|
|
| Sound News | Press
Releases | Archives | Week In
Review | Editorials
| Articles | |
|
![]() |
|
||||
|
|||||
StarSiege - Rick Overman Interviewed : Rick Overman / Programmer / Dynamix Interviewed by : Mark
Date : June 26th 1998 <3DSS> Thanks again for taking the time to do this interview. Perhaps you can start out be telling our readers a little about your background and your current role with Dynamix? <Overman> As far as the gaming industry
goes, I believe I'm a bit of a rarity. I've been working for the same company, Dynamix,
for seven years now. Nowadays it seems like no one stays at the same company for more than
a few years before hopping to the next one. The first few years at Dynamix, I worked with
the R&D team, which was dedicated to creating/investigating new technologies that
could be utilized by any of the projects that were under development. Being on the R&D
team was a great experience. It honed my programming skills and exposed me to the
development of more products than most people ever get to work on in their entire career.
Since leaving the R&D team, I've worked on Aces of the Deep, Silent Thunder, Battle
Drome, Earthsiege2 and Red Baron II. And now I'm Director and Lead Programmer on
Starsiege. Don't ask how I manage to shuffle both these jobs. All I can say is requires a
lot of coffee, which Dynamix enthusiastically provides. :) <Overman> We have been working on Starsiege for about two
years now. During the first half of the development, the core programmers from the
Starsiege and Starsiege: Tribes formed a single team (mostly old R&D team members).
Starting from scratch, we designed and implemented the Starsiege game engine, which both
Starsiege and Starsiege: Tribes is built upon. <Overman> Starsiege will have a
Dolby-encoded Redbook audio sound track. See the next question for the rest of the answer. <Overman> Let's see... I don't want to
get too specific, but there are places you'll hear things that will surprise you. We
started in the game shell, so pay attention, 'cause there are several very subtle but nice
effects that I've never seen done before. In the sim, you spend most your time strapped
into the cockpit of a huge fighting vehicle; obviously, many of the traditional subtle
environmental sound sources wouldn't be audible through a thick cockpit hull, but the
cockpit is a micro-environment all of its own. Also pay attention to the radio chatter
from your co-pilots, support vehicles, and base commander. <Overman> Being the audiophile at heart
that I am, I was sold on positional audio well before I ever dreamt of coding games. When
I got to Dynamix, it just seemed like the logical thing to do. My goal for Starsiege is to
bring it to a new level. <Overman> Let me answer that by telling
a very short story. When I was writing my software 3D positioning code (before 3D audio
hardware existed), I would test it by creating an invisible fly that would move around in
the world making a buzzing sound. I'd set the fly loose, wait a few seconds, then try to
find it and catch it. The moment I could consistently find it and catch it, I knew I had
it right. 3D audio hardware makes the azimuth, incidence, distance, motion, and
environmental cues more accurate. If I can find an invisible fly, imagine what Starsiege
players will do with the 3D audio cues we will bombard them with... <Overman>Each API/approach has its own
strengths and weaknesses. I believe the set of APIs we have chosen to support in Starsiege
will provide an incredible 3D audio experience to the largest number of users. In
addition, that experience will scale with the power of their machine and the 3D audio
hardware that they might install. <Overman> Most 3D audio hardware cards
are performing incredibly complex audio processing, HRTFs, environmental filtering,
occlusion, etc. Given typical game CPU budgets for audio, we just cannot afford to do that
level of processing in software. The software positioning system that's implemented in
Starsiege produces an incredibly believable 3D audio effect with extremely low CPU
requirements. The Starsiege software positioning subsystem targets those gamers who have
minimum machine requirements and do not own 3D audio hardware. <Overman> While QMDX is a great API that
takes care of many of the tedious tasks involved in coding up a sound system and has great
sounding software 3D, it is a proprietary API. Choosing QMDX would have meant abandoning
the A3D and Creative Labs extensions to DS3D. While DS3D is not the most elegant API, it
is 'native' to Windows and is extendable by third parties. <Overman> Starsiege uses version 1.x of
the A3D API. Version 2.0 of A3D is still beta. <Overman> The audible difference between
the two approaches is subtle. The BIG difference is in the implementation. With the
Creative Labs LIVE! Extension, you define world acoustics by specifying the amount of
reverb, damping, etc., by hand. Aureal has taken a totally different approach with A3D
2.0: you actually pass the geometry of your world down to the sound API where it
calculates the acoustical properties of that area. With this much data, A3D has the
ability to calculate true reflections and occlusions, thus producing a much more complex
acoustical model. While I find the A3D 2.0 approach very interesting, to be honest, it
scares me from a development point of view. I don't want to be passing down high-poly
models because it will eat up too much CPU, so I'll need to create low-poly acoustical
models. If the low-poly models do not produce the desired effect, how do I tell my artists
to adjust the geometry to get the correct effect? Great idea, but there's just not enough
control. <Overman> How many channels can your
card play? The Starsiege sound subsystem can scale the number of channels it uses on the
fly, from just a few to the maximum number of channels supported by your card. All you
need to do is move the slider in the preferences menu. Obviously, there's a point where
the number of channels you throw at a card becomes absurd, but you get to make that
choice. <Overman> 3D audio hardware is
definitely the way to go if you want great sound and great frame rates. If you don't have
3D hardware, Aureal's A3D provides great sounding software 3D, utilizing HRTF's (Head
related Transform Functions) with very reasonable CPU utilization. If you're still holding
on to your old processor waiting for that next generation chip to be released, Starsiege's
software positioning will give you great positioning with minimum CPU utilization. <Overman> The folks at LoudMouth ( www.loudmouth.com ) perform all
that magic. <Overman> As far as sound goes, the
distinguishing features in Starsiege will be flexibility, choice, and scalability. <Overman> The Starsiege team is
dedicated to providing the best 3D audio experience possible. You bet we have plans for
the next Starsiege! It's a bit early to discuss specifics without giving out too many
secrets, but when the time's right, I'm sure we'll hook up for another interview. In the
meantime, I'll be spending much more time with Creative Labs, Qsound, Aureal, and
Microsoft to help shape future extensions to the DirectSound API and to streamline the
process for creating superb 3Daudio effects.
Have comments or questions on this feature?
|
|
|