nickdesigner

How do I round numbers in Max/MSP? I'd like to have the option of rounding up or down based on the decimal place that I want. So sometimes I want 44.100002 to round up to 44.1 or sometimes I want it to round up to 44. Or maybe I want to round 3.141592 to 3.142. (ideally the rounding decimal place would go up for above 5, and down for below 5).

I'd like to know the math so that I can do it with audio signal objects. The int works for data, but what is there for MSP objects?

I've seen this somewhere on the forum before, but I'm having trouble finding it, so I figured I'd make a nice post with a lot of clear Tags for others.

How do I round numbers in Max/MSP?  I'd like to have the option of rounding up or down based on the decimal place that I want.  So sometimes I want 44.100002 to round up to 44.1  or sometimes I want it to round up to 44. Or maybe I want to round 3.141592 to 3.142.  (ideally the rounding decimal place would go up for above 5, and down for below 5).

I'd like to know the math so that I can do it with audio signal objects.  The int works for data, but what is there for MSP objects?

I've seen this somewhere on the forum before, but I'm having trouble finding it, so I figured I'd make a nice post with a lot of clear Tags for others.  

rounding-numbers-2

If this is to do with your [index~ ] based patch, then you'll want to use [+~ 0.5] >> [trunc~ ] to convert to integer :)

This is for that patch :P did you have a chance to check out that problem? Unfortunately I need to round 44.10002 to 44.1.

This is for that patch :P  did you have a chance to check out that problem?  Unfortunately I need to round 44.10002 to 44.1.  

This looks to be one of those how-computers-work-with-numbers issues that isn't really an issue. I'm sure someone with a bit more of a computer science background will jump in but to sum it up: computers cannot represent every single real number, instead they will give you the most accurate representation they can using floating point form. For 44.1 this appears in a [number~] box to be 44.100002. This is what you have to work with (not a great explanation I know, maybe ask wikipedia about floating point numbers for more detail).

On a more practical side, what do you want to do with these numbers? There may still be a way to achieve what you want if we know a bit more about your goals.

not to forget that there is an option in maxmsp which lets you choose if you want to see 44.100002 or 44.1

you dont miss much when you just turn correction ON.

thanks guys - I am in agreement. I was getting a bug in one of my patches, which I thought might be due to errors involving these small numbers, but I've since realized that it is something else.

thanks guys - I am in agreement.  I was getting a bug in one of my patches, which I thought might be due to errors involving these small numbers, but I've since realized that it is something else.


Rounding numbers