|
Proceedings
Graphics Interface 2003
Author:
Edited by Torsten Moeller and Colin Ware
Publisher: A.K. Peters
ISBN: 1-56881-207-8
Purchasing: [Amazon.Com]
- US$70.00, [Amazon.co.uk]
- UK£37.02, [TransAtlantic
Publishers] - UK£46.95.
Reviewed: 30th January 2004
Front
Cover Shot:

Overview
This
book is somewhat different from many of the other
books reviewed on this website. In particular, it
is a publication covering an event - as opposed to
a book on a given topic (or field). It is,
essentially speaking, a collection of white papers
that were presented at last years 'Graphics
Interface' conference in Canada.
Graphics
Interface conferences
The
Graphics Interface conferences are not
widely publicized outside of the academic-world,
yet they have been one of the longest running regular
graphics related conferences over the last 30
years. Yes, that's right, the first conference was
over 30 years ago (1969)! It's hard to imagine
just how much the graphics programming world has
changed since then - if it were possible to get
hardcopies of all GI proceedings I'm sure it'd
give a good history of modern computer graphics.
For
those of you not familiar with these sorts of
conferences, they are basically a sounding-board
for the various academics in the field - they go
and present their latest work to other academics
and discuss the pro's/con's of the work. In
general it is the work from these researchers that
eventually filters down to every-day programmers
and real-time/games graphics programmers.
Why
are these white papers of any use?
As
just mentioned, the content of these conferences
will over time become more relevant to the
programmers that actually produce the bulk of
commercial software. There are two ways to look at
this, and consequently, this book:
Firstly,
if it's only just being discussed at the research
level it's not really relevant yet - it's still
just first-drafts and clever ideas. Or, more
importantly, it's still too advanced to simulate
in real-time graphics.
Secondly,
given that the material is only just ready - and
coming from some of the greatest minds in the
graphics world then it's probably time to sit-up
and pay attention. Reading through this work will
prove to be a valuable insight into the next few
years of commercial-level graphics.
The
distinct problem is that most people are inherently
lazy and will go for option 1 - do nothing and
wait till a bit later, and that option 2 does
require that you have your wits about you
and that you're an advanced-level programmer.
Content
The
content in this print is strictly limited to the
material that was presented at the '03 conference
- sounds obvious, but it does mean that the
various papers don't mesh together too well;
unlike a normal book where the author would pick
and arrange material to fit together as best as
possible.
Because
of this, it reads much more like a reference book
than anything else - one where you can have a
quick flick-through when you first get it, but
after that it'll mostly be useful as a reference
when it comes to implementing a technique that has
been discussed.
The
text itself is divided into 10 sections, covering
a total of 36 white papers (roughly 3-4 per
section). These sections are:
Modeling
- Highlights include: An article on
BRDF's and simulations of fluid-solid interaction.
Detail and Context
- Not really much here that's obviously
related to real-time graphics/games
Hardware Methods
- Interesting article on texture based
volume rendering (voxel's)
Input
- Not really related to graphics as such,
but interesting articles that could influence the
input mechanisms used in games..
Rendering
- Various useful papers here, some new
work on ray-tracing as well.
Mixed Reality
- Like with 'Detail and Context', not
much of obvious interest here
Meshes and Surfaces
- A good white paper on View-Dependent
CLoD algorithms.
Multimedia
- A white paper on Human Computer
Interaction in computer games..
Deformable Models
- Deformations are an interesting topic,
and a few clever white papers are presented here
Keynote Speakers
- Not really of any use unless you
attended the actual event.
In
Conclusion
This
book is, as mentioned several times, more of a
reference collection - one that you'd scan read
once and then use as-and-when. The quality of the
print is of a high standard, and follows the
standard formatting used by the majority of
research publications - the writing style is
obviously varied (many different authors) but
generally of a high standard.
Bottom
line is, get a copy of this if you're a hardcore
graphics programmer - but otherwise, you may as
well wait till the topics become a little more
mainstream.
| Good
Things |
Bad
Things |
| •
Interesting - all the topics are rather up
to date |
•
Some aspects either very very specialised
or vague |
| •
Well presented and formatted |
•
Only really suitable for the elite
programmers/developers |
| |
•
Quite expensive for what it is |
| |
•
Being an academic publication, its not the
easiest text to read/understand. |
|