Originally Posted by lovingdvd
Looks like a great setup! I'm curious how you were able to get 9.1.6 going using 4 AVRs. I understand if this is matrixed, but if not, then how is the DSP being handled? For instance how can the Wides be turned on and one system knows what to do with it while the other system is able to account for whatever the wides would be doing? I would think it all has to be "orchestrated", no? Also why does it take 4 AVPs instead of 2?
Thanks, I do enjoy this setup greatly.
The trick is to use two Atmos AVRs to do the main work. It turns out the Atmos renderer isn't very clever at all, it is a clear split between the base layer and the height layer. It doesn't matter if you have 2 or 4 heights, the same amount of audio is extracted from the base layer and the heights are rendered to any and all speakers available.
As a result I set the base layer AVR1 to run 9.1.2, and leave the two heights unused.
The height layer AVR2 is set to 5.1.4 and I leave the 5.1 unused.
AVR1 will render everything to the 11 speakers it knows of and anything that is destined for the heights are sent to the ether and no height information is left in the base 9 speakers thanks to the Atmos processing.
AVR2 will also render everything to the 9 speakers it knows of and anything that is destined for the base layer is sent to the ether while the 4 heights it knows of will render as intended.
Mix the two together and you have 9.1.4 with no overlapping. The second set of AVRs are simply doing Dobly ProLogic II Movie upmixing of the 4 heights to turn them into 6 discreet speakers.
Due to the PLII upmixing I can't run any room correction on AVR2 as that will break the center extraction well and proper.
As a result of the above I can play the 9.1.6 test tones and everything is rendered where it is supposed to be with 0 bleeding. Now DTS:X is a different matter as it is trying to do phantoming between the base and height layer. To minimise this impact without having to upgrade AVR2 to an 11channel model I simply set the height on AVR1 to TM and on AVR2 I set the heights to TF an TR instead of FH and RH.
The reason I don't use the TM from AVR1 is simple, those two speakers will play all sounds destined for the heigths duplicating any sound that AVR2 would send either to the TF or TR. This is why the PLII AVRs are needed.
I have a wall of text how everything is connected/set up in post 5 in this thread.