I'm kinda curious about Forth on the Propeller - does most of the interpreter run in the cog, or does it have to dip into main memory a lot?
There are several different Forth implementations for the Propeller chip. Most have a VM run in a cog that creates the basic stack machine, the stacks can be in hub RAM, but I think TachyonForth uses cog RAM and has limits about depth of call. The word definitions and program are always in hub RAM
Last edited by Martin_H on Wed Apr 05, 2017 5:31 pm, edited 1 time in total.
The Propeller page on Wikipedia says, "There are at least six different versions of Forth, both commercial and open-source software, available for the Propeller."
Naturally this doesn't stop me from wanting to write a Propeller Forth of my own!
Also I'm wondering about a 65xx/Propeller hybrid, similar in some ways to what Jac has done.