Comments on: Floating-point errors got you down?

By: Ganesh

Ganesh — Thu, 28 Jan 2010 10:11:51 +0000

Thans man that really worked like magic! dont know what these flex guys doing

By: Dealing with floating point numbers in AS3 « Ramblings

Dealing with floating point numbers in AS3 « Ramblings — Mon, 30 Nov 2009 08:08:33 +0000

[...] Floating-point errors got you down? – Josh Talks Flash Many developers will never encounter issues when floating-point math fails to return the correct result. Every once in a while, when you’re expecting to see a number like 10.1 in your trace panel, you might actually see 10.1000000001. This isn’t a specific issue with Flash either. By it’s very nature, any sort of calculation with floating-point numbers can result in tiny, nearly insignificant errors anywhere in modern computing. [...]

By: Overcoming rounding errors « e2easy AIR applications

Overcoming rounding errors « e2easy AIR applications — Thu, 25 Sep 2008 20:43:36 +0000

[...] that would have involved many program operations to simulate each BCD calculation. Â Then this blog by Josh Tynjala gave me another idea. Â (Ok, probably not a new idea – but it was a revelation to [...]

By: Flash/Actionscript: Innaccurate math results : Ryan Bosinger

Flash/Actionscript: Innaccurate math results : Ryan Bosinger — Mon, 21 Jul 2008 18:10:52 +0000

[...] I originally found this information/code on a post called Floating-point errors got you down? [...]

By: Ryan

Ryan — Thu, 10 Jul 2008 17:45:11 +0000

Josh, thanks a lot man! I actually did run into this while trying to do some AS based audio fades. Your “correctFloatingPointError” function smoothed that out nicely.

By: Troy

Troy — Tue, 04 Dec 2007 08:18:59 +0000

Good post, Just covered some other strategies when working with floating points here: http://troyworks.com/blog/?p=70, and here's a related function for you collection:

public function notClose(number1:Number, number2:Number, precision:int = 0):Boolean
{     
    var difference:Number = number1 - number2;
    var range:Number = Math.pow(10, -precision);
    trace(-range + " " + difference + " " + range);
    var res:Boolean = ( difference < -range ) || ( range > difference ) ;
    trace("notClose" + res);
    return res;
}

By: TroyWorks » Blog Archive » AS3: Tips for Floating Point Hell

TroyWorks » Blog Archive » AS3: Tips for Floating Point Hell — Tue, 04 Dec 2007 08:17:16 +0000

[...] Math.round, or roundToPrecision late as possible, and only when you have [...]

By: labs.tendercreative.com » Blog Archive » The Birds & the Floating Point Errors

Sat, 17 Nov 2007 21:39:38 +0000

[...] there are a few adequate work-arounds (please note that these following two methods were based on a similar article from ZeusLabs for inclusion in the SASAPI Framework: /** * Correct "roundoff errors" in floating point [...]

By: Pete

Pete — Tue, 13 Mar 2007 17:51:24 +0000

Awesome, thanks. I am doing a financial services app in Flex and I can’t just cavalierly wave my hand and say “it’s just the nature of computing” like Adobe does. I was getting something as basic as 0.08 * 0.7 = 0.55999999999999994. This function will come in very handy, thank you!

By: Daniel Tome

Daniel Tome — Wed, 31 Jan 2007 03:15:59 +0000

It happens the other way around too.
Try this in a blank flash file:

trace(10.00000000000001);//outputs 10
trace(10.0000000000001);//outputs 10.0000000000001

cheers