I see only the following in chm doc.
1.0.46.15 - May 9, 2007
Fixed FileInstall causing the Random command to become non-random in compiled scripts. [thanks Velocity]
I need some clarification on the following quotes from doc.
Code: Select all
Random, , NewSeed
IIUC, the concept of Seed is gone from API from Windows 8. Windows now automatically does it.
The lowest allowed value is -2147483648
the largest number will be 2147483647
Why not a higher range?
If either Min or Max contains a decimal point, the end result will be a floating point number in the format set by SetFormat.
SetFormat is deprecated. If Random was function, then I would use my RoundT() to control the precision.
Without SetFormat, I'm stuck at max precision of 0.6. Is there any other way, that I am not seeing?
occasionally a result can be slightly greater than the specified Max (this is caused in part by the imprecision inherent in floating point numbers).
So: If max is critical to me I have to do an additional validation with Min()?. Why can't Min() be done by AHK itself?
This function uses the Mersenne Twister random number generator, MT19937, written I need to send by Takuji Nishimura and Makoto Matsumoto, Shawn Cokus, Matthe Bellew and IsakuWada.
....
When you use this, send an email to: [email protected] with an appropriate reference to your work.
The license for all my code is PD-Self
So: If I include Random in any of my functions ( for example: RandowPW() ),
I need to send email, as well as put an note in my topic that any users of my function should do the same?
Do NOT use for CRYPTOGRAPHY without securely hashing several returned values together,
otherwise the generator state can be learned after reading 624 consecutive values.
That's very discouraging. I have my own Random function, but I don't know enough about this subject.
So, If I share my native implementation of Random I need to use a more stronger disclaimer?
How fantastic is this MT19937? Can't it be replaced? Can't we do it with Windows API?.
If there is any hesitation owing to a a breaking change, why not an additional Random function that doesn't impose such a usage condition?
Thanks.