On Thu, Nov 29, 2012 at 4:48 PM, Robby Findler <ro...@eecs.northwestern.edu> wrote: > But why does 3.225 round differently than 4.225?
Neither 3.225 nor 4.225 can be represented exactly in binary floating-point. The numbers actually parsed are: > (write-rational-expansion (inexact->exact 4.225)) 4.2249999999999996447286321199499070644378662109375(0) > (write-rational-expansion (inexact->exact 3.225)) 3.225000000000000088817841970012523233890533447265625(0) ____________________ Racket Users list: http://lists.racket-lang.org/users
