[
Lists Home |
Date Index |
Thread Index
]
From: Bill Kearney [mailto:wkearney@ideaspace.net]
>> Why are 3D systems such miserable market failures?
>Probably because making truly engaging 3D visuals requires genuine artistic
>talent.
I consider that to be the number one reason, precisely. 3D game systems
are not miserable failures, but they aren't produced by one man shops
either with great success. At the level of games, it is very expensive,
and the gaming market thrives on proprietary systems.
>It's one thing to cobble up some 2D vector art. It's another thing
>entirely to make something in 3D that also doesn't confuse the heck out of
the
>operator trying to navigate within it.
Once again, yes. And in fact, extensibility of the navigation system is
an important topic to the object model designers.
>Combine this with the bleating mobs
>still trying to limp along on crappy-ass 300mHz machines running Win9x and
>you've a recipe for disaster.
Yes. We're up to our heinies in whinies.
>From a different perspective (pun intended) not everyone has the same
>visualization skills.
Yes. It would be interesting to see if Tufte applies.
>But bottom line, 3D is just danged hard to setup and navigate, and too slow
on
>most boxes out there today. At least from a data visualization standpoint.
>Even vector art, a la SVG, remains a pain in the ass to navigate with
current
>plug-ins.
Yes.
1. Real time 3D depends on the extensibility of the object model and
production
depends on fully integratible library models. As I pointed out elsewhere,
this
is no different than Visual Basic or any other design framework. Navigation
extensibility is a good example.
2. Iron and networks for the 3D systems are now available if not ubiquitous.
So a smaller market but a growing one. People do have to eventually toss out
the old machines and operating systems because new programs refuse to run on
them.
The commodity technology is no longer a barrier and ubiquity of distribution
isn't
required; otherwise, the gaming market would fall apart.
3. Games are a subset of the simulation market. The VRMLers started out
thinking
Gibsonian hyperspace, but in fact, that is just fictionalized simulation.
Online
simulation systems are an important topic. Games are too but that market
still
embraces proprietariness and short life cycle content. Simulation systems
for
some customers require long life cycle content for the same reasons CALS
did.
Can REST support real time simulation or is this the example where RPC is
preferred?
len
|
