>By the way, I am curious. What features do you think
>should be added/modified in the next revision of the m3d
>API?
We need to use it for a year or so before I can answer that properly - and
it shouldn't just be I answering that question.
I can honestly say that it has everything we need to build great games -it
is an amazing powerful API and every single 3D vendor in the mobile space
was involved in its design. We have 20+ games being built for us by a number
of developers and doing some in-house, and so at some point we'll ask that
question, but not yet. So far we're able to turn around a quality game in
around 6 months, including testing. Our experience to date, and that of the
8 or so development houses building games for us, is positive - we're able
to build games rapidly and the reaction to those games has been great. It's
different to other APIs, especially if you've come from a console
background, but it is a different market of course.
The API is also meant to be future looking as well as being practical for
todays handsets, so there is also an element of extendability in there as
well as being able to expose the underlying OpenGL ES layers if there is HW
underneath. That means that you can actually add to the API if you want, use
different file loaders or use specific features of the 3D accelerator
without needing a new API or engine on the handset. As an example - you
could access Imagination Technology's Vertex Shaders from M3G through
extensions without breaking content.
I would expect that in a year or two the industry will start looking at a
new JSR to design M3G version 2. It'll build on the version 1, look at what
the results of JSR 239 were and other standards initiatives, but what goes
in it will be mostly be determined by developers who have been using it and
the state of HW to see what it needs to accomodate in terms of future
technology.
regards
Paul
Received on Sun Apr 25 20:35:54 2004