Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!mips!twg.com!david From: david@twg.com (David S. Herron) Newsgroups: comp.windows.x,connect.audit Subject: Re: X-Windows and XINU Keywords: XINU, X Client WM Message-ID: <8805@gollum.twg.com> Date: 29 Mar 91 23:49:12 GMT References: <1991Mar24.220320.15446@ibmpcug.co.uk> Followup-To: comp.windows.x Organization: The Wollongong Group, Palo Alto, CA Lines: 34 In article <1991Mar24.220320.15446@ibmpcug.co.uk> object@ibmpcug.co.uk (Ken Tough) writes: >- We were told the (8/10) MHz 16bit 68000 would be just too slow to > run the Xwindows library calls. This seems a little hard to swallow Indeed. I know of two X servers running on 68000's. The original NCD terminal (NCD16) and GfxBase's server for Amiga & AmigaDOS. Both had >=acceptable performance, supported >=very good sized displays & etc. While the Amiga server relied on the built in custom chips, it also had to share the CPU with other AmigaDOS processes. The NCD, from what could be told by looking at the motherboard, did not have any coprocessors. >- The available X software seems portable among UNIXes but may be a > big job to port to non-UNIX OS. Will SIGNALs and things like > select() calls be an enormous problem? At least for the Amiga server, Dale claims that X11 "just ports". In general .. isn't X11/Xt code (that is, applications level code) pretty much free of particular OSisms? While I've only been working with this for ~ 1 year, I've only touched any Unixisms (select() and the rest of the socket() interface) once. The rest has been X11/Xt, Motif and XVT in varying mixtures. But I am talking about applications code. The layer insulating the application from those OSims *is* the Xt layer so obviously the system-specific uglinesses occur there. Fortunately you only gotta do it approximately once. -- <- David Herron, an MMDF & WIN/MHS guy, <- Formerly: David Herron -- NonResident E-Mail Hack <- <- "MS-DOS? Where we're going we don't need MS-DOS." --Back To The Future