

Of course I realized pretty soon that I wasn't going to find that.
So then I looked at the random_demo.cpp which was
helpful and led me to look at library headers which wasn't
all that helpful.

Personally I would have found a small tutorial helpful. for example:

The random number library is composed of:

"generators" which generate random numbers and
"distributors" which transform output from a "generator"
into a particular distribution:

A small complete example (from random_demo.cpp)

// simulate rolling a die 10 times

// construct a random simple random number generator
boost::minstd_rand random_number_generator;

// construct a uniform distribution for integers 1 through 6
boost::uniform<int> uniform_distribution(1,6);

// hook the above together to create a die?
// I'm not sure what to call this "variate_generator" didn't help me at all
boost::variate_generator<boost::minstd_rand &, boost::uniform<int> & > die()

// throw the die 10 times
for(int i = 0; i < 10; ++i)
    std::out << die() << '\n'

Notice the lack of typedefs and other stuff to make things
more abstract".  I eliminated these to make the example
less ... uh abstract.

To summarize, a little bit more introduction would have helped
me get going faster.

Robert Ramey