Disclord
900 Club - QQ All-Star
The overhead channel doesn't need full frequency response because we actually have a high-frequency roll-off in our hearing for sounds that come from overhead. And we can't localize low-frequency sounds below about 120 Hz that are above us either, so an overhead channel + loudspeaker need not have full response to give good results. It needs to match the main speakers in the midrange and be flat over the range it does cover, along with matching the power handling capability of the main speakers/channels, but that's about it. Most of our perception of "overhead" comes from side-reflections of the overhead signal being bounced off walls and such or the floor (if its a reflective surface and not carpet), straight into our ears. That's why it's so important in a surround system to have a moderately reverberant listening room - with a moderately reverberant room the speakers can disappear as 'point-sources' and so-called "ambience labeling" then occurs, allowing us to locate phantom sounds with much higher precision and localize sounds in front of or behind each other, giving a much more realistic, "3D" soundfield than if the room were completely dead. The reverb needs to come from all around, evenly in all directions, and not be excessive in the front or back - in fact, you want the most reflections from SIDE walls, since that's how we pick up audible cues to the 'size' of the auditorium, etc... This was discovered by multichannel workers back in the early 70's and has been experimentally confirmed in every listening test ever since. The BBC as well as the National Quadraphonic Radio Committee in the USA also confirmed the results.
Periphonic reproduction was studied extensively by Michael Gerzon (co-inventor of Ambisonics and Meridian Lossless Packing) and he tried and abandoned numerous periphonic layouts - such as a matching floor (down) + ceiling (up) array that used a double (cube) array of speakers. He also tried a tetrahedral layout - the system Dolby is using for Pro-Logic IIz is the worst possible way to use the capability for more speaker feeds. Before we go to an overhead channel we need to get a 5-across-the-front layout like Cinerama, Todd-AO and SDDS-8 into home use - since we are the most sensitive to sounds coming from the front, we need to optimize localization for that direction - adding additional channels in the directions we are least sensitive to is a very poor use of the capability. Why waste speakers/channels for quadrants we can't localize correctly? The MPEG-2 Multi-channel surround system had a 7.1 channel mode with 5 channels arrayed across the front soundstage - but it was always ignored or downright ridiculed by people like Tom Holman who doesn't like that layout - nor does he like movies that have dialog that pans with the actors movements! He thinks films should be remixed to take all panned dialog and lock it to the center speaker. UGH!
If you have an extra Circle Surround, Shure HTS or Dolby Pro-Logic decoder that's not being used you can hook it up across the back channels and send its "surround" output to an overhead speaker - or, connect it between the Center Front and Center Back inputs (Cf to Left In and Cb to Right In on the decoder), and send the "Center" output of the decoder to an overhead speaker and the Left Front output to Cf of your receiver and Right Front output to Cb) The film "We Were Soldiers" was the first (and so far, only) film to use Dolby's "Sonic Whole Overhead Surround" (what an awful name!). The so-called "Extra" output of a cinema "Surround EX" type decoder was used to derive the overhead signal. In fact, Dolby actually used SMART Devices "Center Surround EX" decoders because the "extra" channel on Dolby EX decoders is really the "surround" output of a Pro-Logic decoder and has time delay, noise reduction and too much HF roll-off applied to be appropriate as an overhead speaker feed. The theatrical and consumer Center Surround EX decoders use the Circle Surround II decoding system.
Contrary to what Dolby has stated about the DVD release - that the overhead channel encoding was removed for the DVD mix - the overhead channel IS present on the DVD. This was confirmed by Norm Schneider, owner of SMART Devices, who talked to the mixing engineer that oversaw the DVD release and by comparing the 5.1 channel print master directly with the 5.1 channel retail DVD sound mix. Sadly though, since that single release, no other films have been mixed for Dolby Sonic Whole Overhead Surround.
Regarding using the .1 LFE channel of a Dolby Digital or DTS Digital Surround DVD to add an extra overhead channel - that can't be done - both AC-3 and DTS Coherent Acoustics lossy coders roll off the 'raw' LFE input and then re-sample it at a much lower rate for encoding as the DTS "Core" data, but I don't know if that rolled off LFE is also stored losslessly in the HD "extension" stream or if the roll-off is bypassed and then losslessly coded as difference information. So the LFE track can't hold anything above 200 Hz in either AC-3 or DTS "core". I'm not sure how DTS-HD Master Audio and Dolby True HD encode their LFE channels.
Periphonic reproduction was studied extensively by Michael Gerzon (co-inventor of Ambisonics and Meridian Lossless Packing) and he tried and abandoned numerous periphonic layouts - such as a matching floor (down) + ceiling (up) array that used a double (cube) array of speakers. He also tried a tetrahedral layout - the system Dolby is using for Pro-Logic IIz is the worst possible way to use the capability for more speaker feeds. Before we go to an overhead channel we need to get a 5-across-the-front layout like Cinerama, Todd-AO and SDDS-8 into home use - since we are the most sensitive to sounds coming from the front, we need to optimize localization for that direction - adding additional channels in the directions we are least sensitive to is a very poor use of the capability. Why waste speakers/channels for quadrants we can't localize correctly? The MPEG-2 Multi-channel surround system had a 7.1 channel mode with 5 channels arrayed across the front soundstage - but it was always ignored or downright ridiculed by people like Tom Holman who doesn't like that layout - nor does he like movies that have dialog that pans with the actors movements! He thinks films should be remixed to take all panned dialog and lock it to the center speaker. UGH!
If you have an extra Circle Surround, Shure HTS or Dolby Pro-Logic decoder that's not being used you can hook it up across the back channels and send its "surround" output to an overhead speaker - or, connect it between the Center Front and Center Back inputs (Cf to Left In and Cb to Right In on the decoder), and send the "Center" output of the decoder to an overhead speaker and the Left Front output to Cf of your receiver and Right Front output to Cb) The film "We Were Soldiers" was the first (and so far, only) film to use Dolby's "Sonic Whole Overhead Surround" (what an awful name!). The so-called "Extra" output of a cinema "Surround EX" type decoder was used to derive the overhead signal. In fact, Dolby actually used SMART Devices "Center Surround EX" decoders because the "extra" channel on Dolby EX decoders is really the "surround" output of a Pro-Logic decoder and has time delay, noise reduction and too much HF roll-off applied to be appropriate as an overhead speaker feed. The theatrical and consumer Center Surround EX decoders use the Circle Surround II decoding system.
Contrary to what Dolby has stated about the DVD release - that the overhead channel encoding was removed for the DVD mix - the overhead channel IS present on the DVD. This was confirmed by Norm Schneider, owner of SMART Devices, who talked to the mixing engineer that oversaw the DVD release and by comparing the 5.1 channel print master directly with the 5.1 channel retail DVD sound mix. Sadly though, since that single release, no other films have been mixed for Dolby Sonic Whole Overhead Surround.
Regarding using the .1 LFE channel of a Dolby Digital or DTS Digital Surround DVD to add an extra overhead channel - that can't be done - both AC-3 and DTS Coherent Acoustics lossy coders roll off the 'raw' LFE input and then re-sample it at a much lower rate for encoding as the DTS "Core" data, but I don't know if that rolled off LFE is also stored losslessly in the HD "extension" stream or if the roll-off is bypassed and then losslessly coded as difference information. So the LFE track can't hold anything above 200 Hz in either AC-3 or DTS "core". I'm not sure how DTS-HD Master Audio and Dolby True HD encode their LFE channels.
