Great information - appreciated. There are several interesting decisions made in the design of Java and the abstraction presented to the programmer. If Kotlin is presenting a series of “integer types” - Byte, Short, Int, and Long - why not keep the abstraction consistent and deal with issues in the compiler/optimizer? I’ve been studying the byte code generated by the Kotlin compiler and it is really good. It would seem that Byte operations should map onto Byte as Int operations map onto Int and then hide any details from a programmer.