>A side note--processing is faster if you do your if statement like this:
>
>if the mouseH < 475 then
> if the mouseH > 385 then
> if the mouseV < 393 then
> if the mouseV > 370
I just did a test of the different techniques. Here's the movie script:
on mt0
end
on mt1
if (the mouseH < 475 and the mouseH > 385 and the mouseV < 393 and
the mouseV > 370) then
nothing
end if
end
on mt2
if the mouseH < 475 then
if the mouseH > 385 then
if the mouseV < 393 then
if the mouseV > 370 then
nothing
end if
end if
end if
end if
end
on mt3
if inside(the mouseloc,rect(385,370,475,393)) then
nothing
end if
end
and here's the frame script:
on exitFrame
times = []
m = the milliseconds
repeat with a = 1 to 1000
mt0
end repeat
append times,the milliseconds - m
m = the milliseconds
repeat with a = 1 to 1000
mt1
end repeat
append times,the milliseconds - m
m = the milliseconds
repeat with a = 1 to 1000
mt2
end repeat
append times,the milliseconds - m
m = the milliseconds
repeat with a = 1 to 1000
mt3
end repeat
append times,the milliseconds - m
put times
go the frame
end
The empty handler m0 is just to help time the 1000 repeat loop and
handler calling. On my machine it came out as 8 milliseconds. The
all-in-one-line-if technique took 56 milliseconds. Kerry's suggestion
varied from 22 to 59 milliseconds, and my suggestion (which I'm now
making) of using the Inside() function took 46 milliseconds.
Bearing in mind that this was a 1000 checks, we're not talking about
much time difference per check. If you take away the 8 milliseconds
overhead for calling the handler, that leaves it as 48, 14-51, and 38
respectively. Which way to use may depend on the likelihood of the
test being true, for example if the mouse is within that rectangle of
the screen Kerry's routine would be somewhat slower than mine, and
only marginally slower than the first method. If the mouse isn't to
the right of the screen Kerry's routine would be quite a bit faster.
On average his routine should come out a tiny bit faster than mine.
This is an easy case where checking coordinates isn't that demanding.
Kerry's approach would be even more useful if the later tests were
more demanding, for example:
if the mouseh > 600 and random(6) < calculatepi()
where calculatepi is a function that calculates pi to 10 decimal
places, would be a very slow check every time. Testing the mouse >
600 and only then calculating pi would be a lot quicker most of the
time.
Meanwhile, back at reality, your situation isn't so demanding, and if
you waste an extra 30 micro seconds it's not the end of the world.
[To remove yourself from this list, or to change to digest mode, go to
http://www.penworks.com/LUJ/lingo-l.cgi To post messages to the list,
email [EMAIL PROTECTED] (Problems, email [EMAIL PROTECTED])
Lingo-L is for learning and helping with programming Lingo. Thanks!]
