• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Multichannel System for Music - Standards, Setup, Thoughts, etc.

Kal Rubinson

Addicted to Fun and Learning
Industry Insider
Joined
Mar 23, 2016
Messages
903
Likes
701
Location
NYC/CT
#21
Are you sure? Wouldn't it be possible to just adjust for any such latency in the Lexicon processor?
Some but not any. It depends on what adjustment range there is in the Lex. Most are limited to under 50feet/50msec. What is the normal latency of the 8C? (FWIW, the Beolab 90 was on the order of 90msec..)
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,421
Likes
515
#22
My main complaint (as I've been vocal about in the past) is the lack of exceptional classical performances in surround. I actually think it's a better format for rock/pop fans since they are remixing a lot of classic rock to surround though often fleecing people to buy the expensive deluxe/super deluxe box sets to get the surround mix.

In my hundreds of classical surround discs I have (and I do make a strong effort to research and buy discs with good performances) I estimate fewer than 10% if I'm being generous would have exceptional performance and be in surround compared to recordings that are in stereo or mono. But then those moments when everything does align like Mahler's second symphony with Abbado (great hi def video as well) it's truly stunning on a big screen with good speakers. But those instances are very rare.

Then again I should probably be turning in my audiophile card as I'm currently listening to a Naxos digital transfer Benno Moiseiwitsch playing Chopin's Ballades... scratchy 78 rpm surface noise and all :rolleyes:
What does or does not constitute an "exceptional" performance is obviously highly personal. I agree that the Abbado BD-Video set of Mahler symphonies is truly exceptional, as are some others from the Lucerne Festival by him and by also by the Concertbouw under Nelsons in the Shostakovich 8th. One could hardly wish for better, though I find the Abbado BDs to be a bit uneven in sound quality and less than state the art, particularly in the releases from the earlier years of the Lucerne Festival BD Mahler cycle. Later ones, including also the Abbado Bruckner 5th, are absolutely great. Great Mch sound and HiDef video together can be stunning. And, I had been quite cool to Abbado until seeing these beloved BDs.

I do consider myself a somewhat discerning listener and performance counts, hugely. But, I never flipped out over Karajan's Beethoven, for example, though the critics did. I am fortunate in having had frequent access to world class live music from the Philadelphia Orchestra, the Metropolitan Opera, and an an exceptional local chamber series. Occasionally, one hears a mind blowing performance with the wonderful richness of sound that only live conveys, but I always thoroughly enjoy it in any case. One simply has to enjoy the performance and superb sound that is there for what it is on its own terms. As often or not with recordings, my old favorite performance or a critical rave is frequently displaced by something different, something with newer and fresher insights, and sometimes better sound makes all the difference.

Big names do not guarantee great performances. There are, however, a fair number of big names on Mch SACD, like Boulez, Haitink, Colin Davis, Muti, Tilson Thomas, Levine, Harnoncourt, Ashkenazy, Argerich, the Concertgebouw, SFS, LSO, Boston Symphony, etc. But, I have also been very enthusiastic about discovering less well known performers and performances in excellent sound, like:

--Ivan Fischer and the Budapest Festival Orchestra for Mahler, Dvorak, etc. on Channel Classics
--Manfred Honek and the Pittsburgh Symphony for Richard Strauss, etc. on Reference Recordings
--Osmo Vanska and the Minnesota Orchestra for Sibelius, etc. on BIS
--Stile Antico for Renaissance choral works on Harmonia Mundi
--Mandelring Quartet for Shostakovich on Audite
--Many Classical period, Baroque and before specialist European ensembles on numerous labels
-- Etc., etc. to name but a few

Big names aside, I don't find the "batting average" for "exceptional" musical performances in Mch to be any worse than in my CD or LP collections, each also numbering in the thousands. And, I have found the critiques of music, performance and Mch sound to be a quite useful and reliable guide at:

https://www.hraudio.net/
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#23
Some but not any. It depends on what adjustment range there is in the Lex. Most are limited to under 50feet/50msec. What is the normal latency of the 8C? (FWIW, the Beolab 90 was on the order of 90msec..)
Thanks Kal. From a gearslutz thread I see that the 8Cs have a latency of 34 ms in full mode, 3 ms in low-latency mode. The mc4 seems to calibrate for distances up to 12 m. If I set the surrounds to max distance, and the 8Cs to zero distance (and same physical distance from speakers), that should be more or less right, shouldn't it?
 

Floyd Toole

Industry Insider
Technical Expert
Industry Insider
Joined
Mar 12, 2018
Messages
149
Likes
718
#24
Thanks Kal. From a gearslutz thread I see that the 8Cs have a latency of 34 ms in full mode, 3 ms in low-latency mode. The mc4 seems to calibrate for distances up to 12 m. If I set the surrounds to max distance, and the 8Cs to zero distance (and same physical distance from speakers), that should be more or less right, shouldn't it?
Re. adding center and surround speakers to an already good stereo system. First, they need to be of comparably high timbral accuracy. If not, the added speakers draw attention to themselves, thereby damaging the subtle directional and spatial effects that one intends to add. This requirement may or may not cost additional money, but it does require careful choice. A surround system of comparably timbrally neutral speakers is a treat. If not, disappointment is almost assured. For a single listener or small audience, conventional forward firing speakers are recommended. Absolutely avoid dipole surrounds, but well-designed bipoles are OK.

The sound from the surrounds should arrive later than that from the fronts, and such delays are often embedded in the upmix algorithm. Most processors allow for timing (distance) adjustments which should compensate for modest digital latencies.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,421
Likes
515
#25
Thanks Kal. From a gearslutz thread I see that the 8Cs have a latency of 34 ms in full mode, 3 ms in low-latency mode. The mc4 seems to calibrate for distances up to 12 m. If I set the surrounds to max distance, and the 8Cs to zero distance (and same physical distance from speakers), that should be more or less right, shouldn't it?
Not sure, but could inroom measurements with REW help in getting the distance/timing and levels more precisely accurate?

With passive speakers, of course, there is little of an issue, except with subs. Linear tape measure distances suffice. Active subs often have some DSP delay in their input networks, so tape measure distances are inaccurate.

But, you are seeing why the industry eventually adopted the simple, automated channel distance and level calibration by mic approach for Mch, which even the cheapest consumer AVRs now all seem to have, even without DSP room EQ.
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#26
Not sure, but could inroom measurements with REW help in getting the distance/timing and levels more precisely accurate?

With passive speakers, of course, there is little of an issue, except with subs. Linear tape measure distances suffice. Active subs often have some DSP delay in their input networks, so tape measure distances are inaccurate.

But, you are seeing why the industry eventually adopted the simple, automated channel distance and level calibration by mic approach for Mch, which even the cheapest consumer AVRs now all seem to have, even without DSP room EQ.
Yes, I think I will need to use REW for this...

Re. adding center and surround speakers to an already good stereo system. First, they need to be of comparably high timbral accuracy. If not, the added speakers draw attention to themselves, thereby damaging the subtle directional and spatial effects that one intends to add. This requirement may or may not cost additional money, but it does require careful choice. A surround system of comparably timbrally neutral speakers is a treat. If not, disappointment is almost assured. For a single listener or small audience, conventional forward firing speakers are recommended. Absolutely avoid dipole surrounds, but well-designed bipoles are OK.

The sound from the surrounds should arrive later than that from the fronts, and such delays are often embedded in the upmix algorithm. Most processors allow for timing (distance) adjustments which should compensate for modest digital latencies.
Thanks, dr. Toole! Very useful. Ok, so no cheaping out on the surrounds then.
 

