Java Doubles to Max Floats

bwu's icon

Hi all,

I'm working with an mxj external right now that is outputting doubles to max that are really really small (like 1.1E-30) and max floats don't hold that many decimal places, however I want to compare these types of numbers in max?

Should I just continue to write the comparison part of the program in Java? Is there a way that max can retain lots and lots of decimal places?

Thanks, if anyone is interested, the patch involves hidden markov models and is still in the testing phase, but I hope to share it with the max community very soon!

bwu

Emmanuel Jourdan's icon

Max only represents 32 bits floating points values, so you might want to write your comparison in Java.