Trending Lyricist I

17 years on and still no solution to AV Amp delays!

50 replies
1202 views

As a reasonably techie guy I’ve been helping my neighbour set up his domestic sound system. He has a Sonos driven output in 5 separate rooms. There is a mixture of Sonos 5 speakers, Ports feeding AV amplifiers and an Amp running a pair of speakers. All the rooms are very separate acoustically apart from the lounge, which has a large opening into the kitchen/dinner area. The lounge has a Port feeding a good Sony AV surround sound amplifier and the kitchen has an Amp driving a pair of passive speakers. And so he is suffering from the dreaded ‘stadium echo’ effect when playing music in both the kitchen and lounge. The mechanism is easy to understand; the Sonos kit is all synchronised together but the Sony AV system adds its own delays after the output from the Sonos Port. Because the Sony amp, along with many other AV systems, is designed such that the video output is always lip synced with the TV picture, changing from surround to direct modes does not affect the audio delay. So it is impossible to reduce the Sony Amp audio processing delay to zero to ‘solve’ the problem.

I’ve spent quite some time looking back through the Sonos Community blogs. The earliest reference I’ve found so far to the AV Amp delay problem was 17 years ago

There are many, many other threads along the same topic. The root cause of the problem is well understood and explained but there is always an excuse why a fix can’t be offered. Yes, it would be difficult to implement variable delays that would allow any combination of Sonos and AV amps to work together. But, from the blogs, for most installations you only need to delay a single Sonos output device relative to the one feeding the AV receiver. Following the 80:20 rule that would fix the problem for most people.
Technically it is possible. There might be RAM constraints to how much delay can be introduced. But RAM is cheap. So I do not understand why Sonos is still not offering even a partial fix :-(

This topic has been closed for further comments. You can use the search bar to find a similar topic, or create a new one by clicking Create Topic at the top of the page.

Show previous replies

melvimbe

I think this will depend on where the listening position is, when not ideal in terms of the distance from the speaker plane. Where closer than ideal, this Trifield thing will help imaging where without it there may be a hole in the image at the centre. But where further than ideal, as is the case at my end with the listening distance to the speakers more than the distance between each speaker, it won’t make a difference.

All I can talk from is my experience. I have 3 different systems in 3 different rooms. All are based on Meridian surround sound processors. All 3 started off as simple 2 speaker systems. Once I moved onto watching DVDs on one system and upgraded it to surround sound, there was no going back. Music, speech and everything else just sounds more convincing and lifelike.
Even on my office system which has a far from ideal speaker layout - my head is about 0.75m from the centre speaker - I enjoy the listening experience.
But then I’m a ‘practicalist’ rather than a ‘purist’!

I’m not sure you can really call doing DSP to covert stereo signals to 3 or more signals always, even when it causes syncing issues in multiroom audio, as practical.

Danny

T

TheRealAdrianAuthor

Trending Lyricist I

I’m not sure you can really call doing DSP to covert stereo signals to 3 or more signals always, even when it causes syncing issues in multiroom audio, as practical.

I’m not sure why you think this isn’t practical? There are any number of AV amps out there, running in price from not very much up to many thousands of pounds, that that are all capable of doing the job.

For me there are a couple of things here;

Stereo signals encode a wide sound stage into left and right channels. The volume of any particular sound source e.g. a singer, or a guitar or a cymbal, is varied between left and right channels to represent its position between a nominal stage left and stage right. That is simply how the spatial position is coded into the ‘stereo’ signal.
The simplest way of reproducing the sound stage is to feed the left channel through an amplifier to a speaker on the left and the right channel to a speaker on the right. A basic ‘stereo system’. This works Ok as long as the listener is sitting exactly equidistant between the speakers and both speakers are capable of reproducing the full audio frequency range.
Each musical instrument, singer or other sound source is basically a single point source. Therefore if you can analyse the ‘stereo’ encoded sound stage signal, work out where each sound source sits on the sound stage and feed its signal to a speaker at the equivalent point on the listening sound stage then you will get a reproduction of the sound stage that is less dependant on listener position than having just 2 speakers. The listener feels more that they are ‘in the sound stage’
Given the trivial cost of computer processing power these days, the only practical limitation to having multiple speakers across the reproduction sound stage is the cost of the amp and speaker for each location. Technically any number of speakers is possible.

If it is possible to have a better sound reproduction system in one room, and you are prepared to pay the money for it, but you don’t want/need that in acoustically adjacent rooms, I really don’t understand why people are throwing up objections to a slight tweak to the Sonos system that would make it possible.

If a particular individual is happy with a simple 2 speaker stereo system then that is perfectly OK. But there are audibly better solutions in existence. It seems perverse to me that people who haven’t experienced those better systems are arguing that a simple software fix that would enable basic and more advanced ‘stereo’ delivery implementations to work together in harmony shouldn’t be implemented.

I suppose that is why personally I use ROON with Meridian surround sound kit. I’m very happy. Its just my neighbour who is suffering having splashed out on a Sony AV amp for his lounge and Sonos to feed all 5 rooms.

K

Kumar

Each musical instrument, singer or other sound source is basically a single point source. Therefore if you can analyse the ‘stereo’ encoded sound stage signal, work out where each sound source sits on the sound stage and feed its signal to a speaker at the equivalent point on the listening sound stage then you will get a reproduction of the sound stage that is less dependant on listener position than having just 2 speakers. The listener feels more that they are ‘in the sound stage’
If it is possible to have a better sound reproduction system in one room, and you are prepared to pay the money for it, but you don’t want/need that in acoustically adjacent rooms, I really don’t understand why people are throwing up objections to a slight tweak to the Sonos system that would make it possible.

