On Tue, Oct 22, 2024 at 1:17 PM Wolf Bergenheim <wolf+...@bergenheim.net> wrote:
> Yeah one does not simply write a book like that, however, I think it could > be a worthy effort and interesting. I don't think FreeDOS needs to be a > from scratch distro, but I think there certainly would be space for a DOS > from scratch distro :) > You're just going to have to find the people who want to do it, or write > it yourself. I would be interested in using it, for instance, as one of my > projects I plan to install freeDOS/svarDOS on my HP Mini 210 which I have > lying around... I'd like to try optimizing for as high of a CPU as possible > and see how fast I could make DOS code 🤔 > Let me be the one to burst your bubble. FreeDOS and other DOSes in general don’t really take a huge advantage of 386, 486 or later processors. The compilers and assemblers they rely on also don’t really optimize code for later processors. Things like FPU, MMX, super scalar instruction scheduling, 32-bit registers or API calls and even basic clock frequency management was largely out of reach for the compilers & assemblers. There was bigger concern for backwards compatibility. >
_______________________________________________ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel
