[comp.windows.x] X on PC

rajohn@hubcap.clemson.edu (r alan johnson) (04/28/89)

I'm interested in some sort of prog or dev. tool that will allow me to write
code that I can generate in a PC environment (DOS, not unix) and when then
ported to X will run as well. Simply put, I want some sort of
way to write X apps (or at least use calls that X will understand) that runs
under DOS, without unix, networking, etc. Is this a dumb request? Am I missing
something terribly important? Any and all input will be appreciated. Lacking
help on this problem, is (are) there any tool(s) that will allow dev. and
implementation under DOS that can then be ported easily to a Sun wstation?
Again, and and all help!

				Thanks,
				R. Alan Johnson (rajohn@hubcap.clemson.edu)

p.s. fortran is preferable, but C or some deriv. is OK too.

pajerek@isctsse.UUCP (Don Pajerek) (04/29/89)

In article <5309@hubcap.clemson.edu> rajohn@hubcap.clemson.edu (r alan johnson) writes:
>I'm interested in some sort of prog or dev. tool that will allow me to write
>code that I can generate in a PC environment (DOS, not unix) and when then
>ported to X will run as well. Simply put, I want some sort of
>way to write X apps (or at least use calls that X will understand) that runs
>under DOS, without unix, networking, etc.

Either you're missing something, or I am.  Basically, the problem with
what you want to do is that X applications employ the 'Client/Server'
model; i.e., there are minimally TWO programs running, the client and
the server.  Now, there's no reason why both can't be running on the same
machine, unless that machine happens to be a single-tasking DOS machine.

If you really are restricted to PC's, then you will need two networked
machines. One of them will run an X server (there are several commercial
implementations of the X server for PC's under DOS).  The other will run
your client applications.

>				R. Alan Johnson (rajohn@hubcap.clemson.edu)


Don Pajerek

wastebasket@LARRY.MCRCIM.MCGILL.EDU (der Mouse, really mouse@...) (04/30/89)

>> I'm interested in some sort of prog or dev. tool that will allow me
>> to write code that I can generate in a PC environment (DOS, not
>> unix) and when then ported to X will run as well.

> Either you're missing something, or I am.  Basically, the problem
> with what you want to do is that X applications employ the
> 'Client/Server' model; i.e., there are minimally TWO programs
> running, the client and the server.  Now, there's no reason why both
> can't be running on the same machine, unless that machine happens to
> be a single-tasking DOS machine.

There are not *necessarily* two programs running; one could design a
library with the same interface spec as Xlib but which included the
actions the server would normally take as well.  Alternatively, on an
IBM PC family machine, one could build the server as a TSR program and
have the Xlib calls use one of the unused soft interrupts to request
the services of the server.  (This would imply the equivalent of
synchronous mode, but it's a lot better than nothing.)

I have no idea whether anyone has built such a thing.

					der Mouse

			old: mcgill-vision!mouse
			new: mouse@larry.mcrcim.mcgill.edu
(Replies will NOT reach me; use the signature address.  This is an
 attempt to sidestep the flood of failure messages that inevitably
 follow any mail to xpert.)

paolucci@snll-arpagw.UUCP (Sam Paolucci) (04/30/89)

In article <225@isctsse.UUCP> pajerek@isctsse.UUCP (Donald Pajerek) writes:
->In article <5309@hubcap.clemson.edu> rajohn@hubcap.clemson.edu (r alan johnson) writes:
->>I'm interested in some sort of prog or dev. tool that will allow me to write
->>code that I can generate in a PC environment (DOS, not unix) and when then
->>ported to X will run as well. Simply put, I want some sort of
->>way to write X apps (or at least use calls that X will understand) that runs
->>under DOS, without unix, networking, etc.
->
->Either you're missing something, or I am.  Basically, the problem with
->what you want to do is that X applications employ the 'Client/Server'
->model; i.e., there are minimally TWO programs running, the client and
->the server.  Now, there's no reason why both can't be running on the same
->machine, unless that machine happens to be a single-tasking DOS machine.
->
->If you really are restricted to PC's, then you will need two networked
->machines. One of them will run an X server (there are several commercial
->implementations of the X server for PC's under DOS).  The other will run
->your client applications.