I can understand a six speaker set up with each speaker fed a signal corresponding to each sound source - but what happens then if there can be only one centre speaker convenient to deploy? Re-encode the stereo signal for that situation?
I don't think people are throwing up objections - they are suggesting that the market for such a solution may not be large enough to be of interest to Sonos.

ratty

If a particular individual is happy with a simple 2 speaker stereo system then that is perfectly OK. But there are audibly better solutions in existence. It seems perverse to me that people who haven’t experienced those better systems are arguing that a simple software fix that would enable basic and more advanced ‘stereo’ delivery implementations to work together in harmony shouldn’t be implemented.

Ah, the good old “surely it’s only a few lines of code”. Which simple software fix was that? Delaying a Sonos zone that’s grouped with an AVR, or implementing Trifield in Sonos itself?

K

Kumar

Leaving out the Sonos side of things, at least this sounds more interesting than the Hi Res nonsense. But I can’t see myself deploying six front speakers. And even if I could, for my use case where the speakers are about six feet apart, while the usual listening area is eight to nine feet away, I find it hard to imagine that a 6 speaker front will do better imaging than the stereo illusion from the speaker pair. Or even coded music for just an extra speaker at the centre.

Now if the listening was four feet away for the same speaker set up, I can see a centre speaker adding audible value.

T

TheRealAdrianAuthor

Trending Lyricist I

Leaving out the Sonos side of things, at least this sounds more interesting than the Hi Res nonsense. But I can’t see myself deploying six front speakers. And even if I could, for my use case where the speakers are about six feet apart, while the usual listening area is eight to nine feet away, I find it hard to imagine that a 6 speaker front will do better imaging than the stereo illusion from the speaker pair. Or even coded music for just an extra speaker at the centre.

Now if the listening was four feet away for the same speaker set up, I can see a centre speaker adding audible value.

I’m not sure where this idea of 6 front speakers has come from. I’ve only talked about an extra centre speaker fed by the common parts of the left and right stereo channels.

As for wether the extra channel is worth it, its down to personal preference just like all listening comparisons. As I write this I have just experimented with my office system where I am very close to the front speakers.

With my Meridian surround processor set to Direct mode, the singer’s voice appears to move towards whichever of left or right speaker my head is closest to. This is what you expect from a simple 2 channel stereo reproduction system.

However, if I set the Meridian to Trifield then the singer’s voice remains rooted to the centre of my desk. Instruments etc, tend to stay on the side they are intended to be e.g. bass guitar towards one side. This is a sound stage that, to my ears, provides a better located and more continuous sound stage than basic 2 channel/2 speaker reproduction. But that is only my perception garnered by actually listening to the setup.

K

Kumar

Each musical instrument, singer or other sound source is basically a single point source. Therefore if you can analyse the ‘stereo’ encoded sound stage signal, work out where each sound source sits on the sound stage and feed its signal to a speaker at the equivalent point on the listening sound stage then you will get a reproduction of the sound stage that is less dependant on listener position than having just 2 speakers. The listener feels more that they are ‘in the sound stage

Six speakers came from the words in italics, as an illustrative number. Because the sound stage does not consist of just three sound sources either - left/right/centre - as the word in italics implicitly accept. If 2 speakers are not enough, neither are 3.

melvimbe

I’m not sure you can really call doing DSP to covert stereo signals to 3 or more signals always, even when it causes syncing issues in multiroom audio, as practical.

I’m not sure why you think this isn’t practical? There are any number of AV amps out there, running in price from not very much up to many thousands of pounds, that that are all capable of doing the job.

As I already said, because it’s causing sync issue with your multiroom system. I get that DSP manipulation of stereo audio is common, regardless of whether that be converting to 3 channels, 5, or whatever. I don’t see it as practical to expect Sonos, or any other multiroom audio system really, to allow users to manually tweak delays in any or all speakers in the system to accommodate processing on external systems.

As an aside, I find it interesting that music in never produced in 3 channel audio. Two channel, 5.1 for a while, and now atmos music. With the amount of homes that now had LCR setups and/or soundbars, seems a little surprising. Also, many new cars come with a center channel for audio, even though they only have 2 channel sources. Granted, this is because the manufacturer can use cheaper speakers while giving the impression to customers that they are getting a better audio system worth paying more for.

Danny

K

Kumar

On the other hand, if a centre speaker does the job of anchoring the singer to the centre, I strongly suspect that my “cheap” solution of adding a play 1 between the Quad pair would also do that just as well. As I said, I only ran my trial where the listening position was further away from the ideal one, so this was not noticeable because there was no hole in the centre to start with. I might just try to run the test again some day, and see if the usual consequent hole in the centre of the image when closer to the speaker plane than ideal, is filled in by the play 1. I suspect that may well prove to be the case.

It will still be of no real value to me, because with my listening position further away from the equidistant ideal, there is never any hole in the centre of the illusion created by just the speaker pair. Even if I move a little to the left/right of the listening area.

ratty

As I understand it, the filling in of the hole-in-the-middle effect is a secondary factor. It’s where the listener is well off-axis and the processing removes centre-stage sounds from L&R and dumps them in the C speaker that it’s potentially of most benefit.

Show more replies

Sign up

Login to the community