NorthSky

Major Contributor
Joined
Feb 28, 2016
Messages
4,999
Likes
431
Location
Canada West Coast/Vancouver Island/Victoria area
#27
Kal, could you please comment on an all analog Master volume for all channels, ty.

Most surround sound processors and AV receivers use a digital Master volume control for all channels. You use one in the analog domain?
 
Last edited:

Kal Rubinson

Addicted to Fun and Learning
Industry Insider
Joined
Mar 23, 2016
Messages
903
Likes
701
Location
NYC/CT
#28
Thanks Kal. From a gearslutz thread I see that the 8Cs have a latency of 34 ms in full mode, 3 ms in low-latency mode. The mc4 seems to calibrate for distances up to 12 m. If I set the surrounds to max distance, and the 8Cs to zero distance (and same physical distance from speakers), that should be more or less right, shouldn't it?
Yes, insofar as that one parameter is concerned.
 

DonH56

Major Contributor
Technical Expert
Patreon Donor
Joined
Mar 15, 2016
Messages
2,799
Likes
3,086
Location
Monument, CO
#29
Speed of sound ~ 1127 ft/sec
12 m = 39.37008 feet
So 12 m => 34.93 ms

If I did the math right (no guarantee, check it) then that is not much margin in "full mode". Whether you need more or less depends on physical placement and phase differences among drivers/crossovers/etc.
 

Guermantes

Senior Member
Joined
Feb 19, 2018
Messages
316
Likes
245
Location
Brisbane, Australia
#30
The sound from the surrounds should arrive later than that from the fronts, and such delays are often embedded in the upmix algorithm.
This seems logical for stereo upmixing but aren't the surround delays already "baked" into MCh ambient recordings such as those recorded in concert halls? Do we need separate delay settings based on the type of recording?
 

Floyd Toole

Industry Insider
Technical Expert
Industry Insider
Joined
Mar 12, 2018
Messages
149
Likes
718
#31
This seems logical for stereo upmixing but aren't the surround delays already "baked" into MCh ambient recordings such as those recorded in concert halls? Do we need separate delay settings based on the type of recording?
I should have said that the delays are also in an original multichannel recording - assuming that the recording engineer is up on his/her psychoacoustics. So, for well done multichannel recordings and properly designed upmixers, it is important only to ensure that all channels deliver simultaneously to the listener an impulse in all channels. I hedged a bit because I have heard examples of both naive upmixers and multichannel recordings over the years. Thanks for the reminder.
 

RayDunzl

Major Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
6,863
Likes
3,032
Location
Riverview, FL
#32
Not sure, but could inroom measurements with REW help in getting the distance/timing and levels more precisely accurate?
I'd say "yes".

Sweep each speaker using "Acoustic Timing Reference" - one speaker is the reference, plays a quick tone before the measured spreaker plays the sweep.

Overlay the impulse responses and you'll see the various delays if they aren't already aligned to the microphone location.

Example:

Left and Right speakers well aligned, single speaker measurements, near perfect overlay, third measurement is right speaker delayed 0.1ms, still using the left speaker as the timing reference

1527210693539.png


Here, left and right speakers playing simultaneously (combined impulse response), with a 0.1ms delay on the right speaker:

1527210830742.png


So, use single speaker sweeps to repair the alignment, and, as a check, send the signal to all speakers simultaneously. If they are all aligned, there will be a single impulse.

Here, both speakers playing the sweep, well aligned with the microphone, and no added delay on the right, so a single impulse is seen:

1527211041292.png
 
Last edited:

Guermantes

Senior Member
Joined
Feb 19, 2018
Messages
316
Likes
245
Location
Brisbane, Australia
#33
Out of interest, I tracked down the reference to upmixing a Jordi Savall recording to MCh which was done by one of the sound engineers, Manuel Mohino:

"For the multi-channel SACD re-mastering I used a Sony DRE-S777 Sampling Digital Reverb in order to create the surround stereo field. The good thing with this vintage reverberator is that Sony's engineers have been in this same church, (Cardona) where the original recording was made, to record its acoustics. Sony released this very first real time sampling reverberation in 1999. Sony Music knew about Jordi Savall recordings and they thought it was a good idea to sample the acoustics in which Jordi was recording. So the SACD surround re-mastering of Jordi's recordings made in Cardona can benefit from the real sound of the church, as if they were originally recorded in surround."

