Stereo imaging and center channel

[TankTop]

Like the title says, how does a center channel effect stereo imaging? Or better yet, how does surround sound processing sending so much information to the center channel effect imaging? I think Atmos and DTS X are supposed to address this (maybe) but I was running 3.1 for awhile after running 2.1 and I “felt” like I lost imaging with 3.1. Anyone have any theory or fact to either support or bust my “feeling”?

 

[pozz]

The center channel replaces the phantom center, so it's no surprise. The cues we use to construct the phantom image aren't there anymore. There's a lot of research about this problem. Helmut Wittek and Francis Rumsey have written a great deal about it.

There are distinct differences of opinion here. For clarity, the material played by the center channel is easily better when compared to the equivalent over a stereo pair. But despite its limitations (dependence on early reflections for a sense of envelopment, for example), stereo is the most effective setup for the phantom image cohesion. That trade-off between stereo and multichannel systems hasn't been resolved yet.

 

[TankTop]

The center channel replaces the phantom center, so it's no surprise. The cues we use to construct the phantom image aren't there anymore. There's a lot of research about this problem. Helmut Wittek and Francis Rumsey have written a great deal about it.

There are distinct differences of opinion here. For clarity, the material played by the center channel is easily better when compared to the equivalent over a stereo pair. But despite it's limitations (dependence on early reflections for a sense of envelopment, for example), stereo is the most effective setup for the phantom image cohesion. That trade-off between stereo and multichannel systems hasn't been resolved yet.

Perhaps some sort of stereo cohesion between all the speakers in a Atmos/DTS-x using something similar to Dirac to create 3-dimensional sound. I wonder if we have the processing power to achieve this yet?

 

[pozz]

It's really an experimental field at this point.

There seems to be some kind of unbridgeable disconnect between creating or recording music or soundfields in general with a discrete number of channels, and then playing them back over an undefined system. I imagine the way to solve this is through an object-based system like videogames use. I'm not sure if such a system would be complimentary to stereo or not.