Too many for him or for his opponent?
Did he complain that they came at the wrong time as well?
(One 4-4 after the first few rolls is usually too many.)
Kidding aside
It is nice that he is at least specific.
You can easily test that now by seeing what a the history of rolls.
If someone has a bit of statistics knowledge, it is possible to see if
the player's experience is just bad luck.
A good book of quality control and sampling will give you a hint about
how big a sample you need and how to determine if the counts are inside
or outside the expected norm.
The gamesgrid reference looks interesting.
A small, not too scientific, suggestion to increase the randomness.
Take the least significant part of the current time (last 2 digits of
the milliseconds) and loop that many times through a tight loop calling
Math.Random and keep the last one.
The code is small and quick and it adds an element of real
randomness(current time) to the algorithm.
If the game is for money and the client side code can be hacked for
profit, you need to move more of the game engine onto the server and
only present results and table state on the client side.
Dice rolls can not be generated by the client in that case. Your client
could be easily replaced by a version of your client that lets the
player chose his own rolls which makes Backgammon not fun for the
opponent but a really lucrative way to pass an afternoon for the guy
choosing his rolls. You probably only need to chose 3 or 4 rolls during
a game to win almost every time. (An interesting test on a real board
for someone with time to kill and a friend who does not mind losing.)
Ron
Jayson K Hanes wrote:
You should be using random(ness) from the server side of things, and not
relying on flash (especially not clientside)...
Otherwise.. on a pay site.. its not too unreasonable to bet someone
sooner or later will decompile your client swf (the game) and modify it
to improve their odds of winning more often.
-Jayson
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:flashcoders-
[EMAIL PROTECTED] On Behalf Of Dimitrios Bendilas
Sent: Thursday, November 17, 2005 1:13 PM
To: Flashcoders mailing list
Subject: Re: [Flashcoders] Math.random()
Hey Ron,
Thanks for the input.
Actually, what that member said is that there are too many 4-4
among the doubles.
But I haven't seen the data, so I don't know.
Regards,
Dimitrios
_______________________________________________
Flashcoders mailing list
Flashcoders@chattyfig.figleaf.com
http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
_______________________________________________
Flashcoders mailing list
Flashcoders@chattyfig.figleaf.com
http://chattyfig.figleaf.com/mailman/listinfo/flashcoders