Full text here: http://www.musicweb-international.com/classrev/2016/May/Beethoven_sy3_AVSA9916.htm

This is the unit used:
sony-dre-s777-444583.jpg


I have this disc but haven't yet heard it in a MCh setup as mine is packed away until after we move house.

This makes me wonder, are there domestic upmixing units (AVR or separate processor) that include impulses of concert halls or reference rooms? I know my Sony 3400ES has some music presets that are apparently based on particular rooms.
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#35
Thanks for the input on delay etc guys. In case others are interested, here's what I found out about the Lexicon MC-4 unit: As mentioned, it provides up to 35 ms delay (12 meters) for each channel. The upmix algorithm seem to be set to adding an additional 15 ms delay by default, but that can be tweaked up and down, with maximum 30 ms of delay. I assume that there is no added delay when playing multichannel recordings.

I will be trying @RayDunzl 's method for using REW for this! But first I need to get some surround speakers.
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#36
Another thing on the upmix topic: It looks like Harman have developed an algorithm for upmixing to surround that is even better than Logic 7. It seems to have been ready for prime-time several years ago, but no sign so far of implementing it in receivers or processors... that's Harman for ya.

The result of all that objective and subjective testing are some pretty intriguing innovations. QLS, shorthand for QuantumLogic Surround is one of these advancements. This technology can create a genuine surround-sound experience from a stereo or even mono track of music. Using advanced software it slices and dices like a Ginsu-wielding sushi chef.

Let’s say you’ve got a single-channel song that’s comprised of three elements, a singer, a guitar and some drums. In a mono track all of these parts are mixed together into a jumbled mess. What QLS does is seamlessly isolate each of these constituent elements so they can be sent to a separate speaker of a surround-sound system. For instance the vocals could come from one in the middle, the guitar from a speaker off to the side and the drums from behind the listener. There are numerous settings for this.

These extracted parts are identical to the original, just broadcast from different areas. This means music sounds the way its creator intended it to. QLS aims to eliminate artifacts and phase issues that lesser systems can introduce. It’s a significant step up from its predecessor, Harman’s Logic 7.

Not surprisingly QLS is dramatically more effective with stereo tracks than mono ones since there’s much more data for it to manipulate. Pierce said, “We’re blown away by it,” though he may be a little bit biased.

QLS may sound like something out of a science-fiction movie but in truth it works extremely well. Pierce demonstrated its capability by playing a handful of songs and it absolutely made the sample music come alive, including a mono track of “Penny Lane” by The Beatles.
 

Fitzcaraldo215

Major Contributor
Joined
Mar 4, 2016
Messages
1,421
Likes
515
#37
Out of interest, I tracked down the reference to upmixing a Jordi Savall recording to MCh which was done by one of the sound engineers, Manuel Mohino:

"For the multi-channel SACD re-mastering I used a Sony DRE-S777 Sampling Digital Reverb in order to create the surround stereo field. The good thing with this vintage reverberator is that Sony's engineers have been in this same church, (Cardona) where the original recording was made, to record its acoustics. Sony released this very first real time sampling reverberation in 1999. Sony Music knew about Jordi Savall recordings and they thought it was a good idea to sample the acoustics in which Jordi was recording. So the SACD surround re-mastering of Jordi's recordings made in Cardona can benefit from the real sound of the church, as if they were originally recorded in surround."

Full text here: http://www.musicweb-international.com/classrev/2016/May/Beethoven_sy3_AVSA9916.htm

This is the unit used:
View attachment 12856

I have this disc but haven't yet heard it in a MCh setup as mine is packed away until after we move house.

