The Height Channel
Dear Gary: I am a longtime subscriber to Widescreen Review and a twenty-five-year user and proponent of multichannel sound for film and music reproduction. I have followed the evolution of the audio side of the home theatre market very closely. I have adopted the latest innovations such as DTS, Dolby Digital, rear center surround channel, tactile sound transducers, etc. into my system as soon as they were announced. Now, with the publication of the “height channel” article in Issue 68, January 2003, another method of layering the sound space was added. After reading your article, I quickly hauled out a “retired” Dolby Pro Logic decoder. Using your description that the channel vectors for height are center front/center rear, I deployed a matrix made up of the center front, center rear, and the derived Pro Logic height channel. The results were, to say the least, spectacular, and not only with movies. In fact, I found that with my particular setup and room, the height channel added the final layer of 3-D sound. My current electronics configuration includes stereo music run through my Yamaha DSP-A1 soundfield processing, with additional external processing added; stereo tactile sound transducers; front, side, and rear effects; center rear; and a derived center front made from the sum of left front and right front (à la the Meridian scheme); and now the height channel. Even my wife and friends who are not “afflicted“ with my particular “audio mania” immediately commented on the magnitude of spatial improvement. I brought them in separately and did not tell them what I had changed but just asked them to describe what they heard. Each one, (more than half a dozen in all), said essentially the same thing...“It sounds so much more 3-D, you can pick any instrument or voice and listen around, in front, and behind it.” I had been running the setup using the old decoders for a few months but wanted to try improved matrix technology for rear center and height. The SMART CS-3X, Jr. really appealed to me as the solution, especially with your recommendation. I purchased a single unit intending to use it for rear center and height per your article. I did wonder how the unit could derive the center front vector required as it receives input from the left/right surround channels only. I figured there must be something I was missing. Anyway, I hooked the unit up in place of the two Pro Logic decoders I had been using for rear center and height. Separation and stability did improve, but upon careful listening to stereo music I concluded that the overall sound had lost some of the highly detailed point source specificity that was so apparent using two decoders. I called SMART Theatre Systems to try and get technical details of the matrix for what they call the “envelop” or height channel, and was able to speak to their chief engineer. We had a great discussion on their technology and he did confirm that they are not using the center front/center rear vectors but instead are using the “out-of-phase” information from the left and right surround channels for the Envelop while using the “in-phase” information to derive the center rear channel. That scheme would equate exactly to the sound difference I had been hearing—a more diffuse and less precise delineation of the height soundfield. For my particular purposes, use of the matrix for both music and movies, this does not provide what I require. So I decided to recreate my original architecture of two decoders, one for center rear one for height. I purchased another SMART CS-3X, Jr. and now I have audio nirvana on STEROIDS! The cost factor of buying two of the SMART CS-3X, Jr. units is so low (direct from SMART for $129 each), that it was a no-brainer. I am going to roll this solution out to a few of my fellow addicts who have setups similar to mine as I believe the result is so wonderful. So thanks to WSR for espousing this added layer of sonic realism and equal measures of thanks to SMART Theatre Systems for a great product that offers such tremendous value. In closing, I guess I would like to ask...so which vectors are the “correct” ones for the height channel, or is it a matter of which type of height soundfield integration one prefers?
John Foxx Bissell
Managing Editor Perry Sun Comments:
First of all, I’d like to thank you for generously sharing your experiences and your enthusiasm. I’m always excited to hear that readers are tweaking with their setups, with the end goal of deriving the greatest possible enjoyment from their home theatre experiences. This novelty of experimentation doesn’t seem to be too common these days, as we have seemingly accepted discrete 5.1 as a configuration which should not be “tampered” with. I remember quite fondly the days of Dolby Pro Logic and LaserDisc (prior to Dolby Digital), and hearing about enthusiasts experimenting with various decoding and processing schemes to try and derive some impression of stereo separation in the surrounds, which otherwise were plain mono. Back to the subject of the height channel. From our viewpoint, you are correct in that it should theoretically be derived from the center front and center rear channels. But perhaps more broadly, height information can come from any combination of the front channels, and any or all of the surrounds. You may even want to consider allowing for this, by combining the three front channels as one input channel, and adding up the three surround channels as the second input channel for your height channel matrix decoder. The use of the out-of-phase channel in a Surround EX-type configuration for height effects does not relate to inter-channel coherence per se, but rather, serves as an additional opportunity to matrix-encode specific channel information.
You can E-mail Widescreen Review @ mailto:editorgary@widescreenreview.com