[
https://issues.apache.org/jira/browse/MATH-1300?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15065484#comment-15065484
]
Rostislav Krasny commented on MATH-1300:
----------------------------------------
According to the code comments both the CM and the Mantissa implementations of
MersenneTwister are based on the same code "developed by Makoto Matsumoto and
Takuji Nishimura during 1996-1997". The main difference between them is that
Mantissa implementation extends the standard java.util.Random and uses JDK's
nextBytes() method while the CM implementation extends its own
BitsStreamGenerator with its own nextBytes() method implementation. So the
reference behavior for the nextBytes() method must be the java.util.Random
since it is a part of the JDK and so it is the standard.
BTW I looked at my code again and found that & 0xff operations are not needed.
Narrowing Primitive Conversion just discards all unneded bits above the byte,
so we don't need to do it by ourself before int to byte type casting.
http://docs.oracle.com/javase/specs/jls/se7/html/jls-5.html#jls-5.1.3
So I propose this new code of the BitsStreamGenerator#nextBytes() method with
even a little bit better performance
{code:java}
@Override
public void nextBytes(byte[] bytes) {
int random;
int index = 0;
for (int max = bytes.length & 0x7ffffffc; index < max;) {
random = next(32);
bytes[index++] = (byte) random;
bytes[index++] = (byte) (random >>> 8);
bytes[index++] = (byte) (random >>> 16);
bytes[index++] = (byte) (random >>> 24);
}
if (index < bytes.length) {
random = next(32);
while (true) {
bytes[index++] = (byte) random;
if (index < bytes.length) {
random >>>= 8;
} else {
break;
}
}
}
}
{code}
> BitsStreamGenerator#nextBytes(byte[]) is wrong
> ----------------------------------------------
>
> Key: MATH-1300
> URL: https://issues.apache.org/jira/browse/MATH-1300
> Project: Commons Math
> Issue Type: Bug
> Affects Versions: 3.5
> Reporter: Rostislav Krasny
> Attachments: MersenneTwister2.java, TestMersenneTwister.java
>
>
> Sequential calls to the BitsStreamGenerator#nextBytes(byte[]) must generate
> the same sequence of bytes, no matter by chunks of which size it was divided.
> This is also how java.util.Random#nextBytes(byte[]) works.
> When nextBytes(byte[]) is called with a bytes array of length multiple of 4
> it makes one unneeded call to next(int) method. This is wrong and produces an
> inconsistent behavior of classes like MersenneTwister.
> I made a new implementation of the BitsStreamGenerator#nextBytes(byte[]) see
> attached code.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
