From: Mark Pflaging <mpflaging@ddlomni.com>
Subject: Re: VRML: Why VRML can't go anywhere
Date: Fri, 03 Oct 1997 11:16:04 -0400
Message-ID: <34350C34.2F51EB9C@ddlomni.com>
Organization: DDL Omni, LLC


Greg,

	I will add my voice as one of the frustrated users of current
VRML browsers.  Another limitation that you failed to mention was the
lack of support for many new low-cost 3D accelerators.  Its annoying
to be able to play games like quake at 30 fps, 640x480, then when you
try to look at your friends' VRML model, software-only, so try 10 fps
at best.  (and yes, my friends are just as skilled as the artists at
id software (grin)).  WorldView even claims to support 3D acceleration
thru Direct3D: Any Direct3D accelerator is supposed to accelerate
WorldView.  WRONG!  Several 3D accelerators only work in full screen
mode.  Since the browsers have menus, borders, and all the other nice
windoze accoutrements, they don't really run "full screen", and thus
the plug-in can't run full screen, and thus the plugin is often not
accelerated.

	I may be asking for the sun and moon here, but how about a
portable VRML viewer for OpenGL in the public domain (or close)?  Why
couldn't there be a GNU VRML viewer?  Yes I am talking about full
source code.  Of course there would still be plenty of room for
vendors with authoring tools, device drivers, new devices, optimizers,
hardware acceleration, or maybe even (oooh oooh) *embedded* versions
of it for set-top TV boxes or video games or... car dashboards....
The only question would be, who can develop this OpenVRML for free?
It would obviously have to be in C or C++ or other portable language.
(Would it be a "standard"?  No...  emacs is not standard... just
indispensible :)

	VRML is still going somewhere, I think these issues may work
themselves out some other way.  We must have patience.  Its simply not
quite ready for mainstream yet.

	-Mark Pflaging
	DDL Omni, LLC
	mpflaging@ddlomni.com