Re[2]: demon_summon spec_proc

From: Gary Barnett (gbarnett@polarnet.com)
Date: 01/17/97


At [Fri, 17 Jan 1997 12:24:46 -0500 (EST)]
Sammy <Samedi@cris.com> wrote:

>  
>  On Fri, 17 Jan 1997 BuckFloyd@aol.com wrote:
>  
>  - >   if (number(0,100) > number(0, 100))
>  - >   {
>  - >     if ((number(0, 100) < 50) && (number(0, 100) > 85))
>  - >     {
>  
>  Wow that really is odd.  To each his or her own, I guess, but it's a
>  little wasteful of processor time.  I think somoene who's had a statistics
>  class could break those 4 random numbers down and turn it into one.
>  
>  This is my take on it:
>  
>  - >   if (number(0,100) > number(0, 100))
>  
>  This looks like a 50-50 chance to me, which could be changed to:
>  
>    if(number(0, 1))
>  
>  And add these in:
>  
>  - >     if ((number(0, 100) < 50) && (number(0, 100) > 85))
>  
>  Hmm another 50-50, and a %15 chance.  Since these are all and'ed together,
>  if my wild guess at statistics is correct, you've got a 50/100 + 50/100 +
>  15/100 chance of getting through all the ifs, or 115/300, or a 38%.  Am I
>  close?  If so, then you could consolidate all four random number
>  generators into this one line:
>  
>    if(!number(0, 37))
>  
>  Sam
>  p.s. Yeah I know my numbers aren't exact, it's the formula I'm wondering
>  about :P
>  

Hmm. I was going to post much the same thing, but I had figured it at
about 4%.   50% 50% 15% .. #1 falls through half the time.. #2 does the
same so that yields about 25% .. 15% of 25% would be about 4%. So:

if (number(1,100) =< 4 ) {
}

would seem to be equivalent, yet use around a third of the processor
time. Though, seeing as it's not a string operation, one wonders how
much of an impact it really has.

--Mallory



+-----------------------------------------------------------+
| Ensure that you have read the CircleMUD Mailing List FAQ: |
|   http://cspo.queensu.ca/~fletcher/Circle/list_faq.html   |
+-----------------------------------------------------------+



This archive was generated by hypermail 2b30 : 12/18/00 PST