Interesting read on new Xbox audio tool (1 Viewer)

[Dean Cooper]

http://www.teamxbox.com/news.php?id=1762 I can't wait to hear it in action in full DD 5.1.
Dean

 

[Morgan Jolley]

How much you wanna bet they tested it in a game with no graphics? Its probably going to eat up processor power. I personally don't see what the big deal is. Is it just an easy way to tell the game to make a different noise if the player steps on sand or mud and then make the appropriate noise?

 

[Gary King]

If the system works like I expect it would, the only real cost should be a reduction in the number of available voices.
Full physical modeling of sounds would not be something that could be done in real-time (modeling spherical pressure waves constructive/destructive interference, etc. is extremely expensive). However, given the hardware capabilities of the MCP, you could model a final sound as the sum of multiple waveforms, with programmable effects applied to each waveform independently, so that frequency-dependent effects (such as sound diffusion) can be better modeled.
It sounds like a very cool technology.

 

[David_LM]

How much you wanna bet they tested it in a game with no graphics? Its probably going to eat up processor power. I personally don't see what the big deal is. Is it just an easy way to tell the game to make a different noise if the player steps on sand or mud and then make the appropriate noise?

I'll bet you. Even if eats up processor power, the XBox has alot of processor power to eat up, remember Morgan? That is why MS built such a powerful box, to allow for cool features such as this. Just another reason why the XBox will be the next cool thing and the Cube will fall a little short.

 

[Sean Oneil]

I personally don't see what the big deal is. Is it just an easy way to tell the game to make a different noise if the player steps on sand or mud and then make the appropriate noise?

Easy? You should read a little more closely. It states that it is NOT just another sample retrieving method as your description implies.

 

[Bryce D]

Morgan,
This is a development tool... i.e. software for development pusposes.
The actual sound processing chip on teh X-Box will be a custom unit from nVidia. Of which will most likly do almsot all audio calculations on chip, rather then offloading it to the processor.
Even on my dsesktop machine (not my gaming rig), with a whimpy PII-400 and 128megs of RAM, and an SB Live! with full 4 channel audio turned on in Giants, its only pulling ~6% CPU usage for sound. Granted its not 5.1 but then again its not designed for 5.1 and the X-Box chip will be.
The hit on the hardware will not be that drastic.
[Edited last by Dan Brecher on August 30, 2001 at 02:36 PM]

 

[Dan Brecher]

Attention
I have re-opened this thread, (in addition, re-editing comments made in an above post). Allow me to stress, once again, that insults toward other members will not be tolerated on this forum, such advances constitute as a break of forum rules.
Any forum administrator has the right to delete such posts without need of explanation.
Please do not stray off-topic.
Thank you,
Dan Brecher
-Administrator

 

[Morgan Jolley]

If this tool can do something like alter how a noise sounds through walls or change the volume based on the environment or do echoes then I think thats cool. The only problem is that there probably won't be enough time to include it in the launch games. I'm wondering just how much of a difference this will makes in games, though. Yes, it will be great for games like Rainbow Six or Unreal Tournament, but what games for the XB could really use this?
Nonetheless, in 5.1 this would probably be really cool. I hope they use it for Silent Hill 2.

 

[Gary King]

The MCP can already perform simple occlusion and reflection effects; however, it performs the effect on individual voices (in fact, the SB Live does that, too). In reality, a brick wall doesn't occlude all sound equally. The amount of sound transmitted depends on the frequency of the waves making up the sound.
My guess is that this new system helps developers take sounds and dice them up into multiple frequency bands (e.g., 20Hz-1KHz, 1KHz-8KHz, 8KHz-20KHz), and then perform the hardware effects on each band individually (so, instead of using 1 of the 64 voices, this example would use 3). It isn't 100% physically accurate (which would require an infinite number of frequency bands), but it would be a better approximation than what is currently done.
Of course, I could be totally wrong (the article didn't have much information), but that's what I'm guessing it does.
And the launch games should sound good enough without this technology... this 'll be a little something extra for second-gen titles.

 

[Dean Cooper]

I agree with Gary on his assessment of the article. An example of where a developer could use this to his/her advantage is a racecar passing you. I as a developer would create a simple whine noise for the cars engine and have this tool render the sound to raise pitch as it approaches and lowers pitch as it passes and moves away while changing the voice dynamically for object reflections in real-time. Another example I can think of is dynamically rendering changes to the sounds of your weapon depending on your environment reflections. It should really help the developer add ambience and depth to the way a game sounds rather than having flat sound effects. It’s an exciting idea for sure, I wonder what developers will come up with using it.
Dean
[Edited last by Dean Cooper on August 30, 2001 at 05:29 PM]