I assume using baseconvert has the same slowdown issue? A function call from library, so not as fast?
put the last char of baseconvert(x,10,2) is 1 into isOdd On Mon, Aug 5, 2013 at 10:56 PM, Mark Wieder <mwie...@ahsoftware.net> wrote: > Mike- > > Monday, August 5, 2013, 6:38:09 PM, you wrote: > > > It's fascinating that the chunk operator is faster than the integer > modulo > > or even the FP division. > > The "last char" chunk can be determined quickly with a pointer. > Math operators are turned over to the math library, which will > necessarily be slower due to the overhead of the function calls. If > the compiled code were optimized, a divide-by-two operation would end > up being a bitwise right shift, which would be as fast as the chunk. > > > The debate over computational efficiency and what you should and > shouldn't > > do are interesting when you have BIG loops, but did anybody bother to > > figure out how many iterations it would take before you would even > notice? > > You aren't using a tool like LC because you are worried about squeezing > > cycles. > > No, but I sometimes need to optimize to squeeze cycles *because* I'm > using a high-level tool like LC. > > -- > -Mark Wieder > mwie...@ahsoftware.net > > > _______________________________________________ > use-livecode mailing list > use-livecode@lists.runrev.com > Please visit this url to subscribe, unsubscribe and manage your > subscription preferences: > http://lists.runrev.com/mailman/listinfo/use-livecode > _______________________________________________ use-livecode mailing list use-livecode@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-livecode
