jennings@lll-crg.ARpA (Richard Jennings) (06/13/86)
I have just started to play with Turbo Prolog, and while it may be a limited Prolog it provided a interesting window system for strict PC compatibles. For people interested in human interfaces, and PC's it is (it seems to me) useful. As a consequence of the packages emphasis on using the PC efficiently, I think a generic MSDOS version, while not impossible, would be *very* hard because the (screen) I/O is optimized for the IBM PC. From my perspective, it seems that it is more of a user interface breadboarding system than a language compiler; which will be my use. Rich.
kim@mips.UUCP (06/15/86)
[ "Send lawyers, guns, and money ..." ] > As a consequence of the packages emphasis on using the PC efficiently, > I think a generic MSDOS version, while not impossible, would be > *very* hard because the (screen) I/O is optimized for the IBM PC. > > From my perspective, it seems that it is more of a user interface > breadboarding system than a language compiler; which will be my > use. > > Rich. It is a fine idea to exploit the PC capabilities to the max ... I have no problem with that. But since many of us must run MS-DOS (TI's, DEC's, Fujitsu's, etc.) we would be happy having just the "bare" compiler, linker, etc. without having to have them tied together with windows and such. If a "front-end" is really required, the TurboPascal editor/control-menu is more than sufficient. Sigh! /kim -- UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!kim DDD: 408-720-1700 x231 USPS: MIPS Computer Systems Inc, 930 E. Arques Av, Sunnyvale, CA 94086 CIS: 76535,25
bc@cyb-eng.UUCP (Bill Crews) (06/17/86)
> > As a consequence of the packages emphasis on using the PC efficiently, > > I think a generic MSDOS version, while not impossible, would be > > *very* hard because the (screen) I/O is optimized for the IBM PC. > > > > Rich. > > It is a fine idea to exploit the PC capabilities to the max ... I have > no problem with that. But since many of us must run MS-DOS (TI's, DEC's, > Fujitsu's, etc.) we would be happy having just the "bare" compiler, linker, > etc. without having to have them tied together with windows and such. > > /kim Besides, smart use of DOS screen I/O (like, for instance, using nansi.sys and large, raw writes to DOS) can net the application developer screen I/O that approaches the speed of direct screen update. And it can be portable! You'd be surprised. -- - bc - ..!{seismo,topaz,gatech,nbires,ihnp4}!ut-sally!cyb-eng!bc (512) 835-2266
kim@mips.UUCP (Kim DeVaughn) (06/20/86)
[ "Send lawyers, guns, and money ..." ] > Besides, smart use of DOS screen I/O (like, for instance, using nansi.sys > and large, raw writes to DOS) can net the application developer screen I/O > that approaches the speed of direct screen update. And it can be portable! > You'd be surprised. > -- > - bc - Yep. Even using the straight ol' MS-DOS int21 calls one *can* get reasonable performance. For example, the TurboPascal editor is alot faster than several specialty editors I have seen ... and in the generic MS-DOS version, that's all it does use (int 21's). In fact, the TurboPascal editor is my editor of choice for just about everything except applications that really require "word-processing" capabilities (since <sniff> Mansfield Software's KEDIT is only available for PC-DOS machines). /kim -- UUCP: {decvax,ucbvax,ihnp4}!decwrl!mips!kim DDD: 408-720-1700 x231 USPS: MIPS Computer Systems Inc, 930 E. Arques Av, Sunnyvale, CA 94086 CIS: 76535,25