You're reading it wrong!

I'm pretty sure the internal sound processing capabilities aren't tied to the output encodings (for eg, XB1 varied audio output capabilties based on the AV pack in use).

My expectationguessthingthatholdsnolegalweightwhatsoever is that it'll be a similar situation here.

One thing to add.

To the best of my knowledge, most every game studio records sounds in 48KHz (or lower, usually) and not 96KHz, and they usually release it in a lossy format like mp3 or something similiar and not in PCM format to save a LOT of space.

Seriously, download an app to let you brows/decode the sound files on an XBox game disc, you'll be surprised at the amount of 22KHz, 8-bit stuff you find.

"Beloved" DTS? The DTS/AC3 debate is about as useful as any other religious war. DTS uses clever phasing tricks to encode the various channels (losing some fidelity in the process) and AC3 is a "regular" lossy compression; DTS provides better dynamic range but loses some separation of the channels, and if the sound is generated on the fly and output directly from source it's largely irrelevant. (I assume the XBox 360 has an SPDIF connector to plug directly into a surround system, so there's no DTS/AC3 entering into the equation.)

I also find it hard to believe that under normal circumstances (i.e. with ambient background noise, consumer level speakers/headphones, etc.) you can tell the difference between 96/24 and 48/16 -- don't be fooled by the demo DVD-Audio that came with your Audigy2 NX, as the big difference there is the 5.1 sound, not the sample rate. Install the Creative DVDA player on a machine with an AC97 6-channel card, use the same speakers, and tell me you can _honestly_ spot the difference; I've tried it, and I could (if you concentrated *really* hard) notice changes in the top-end, but if you're listening to the sound rather than trying to spot the difference it's not an issue. (Also bear in mind that most of the sounds in games are low to mid range (explosions and the like), so are pretty much indistinguishable without using expensive monitoring equipment.)

Don't get me wrong, 96/24 is "better" than 48/16 (or 44/16), but unless you've spent hours with a pink noise generator, EQ and spectrum analyser and have been moving furniture around to get a perfectly balanced listening environment then the benefit is lost through basic physics. If your speakers cost less than several thousand dollars and if you don't have a dedicated listening zone (no soft furnishings, no sharp edges, and with everything damped to avoid resonance) then don't sweat it -- even the presence of your body is going to screw it all up!

</rant>