As a typical Pi hobbyist, there's usually no point, since the programs are usually small and you're usually supporting only one target. But as a retired software engineer for 45 years, take my word for it that it's typical in the pro world, to support multiple targets, to use a fast build machine, to minimize the number of build setups to maintain, and no doubt other reasons I've forgotten or never bothered to figure out since it's what we almost always did -- even when building in and for the cloud, and even when there's only one cloud target that doesn't match our build environment, and even when different team members use different build environments so we need to maintain build setups that work regardless -- not a trivial amount of extra work. The idea is to let each developer choose what their most effective and productive environment is.Why are you trying to make life so complicated for yourself?
I doubt I'll bother doing that for my Pi work (fun) now that I'm retired -- it's too much like work used to be. Coding Pi is 90% of the fun stuff work used to be without the 40% of annoying/tedious stuff work used to be! (And yay, no meetings.)
Regardless, I'd definitely develop using Linux, even though my laptops are Windows. I'd either set up a VM or use WSL. But that's probably just because I'm used to it, having never been a Windows developer.
Statistics: Posted by jlearman — Sat Dec 27, 2025 12:30 pm