Hi Pat, nice to hear from you :-) > If you folks were to extend the DOS API, in which direction would you > take it? Would you go with POSIX/UNIX/Linux, win32 or simply provide > the same API with extensions for modern functionality and platforms?
Well OS/2 once was a pretty nice and straightforward extension of DOS, but then... I guess it would be a bit too much work to turn FreeDOS into an OS/2 clone. As for the Win32 clones, there are already ReactOS and Japheth's impressive HXRT HX DOS Extender :-). I think one thing that would be pretty useful now would be a virtual SoundBlaster. Even my usually-very-compatible (better than CMedia crap) ForteMedia FM801 based PCI soundcard (thanks to MartinS for that :-)) stopped to work DMA-wise on my new AMD socket AM2 board. A good starting point would be MPXPLAY (hardware drivers) plus JEMM386 (plugin system and virtual hardware support) plus DOSEMU (implementation of virtual SB). This is not DOS API related, though. DOS apps tend to use the hardware directly, and extending the API would only help newly written DOS apps. Modern disks / optical drives / CPU / RAM are no real problem for DOS, but you could think of funky ways to waste the lots of RAM and CPU cycles and cores you have today ;-). Modern graphics cards work okay with DOS, but VGA BIOSes are getting sloppier. Modern network cards often work if you use a NDIS wrapper or similar, which is probably thanks to Ghost still needing DOS drivers. A generic PXE driver exists but not as open source. Next are probably modern storage thingies: USB storage (harddisk, flash, floppy, optical) support in BIOS is getting better (and USB keyboard / mouse support in BIOS works fine as well) but it might still be useful to have a DOS driver to get USB2 speeds and to get less interference (BIOS USB drivers can be unstable and can be CPU hogs and stuff). For PCMCIA / CF, a driver would be nice: Think about a combination of a RAMDISK, kernel initdisk.c, some PnP stuff, Deskwork's donated PCMCIA component source and the non-DMA IDE code of UIDE/UDMA for that :-). Supporting SD cards and similar would not hurt either, but I am not sure what sorts of drivers you would need and how much BIOSes help. For the DOS KERNEL API, it might be a fun idea to be able to hotplug partitioned drives, would make driver writing slightly easier... So I guess the answer to your poll is clear: Support more hardware! Support for software is already sufficient (DOS API and extenders) as it is, if you were to clone OS/2, Win32 or Linux, you better use OS/2(-clones), Win32(-clones) or Linux themselves and their DOSEmu and similar windows for your DOS apps :-). Eric PS: In the long run, a gparted-with-ntfsresize port and supporting the new > 2 TB partitioning scheme would not hurt, but then, there already are very easy to use GPARTED boot ISOs, so why use DOS here? PPS: Maybe drivers for certain VM properties would be nice, too? Many people use VMWare, Virtual PC, Bochs, Qemu, ... for DOS now. ------------------------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Still grepping through log files to find problems? Stop. Now Search log events and configuration files using AJAX and a browser. Download your FREE copy of Splunk now >> http://get.splunk.com/ _______________________________________________ Freedos-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/freedos-devel
