Hi, On Wed, May 2, 2012 at 8:03 PM, Martin Kelly <martin.kelly90...@gmx.com> wrote: > > Don't get me wrong lol. I am running both Linux, BSD and FreeDOS currently, > and FreeDOS is amazingly fast in comparision to Linux.
Linux is great but my main gripes are 1). build complexity, 2). too many incompatible distros, 3). little binary stability, 4). relatively high requirements (and disdain for anything older than P3/1.0 Ghz or just compatibility in general). >>Though alternatively, others (e.g. late genius Pat >>Villani) already had ideas of moving to a 32-bit kernel built via GCC using >>FreeBSD driver framework. > > Sounds Interesting, but wouldn't that mean that DOS would essentially not be > DOS lol, it would DOSBSD???? :P The idea was that having a proper 32-bit OS with built-in support for some things (networking, USB) via external [stable] drivers would be a good fit a la NTVDM (thanks to V86) or whatever. Having a built-in solution would prevent a lot of workarounds, hacks, and kludges (I suppose). Granted, this idea was before Bret and Mike's efforts, so maybe it's less critical now, but the idea was build upon what's already out there as some modern features are considered important these days. >> > -A secondary 32/64-bit capable kernel that *doesn't* replace the > current >> > FDOS kernel but expands upon the existing kernel by writing new 32-bit >> > and 64-bit binaries/ code. > >> Unlike DJGPP or DOSEMU or FreeDOS-32? > > No the intention was different to DOSEMU but similar to FreeDOS-32. > > DOSEMU is an emulator ontop of a host system, practically the DOS version of > virtualbox and the like. Both were hosts with built-in DPMI running natively in 32-bit protected mode using V86 for 16-bit stuff. The only difference is that FreeDOS-32 was never very stable or useful except as a proof of concept (sadly). > The x-windows system was only a small part of the idea. DJGPP just does what > essentially the idea of what cygwin on DOS would have done. Cygnus used to use DJGPP, so that's no surprise. It's just that some people "switched" exclusively to Windows for oddball reasons, hence DJGPP kinda got left in the dust. I'm telling you right now that a LOT of software could work well with DJGPP (and has in the past) but has been dropped for vanity reasons. Considering how stable and powerful DJGPP is, that's a big shame. > The secondary kernel design was more about enabling SMP > and expanding memory access to run more intense applications (bloatware lol) > like firefox. Many have tried and failed. (And Firefox is way way too big and complex, IMHO. Years ago people here dreamed of porting Dillo as a lightweight alternative, and even that was considered way too complex for us. I consider it no small feat that Dillo was ported to DOS [DJGPP ftw!], and I was pleasantly surprised! > Not to re-design the whole system. Whether SMP or greater RAM can be > achieved through a DOS > extender i don't know, or even if SMP is already implemented? > I was of the assumption and knowledge that greater than 4GB of RAM access > requires a 64-bit > kernel or PAE to address that higher RAM. No, there is no SMP or PAE support, at least not in FreeDOS proper. Some have tried looking into SMP, but apparently even their one-app hacks weren't feasible or stable enough. As for PAE, nobody has really needed it because "most" DOS apps don't use 500+ MB of RAM (yet). But like I said, some PAE hacks do exist (untested by me). ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ Freedos-devel mailing list Freedos-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/freedos-devel
