Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ulowell!apollo!jch
From: jch@apollo.COM (Jan Hardenbergh)
Newsgroups: comp.graphics
Subject: Re: Flavors of PHIGS?
Message-ID: <433d41ed.137b8@apollo.COM>
Date: 15 May 89 16:53:00 GMT
Organization: Apollo Computer, Chelmsford, Mass.
Lines: 30

> From: gil@limbic.UUCP (Gil Kloepfer Jr.)
> Keywords: PHIGS commerical implementations hardware platforms
> Date: 12 May 89 14:27 GMT

> 	1.  Suppose I were to develop under a SUN hardware platform?  Why
> 	    would I use TGS FIGARO over SunPHIGS or vice versa?  Does one
> 	    or the other support more different types of digitizing tables
> 	    and plotter devices?  Do any require OS-specific calls to
> 	    initiate graphic "windows"?

Since PHIGS is now a real standard you want to make sure the PHIGS you use
complies with ANSI X3.144-1988. Template's Figaro has been aropund for years
and was orginally based on earlier drafts of the PHIGS standard. The October
1987 draft became X3.144-1988 AKA PHIGS88.

Now that it is a standard there should be fewer questions about protability.
Another thing pulling for partability is a few heavy weight customers demanding
common data record representations for common device initialization.

I think vendor's native implementations will provide better integration with
other system utilities. For example, Apollo's Domain/PHIGS allows PHIGS to
be initialized in an X window supplied by the application. This provides
the full performance of a native PHIGS along with the niceties that X can
offer for building UI's. I believe other implementations provide this also.
This will also be trivial when PEX becomes an X consortium approved extension.

PEX is the PHIGS extension to X. But, it is not scheduled to be "blessed"
until late 1990 when the sample implementation is given to MIT.

-Jan Hardenbergh    - jch@apollo.com - (508) 256-6600