This makes me wonder, are there domestic upmixing units (AVR or separate processor) that include impulses of concert halls or reference rooms? I know my Sony 3400ES has some music presets that are apparently based on particular rooms.
There seem to have been various attempts pre-2000 to synthetically generate Mch surround effects in concert recordings. After that, discrete recording started to become commercially dominant, and I believe newer Savall recordings were all discrete.

I do not know of anyone today still using synthetic methods. I do not think Mch synthesis on recordings would be considered competitive or acceptable today. And, the equipment and know how in discrete recording in the hall has steadily improved among the fraternity of engineers dedicated to Mch, of course including those in cinema from which Mch music has evolved. However, my industry sources tell me that the large majority of engineers working in classical Mch recording are former musicians or have a strong musical background to go with their technical knowledge.

I have used an Integra HT prepro in the past up until about 5 years ago, and it had a plethora of various Mch upmixing modes, including several from Dolby and DTS with different versions of each, as well as numerous proprietary Symphony, Cathedral, Jazz Club, etc. modes. Take your pick. None on my prepro were adjustable except by varying channel level trims by ear or by just selecting a different algorithm. The older Symphony, Cathedral, etc. modes seem to date back to developments possibly by Yamaha in '80s Quad-era consumer gear and quickly copied by other mainly Japanese mfrs. Those on my Integra were lower in resolution and considered inferior to the newer Dolby/DTS families of algorithms. But, offering a lot of them, each having specific effects for better or worse, seemed more of a "how many of these can we load into the processor" features game than a genuine quest for improved sound quality.

After my curiousity about these various schemes was exhausted, I did settle on DTS Neo6 for awhile on classical music. But, ultimately, I abandoned it in favor of just listening to stereo as stereo while dramatically increasing my listening to discrete Mch recordings.

I never heard Harman's upmixing algorithm, unfortunately. But, I am suspicicious of any synthetic, modeled approach in dealing with the complexities of natural musical sound fields in the hall, though they might slowly evolve. And, since my library of discrete Mch recordings is so huge, I no longer desire upmixing in my preferred listening. I have not bought a CD in over 10 years, yet I am musically quite satisfied with the selections I have.
 
Last edited:

Floyd Toole

Industry Insider
Technical Expert
Industry Insider
Joined
Mar 12, 2018
Messages
149
Likes
718
#38
Another thing on the upmix topic: It looks like Harman have developed an algorithm for upmixing to surround that is even better than Logic 7. It seems to have been ready for prime-time several years ago, but no sign so far of implementing it in receivers or processors... that's Harman for ya.
In my view, Quantum Logic is more accurately described as an alternative to Logic 7, not an advancement on it. Logic 7 pretty much left the front soundstage alone and added pleasant enveloping sounds from the surround channels. It treated classical music with some delicacy, but some listeners wanted more from popular mixes. Quantum Logic uses clever processing to separate at least some of the players in a band and the ambient sounds surrounding them, and then, depending on decisions by humans not involved with the original recording, it can deliver those sound components to different channels or combinations of channels. So, the result can range from tasteful upmixing with added envelopment, to almost "middle of the band" illusions.

It was considered as a significant advantage for multichannel in cars, where background sounds swamp most subtleties in timbre and space. In quiet homes the results are remarkable demonstrations of the ability to extract individual sound sources in recordings, but the overall directional and spatial effects that I have heard cover a wide range of possibilities. The timbres seemed to be well preserved. I have heard none of the current offerings, so I await evidence of the playback alternatives that may have been chosen . . .
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#39
There seem to have been various attempts pre-2000 to synthetically generate Mch surround effects in concert recordings. After that, discrete recording started to become commercially dominant, and I believe newer Savall recordings were all discrete.

