[ANN] Help with a LiveCode Presentation/Demo

[Kay C Lan]

Monte,

Very interesting, but you've made a mistake as well. Both your code and
Peter's produce exactly the same string, there is no exponential growth.

I created the following script:

on mouseUp
  -- yes it's not a million cause I couldn't wait
   put 100000 into tRepeats
   put the long seconds into tStart
   put "Hello" into tHW
   repeat tRepeats times
      put tHW & "!" into tHW
   end repeat
   put the long seconds into tEnd
   put tEnd - tStart into tTimeTaken
   put "INTO took " &tTimeTaken & cr into msg
   put the long seconds into tStart
   put "Hello" into tWH
   repeat tRepeats times
      put "!" after tWH
   end repeat
   put the long seconds into tEnd
   put tEnd - tStart into tTimeTaken
   put "AFTER took " &tTimeTaken & cr after msg
   if (tWH = tWH) then
      put "TRUE" after msg
   else
      put "FALSE" after msg
   end if
end mouseUp

On my machine both strings end up equal. This appears to be like 'repeat
for each' which we've all learnt is generally much faster than any other
repeat loop. For reasons I certainly can't explain, LC seems to handle 'put
after' exponentially faster than 'put into'.

As Peter cautioned, speed comparisons are fraught with comparing Apples
with Cheeses, but in this case, it made me look into V8 and I see, just
like SC and MC learnt a long time ago, a better optimised engine makes a
world of difference. It also emphasis what Richard Gaskin keeps telling us,
that if speed matters, you'll need to skin the cat a couple of different
ways to actually determine which IS the fastest LC code.

I've now amended my Google Spreadsheet to indicate LC and JavaScript are
similar in speed.

Thanks

On Wed, Jun 5, 2013 at 4:36 PM, Monte Goulding
<monte at sweattechnologies.com>wrote:

> Actually the original code was adding the string and ! To itself so it
> grew exponentially rather than linearly. The final string would have been
> very very very long...
>
> --
> M E R Goulding
> Software development services
>
> mergExt - There's an external for that!
>
> On 05/06/2013, at 6:07 PM, Peter W A Wood <peterwawood at gmail.com> wrote:
>
> > Thanks. It does indeed take only 165 milliseconds (on my machine). I
> guess that does re-inforce my warning of the dangers of such comparisons :-)
>
> _______________________________________________
> use-livecode mailing list
> use-livecode at lists.runrev.com
> Please visit this url to subscribe, unsubscribe and manage your
> subscription preferences:
> http://lists.runrev.com/mailman/listinfo/use-livecode
>