Originally Posted by appelz
It probably isn't that much different than what DTS:X Pro is doing to increase their channel count beyond 7.1.4.
What DTS:X Pro does is exactly what I'm doing with Top Middle. I'm not trying to improve rendering panning "resolution" so much as to get solid imaging directly overhead in a room that's too long for just four overhead speakers. Panning resolution can improve with pre-rendered object data, but it's a question of whether it's really noticeable, particularly given off-screen sounds have no frame of reference as to where they 'should' appear. You will get different locations of overhead objects using "Tops" versus "Heights" as Tops only use 1/2 the ceiling and put more sounds directly overhead. Which is actually correct for where it's "supposed" to be. You'd need a Trinnov system with BOTH Heights and Tops to even figure that out. Beyond that it's subjective to whether "Heights" or "Tops" are preferred as one puts more sounds directly overhead and the other lets sounds pan further around the room and align better with music sources like Auro-3D that are designed to be played back with "overhead over bed" layouts.
You keep moving the target here.
I haven't moved a darn thing. You say one thing, retract a bit and then say it again. It's hard to follow and frankly, I don't know what your point is other than my original assumption that you want everyone to believe higher speaker counts always equal "better" to push sales in that direction.
You also mention mixing/matrixing, neither of which is happening in your system.
This is where you cross a line with me as you're basically calling me a liar about my own system when I know full well what's in it. This so-called "cascading" surround processors (there's nothing cascaded
about them so I don't know WTF you come up with that term; there's exactly one processor per channel to extract a center-point, exactly as DTS:X Pro does with Neural X.) But that is ONLY
for "Top Middle". As I'VE ALREADY SAID SEVERAL TIMES
, my front wide and SS#2 speakers are MATRIXED
(two channels added summed together with a +3dB correlated maximum and summed uncorrelated canceling each other out where equal) using ACTIVE MIXERS made by ROLLS. They are NOT using any kind of Pro Logic Decoder setup. That is EXACTLY what the Lyngdorf MP-60 does internally for its extra two matrixed channels. I've said that already. You seem to want to insist to dispose of what I actually said and make something up instead and then accuse me of putting words in YOUR mouth. Unbelievable.
You are cascading surround processors, and using ProLogic or some other center channel extraction to create your 11.1.6 system from 9.1.2.
No. I'm not. I've tested 9.1.2 before, but I don't use it. I am creating 11.1.6 from 7.1.4. Top Middle is created using Steered Pro Logic (very similar as Neural X does internally). FW and SS#2 are matrixed from L/R Main + Side Surround = FW and Side Surround + Rear Surround = SS#2. They use ROLLS active mixers to do the summation and set levels, which allows me to actually MOVE the phantom image for side surround forward or backward as desired. This allows me to put the surround speakers in-between rows of seats (BTW, all surround speakers are over 3 feet from the listener save one beside the left middle seat, which is barely audible to that location given it's pointed away from it. NOTHING is playing into
anyone's ears anywhere!
It is a fact that that Dolby Atmos contains objects that are encoded with spatial coordinates, size, and diffusion characteristics, in addition to the bed channels.
You mostly argue about more panning precision with more speakers. That's fine, but when you have people here who cannot hear any improvement with 6 overheads than with 4 overheads in their "typical?" rooms, one has to question at what size room is the extra precision more or less inaudible or more to the point, meaningless? Assuming, an extra set of speakers correct the course of an object's travel through the room even by a couple of feet, can I hear that change reliably sitting 10 feet away? Does it matter if the helicopter veers 18 inches inward with the extra speaker than without it? Would I even notice it? As the room get larger, the margin of error increases and that object is "off course" by maybe 10 feet or more and perhaps that is objectionable sounding for some seats in a larger theater for some reason. But there are other issues as the theater gets larger that are far more important, like phantom images being solid between speakers when the speakers are getting ever further apart. That's why I'd prioritize more speakers for larger rooms or when there's more than one listening seat that's important to you that your guests get a great experience. That's what I've said from the start. Getting a slightly more accurate object travel path with no visual frame of reference is near meaningless and becomes ever more so the smaller the room gets.
To compare to something visual, it'd be like saying you need 4K with a 55" screen when you plant to sit 20 feet away. You couldn't tell 480p from 720p at that distance, let alone 4K. A 4K set may not hurt anything, but it's not going to improve anything either (excluding HDR).
Phantom imaging only works for one seat.
Correct. But I said from the start that more speakers would improve accuracy for off-axis seating and for keeping panning resolution acceptable as the room size increases. Go back to my first post. That's EXACTLY what I said.
In the case of "center channel vs phantom center", it has also been researched and proven that dialogue intelligibility improves when a dedicated center channel is used (large dip around 2kHz), so clearly not identical.
Tell that to my ears when I turned the phantom center on/off to compare my three identical B15 speakers in my original 6.1 setup. Whatever research "proved" I sure as heck couldn't hear any difference kicking it on/off instantly (Atmos systems are slow to switch but old school 5.1/7.1 systems were often instantaneous).
In addition, when dialogue is constrained to the center channel, you can avoid masking of other sounds, by having the L&R speakers reproduce a phantom image for those sounds. This same effect applies to every possible stereo pair in the system.
You do realize you're arguing you need multiple speakers for better quality (less "masked" sounds from a speaker and yet everything you hear on the receiving end is heard by just TWO eardrums, right?) Personally, I'd question this. You'll get more interference and cancellation from using multiple speakers than just two. There's a reason some of the best loudspeakers in the world are two-way designs. My Carver AL-III ribbons cover 200Hz-20kHz with a single line-source ribbon. There's no "loss of resolution" or "more masking". Those are some of the finest reproductions of sound I've ever heard. I can often hear detail on those speakers I cannot on regular driver speakers. Line sources cancel the room out vertically, so the "extra resolution" is from less muddling of reflections from the vertical plane.
Frankly, I've grown tired arguing pointless details about something that is very subtle and subjective in smaller rooms at best and a waste of money at worst for a single listener. If you want to slam my post, go ahead. I'm done here.