Re^2: Random generator fail

[Anonymous Monk]

in reply to Re: Random generator fail
in thread Random generator fail

Ok I have better understanding now, Bill your solutions works great!

@MidLifeXis What is rational for rand to ignore $_. Appearently you pointing out that, solves my problem by using

perl -wle "use 5.10.0;  foreach (1..100) { say rand($_);};"

1.94992065429688
56.7288208007813
52.1597290039063
18.6995849609375
6.6961669921875
68.6776733398438
65.6177978515625
62.72314453125
52.1324157714844
...

But I think using additional line  { say rand($_); is completely un-neccessary. Thank you.

[Anonymous Monk]

forgot to add this:

perl -wle "use 5.10.0;   say int foreach (1..100) ;"

Works as rand() should have.

[Anonymous Monk]

This one is better

perl -wle "use 5.10.0;   say abs foreach (-5..5) ;"
5
4
3
2
1
0
1
2
3
4
5

[NodeReaper]

This node was taken out by the NodeReaper on Dec 07, 2012 at 11:40 UTC
Reason: [Athanasius]: Reap: Duplicate of Re^4: Random generator fail.

You may view the original node and the consideration vote tally.

[MidLifeXis]

Works as rand() should have.

Either your assumptions are wrong, or the documentation for rand does not match the implementation. Which do you think it is?

[MidLifeXis]

Let's step through this:

foreach (1..100) {
    say rand($_);
}

When $_ is equal to ...
• ... 1, rand($_) cannot be larger than 1.
• ... 2, rand($_) cannot be larger than 2.
• ... 3, rand($_) cannot be larger than 3.
• ... 4, rand($_) cannot be larger than 4.
• ... 5, rand($_) cannot be larger than 5.
• ... 6, rand($_) cannot be larger than 6.

Look at one of my earlier posts on this thread. The documentation for rand states that 0 <= rand($x) < $x if $x > 0. To accomplish what you are trying to do would require that you do this:

say 1 + rand(99 - 1) for (1..100)

In my previous node, MAX = 99 and MIN = 1. The 1..100 is the number of iterations you wish to use. If you use rand($_), your results will be biased toward the lower end of the counter range.

[Anonymous Monk]

I agree with bias at lower end. Great, that is pretty much what I was looking for.

Thank you

[BillKSmith]

My rand 98 generates random numbers in the range (0<= number < 98). Adding one to every random number changes the range to (1<=number<99). Using rand $_ would generate a very different sequence of numbers. In fact, the sample output that you show cannot be the first few numbers as your dots suggest. It is possible that they are the last few numbers.