Of course, you could also use an Amiga which has a multi-tasking OS and
which runs the server and clients at the same time.  X11 has just been
released on the Amiga, and a bunch of clients are also included.  Xlib
and the toolkit libraries have been promised in the near future.


-- 
					-+= SAM =+-
"the best things in life are free"

				ARPA: paolucci@snll-arpagw.llnl.gov

Dion_L_Johnson@cup.portal.com (05/02/89)

R. Alan Johnson wrote:
[ .. some deleted stuff .... ] 
If you really are restricted to PC's, then you will need two networked
machines. One of them will run an X server (there are several commercial
implementations of the X server for PC's under DOS).  The other will run
your client applications.
>				R. Alan Johnson (rajohn@hubcap.clemson.edu)

Or, if you prefer to avoid networked DOS,  there are UNIX
implementations that run on PCs!
SCO is just starting to ship developer versions of SCO Xsight under
SCO UNIX for 386 machines.
- Dion L. Johnson  (dion_l_johnson@cup.portal.com)




Don Pajerek

bob@eriador.prime.com (05/05/89)

>> Or, if you prefer to avoid networked DOS,  there are UNIX
>> implementations that run on PCs!
>> SCO is just starting to ship developer versions of SCO Xsight under
>> SCO UNIX for 386 machines.
>> - Dion L. Johnson  (dion_l_johnson@cup.portal.com)


I think that most of these responses miss the point.  There is nothing
about X which *dictates* that it run over a network or on a multitasking
system.  Period.  

You could easily right a graphics library on the PC which had the same 
API as Xlib.  You could even do event handling the same way.  You could then
take some X application and recompile it on a PC and it would work.

bob pellegrino
Prime Computer, Inc.

herrarte@mcs.anl.gov (Virginia Herrarte) (02/27/91)

I would appreciate some advice from anyone using X on a PC as to what the best
packages available are.  I use X on a Sun workstation at work and I am trying to decide
if it would be worthwhile to try to get it on my PC at home.

Thanks.

Vicky Herrarte
herrarte@achilles.ctd.anl.gov
herrarte@mcs.anl.gov

cressler@hpcupt1.cup.hp.com (Scott Cressler) (02/28/91)

You sort of asked two questions: what is the good "X on a PC" package, and 
how do I get X at home?  I don't have a good answer on the latter, partly 
because "it depends" and partly because *I* don't have anything.  However, I'd
just throw out the good solutions I've heard being used: a GraphOn X terminal
(which could work across a modem to your work workstation) or some sort of 
UN*X on your PC at home (essentially converting your PC into a workstation).
I also believe something called DesqView (which I have not looked into much
but which I think is sort of a MS Windows competitor) might support local X.

As to "X on a PC", i.e. *displaying* X windows on a PC, XVision from VisionWare
(UniPress or GSS in the U.S.) is *excellent*.  It runs under (with?) Windows 3.0
on a PC (with 2 meg, a LAN card, and networking software :-) and will display
either an entire X environment in a Win3 window or (and this is the part I 
love) a single X client window in a single Win3 window.  If you're widgets are
similar enough to Win3 (e.g. OSF/Motif), you wouldn't even have to know it was
an X client!

Scott.

dga@cs.brown.edu (Daniel Gerardo Aliaga) (03/24/91)

This might have been asked before but, is there any X11 server that runs
on a 386 or 486 machine (MIT X11 R4 if possible). I now there is X client
equivalent that allows you to run off a X server, but I would like the PC
to be the server and client.

Thanks,

dga


 XXXXXXXXXXXXXXX       XXXXXXXXXXXXXXXXXX      Daniel G. Aliaga '91
          XXX         XXX     XXXXX            Brown University
        XXXXXXXXXXXXXXXXXX 
         XXXXXXXXXXXXXXXXX		 "Real men write self modifying code"