MORE SPEEDTESTS: http://scratch.mit.edu/studios/795672/ ------- This is a test to find the overhead in using broadcast-and-wait compared to directly calling a custom block. The difference may seem fairly large here, but it is doing them 100 times, so you have to divide the difference between the two by 100 to find how much overhead there is for each broadcast-and-wait. In other words, as long as you're only doing *a few* broadcast-and-waits in each 'frame', it's not going to cause significant slowdown. Note: we can't put the 100 broadcasts into a repeat loop, since Scratch only runs such loops at 30fps (unless using turbo). That's why they're stacked on top of each other instead!