I do not know of anyone today still using synthetic methods. I do not think Mch synthesis on recordings would be considered competitive or acceptable today. And, the equipment and know how in discrete recording in the hall has steadily improved among the fraternity of engineers dedicated to Mch, of course including those in cinema from which Mch music has evolved. However, my industry sources tell me that the large majority of engineers working in classical Mch recording are former musicians or have a strong musical background to go with their technical knowledge.

I have used an Integra HT prepro in the past up until about 5 years ago, and it had a plethora of various Mch upmixing modes, including several from Dolby and DTS with different versions of each, as well as numerous proprietary Symphony, Cathedral, Jazz Club, etc. modes. Take your pick. None on my prepro were adjustable except by varying channel level trims by ear or by just selecting a different algorithm. The older Symphony, Cathedral, etc. modes seem to date back to developments possibly by Yamaha in '90s Quad-era consumer gear and quickly copied by other mainly Japanese mfrs. Those on my Integra were lower in resolution and considered inferior to the newer Dolby/DTS families of algorithms. But, offering a lot of them, each having specific effects for better or worse, seemed more of a "how many of these can we load into the processor" features game than a genuine quest for improved sound quality.

After my curiousity about these various schemes was exhausted, I did settle on DTS Neo6 for awhile on classical music. But, ultimately, I abandoned it in favor of just listening to stereo as stereo while dramatically increasing my listening to discrete Mch recordings.

I never heard Harman's upmixing algorithm, unfortunately. But, I am suspicicious of any synthetic, modeled approach in dealing with the complexities of natural musical sound fields in the hall, though they might slowly evolve. And, since my library of discrete Mch recordings is so huge, I no longer desire upmixing in my preferred listening. I have not bought a CD in over 10 years, yet I am musically quite satisfied with the selections I have.
This all makes perfect sense. It seems logical to me that discrete multichannel will be superior in an objective sense. Virtual reality (i.e. upmixed ambience) can't beat reality (real ambience), that's just how it is. But I'm still curious to see how the Logic 7 performs. It does seem to me like people who have tried it prefer that algorithm to most of the others available. I'm very sensitive to hifi sounding "artifical", as compared to real acoustic music in real venues, so if it doesn't sound convincing to me, I won't be slow to notice it.

One major advantage of getting upmixing to work at home though, is that it makes it possible to listen in "multichannel" to a variety of two channel recordings through streaming services. As far as I know there are no streaming services who provide multichannel recordings for streaming, not even qobuz. But as said I'll also try to get hold of some SACDs. If I really like it I might even try to get a computer connected. I just need to get the installation itself off the ground first.
 

oivavoi

Major Contributor
Joined
Jan 12, 2017
Messages
1,216
Likes
638
#40
In my view, Quantum Logic is more accurately described as an alternative to Logic 7, not an advancement on it. Logic 7 pretty much left the front soundstage alone and added pleasant enveloping sounds from the surround channels. It treated classical music with some delicacy, but some listeners wanted more from popular mixes. Quantum Logic uses clever processing to separate at least some of the players in a band and the ambient sounds surrounding them, and then, depending on decisions by humans not involved with the original recording, it can deliver those sound components to different channels or combinations of channels. So, the result can range from tasteful upmixing with added envelopment, to almost "middle of the band" illusions.

It was considered as a significant advantage for multichannel in cars, where background sounds swamp most subtleties in timbre and space. In quiet homes the results are remarkable demonstrations of the ability to extract individual sound sources in recordings, but the overall directional and spatial effects that I have heard cover a wide range of possibilities. The timbres seemed to be well preserved. I have heard none of the current offerings, so I await evidence of the playback alternatives that may have been chosen . . .
This is very interesting. So for classical use - where I personally feel that more ambience and more "information" is most lacking in conventional stereo - it might be that Logic 7 is already good enough? The ideal would be to have a processor where can choose between both of these algorithms, of course!
 

Similar threads

Top Bottom