The Weak Links of Today's Audio

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Disagree with the view that stereo is going to die off anytime soon. There is no standard for multichannel speaker placement and level that dominates the market AFAiK. If you plan to move your speakers around and change settings for every album, have fun.

Disagree that loudness wars are dwindling. AFAIK most of the hi res remasters targeted for streaming and SACD release are often lower DR than the CD that preceded them.

There is a standard with movies.
Do you have to move your speakers for different movies? Most recording studios already have 5.1 channel setups and know how to use them. And many homes also have multichannel systems. Movie music has been recorded and mixed multichannel for over 20 years, so the problem is distribution. Also, tidal etc will need 3 times the bandwith.
 
Last edited:
With all respect to Toole, I find this a bit too easy. If two microphones cannot capture the soundfield, how can two ears?

It was from a post in a forum so Toole shorthanded it. I assume when he says multichannel he means the entire ecosystem that includes multichannel. In other words, not converting source material recorded and mixed for two-channel stereo.

But I don't know if Toole was aware of what Google's subsidiary Deep Mind can do today when he made that post. Yes, an array of directional microphones can capture an environment. However, wavenets will rerender all past recorded music as if it was originally recorded in a perfectly anechoic chamber then place that recording into any environment you want on the fly.

WaveNet: A generative model for raw audio | DeepMind
 
Regarding bandwidth. Fiber, 5G, and SpaceX Starlink are going to supply it.

Fiber and 5G are obvious but SpaceX is less obvious. SpaceX revenue is $2 billion a year from launch contracts. They're building Starship to launch Starlink which will generate ~$70 billion a year. I'm just saying that to identify incentive. SpaceX is going to be a communication company.

Starlink

We're going to have global broadband fairly soon because launch costs are falling so rapidly. You can fly low altitude satellites that burn up because they're so cheap to launch.
 
AX tech editor
Joined 2002
Paid Member
Try this:
Mono noise through 2 speakers (normal stereo setup), then listen to it while moving your head a tiny bit.
You'll hear a massive amount of couloration.
This is the reason why stereo is flawed, it's impossible to recreate the phanthom centre image correctly. Only with a centre channel the centre image can be created correctly.

But when I am in a concert hall, it is just the same. Why doesn't that not bother me? Or do I just accept it because that's how it is?

Jan
 
Wouldn't a separate speaker for each instrument in the proper place sound more realistic, not to mention no phantom imaging moving or sweet spot. Also reduces the load of each speaker, amp, etc. I would love to hear a string quartet thru 4speakers, one instrument per speaker. (and if the cello isn't loud enough you can turn up just the cello.) Yes not a practical solution but multichannel audio is a step in that direction.
 
With all respect to Toole, I find this a bit too easy. If two microphones cannot capture the soundfield, how can two ears?
It is not necessary to re-create the soundfield of the original performance. It 'only' is necessary to recreate the air pressure variations at the eardrum that resulted from the original performance.

The difference may seem unimportant, but it isn't.

I have been at demonstrations of immersive sound systems, up to 19.7 channels. Incredibly complex and expensive, in an effort to recreate the soundfield. Not convincing.
Jan

Next time your at a concert try to hear how little direct sound energy reaches your ears. Even better try to find a recording done from the audience. This is the sound field of the original performance. You won't like it.

Did the multichannel demos you went to use a properly recorded and well mixed multichannel source (no one mixes for 19 channels so they probably used DSP to fake most of them, with inferior results).
 
In the longer term, I think they'll be able to successfully move the instruments in space as though they placed a speaker in each instrument position. They're already trying in larger spaces. I think they'll be able to do it in homes because we have access to so many microphones now in assistant devices. It's not at all ridiculous to think they'll place microphones on each speaker to activate noise cancelling as needed. Harman sort of does that with low frequency cancellation between subwoofers.

When? I don't know. But I hope we at least get three channel stereo soon. Then maybe DML TV screens. TVs are so big now they should be sound sources rather than just boundaries.
 
Administrator
Joined 2004
Paid Member
So speaking of dsp, is there a way to sum a mono center channel using dsp.......I’ve been wanting to mess around with a three channel setup for awhile now.
That would depend on your DSP. In the music player JRiver, it's easy to mix any channels you want. I've used it to make a center mix. You can also set to to decode multichannel from 2 channel. I suppose Equalizer APO could do the same.

On my DCX2496 there is also a sum function usually used for subs, but it could be whatever you want. I have also used it for a mono center.
 
You can put the best speakers and the rest of replay electronics in a room with poor acoustics, it will sound bad.

The best loudspeaker is a human voice. Humans -- and our ancestors -- adapted to voice in open and reverberant spaces. We quickly adapt to new rooms.

The problem is your voice reflects in the room. The reflections become additional sources that add up to the room curve when measured with an omni microphone. Our brains adapted to best make sense of those reflections.

If you equalize the room curve you mess up the nature of the reflections. That's something for loudspeaker DIYers to consider. Build the EQ into the speaker itself, don't hand it off to an equalizer with discrete control used after the fact. Instead, use tone controls that keep the overall frequency response curve intact to adjust for recording errors.

When it comes to room acoustics you can improve your room. A better room is better than a bad room, just like it is with human voices.

The question then for the audio system DIYer (beyond just DIYing the loudspeaker) is whether it is better to address the room in the loudspeaker than in the room itself.

That's where, I think, the M2 vs Salon 2 challenge comes into play. If we use a wide selection of today's 2-channel recordings with all their flaws I think it's better to go the Salon 2 route. If a particular DIYer gets high satisfaction from a selection of really well done recordings and doesn't really care about a wide selection then they might get more satisfaction with the M2 route.

In the future when all recordings are good, the types of loudspeakers that produce the highest general satisfaction ratings might be M2 styles.
 
The best loudspeaker is a human voice. Humans -- and our ancestors -- adapted to voice in open and reverberant spaces. We quickly adapt to new rooms.
The topic is about replaying sound.

The problem is your voice reflects in the room. The reflections become additional sources that add up to the room curve when measured with an omni microphone. Our brains adapted to best make sense of those reflections.

If you equalize the room curve you mess up the nature of the reflections. That's something for loudspeaker DIYers to consider.
I've heard of "house curve" but not sure what you mean by "room curve". :scratch:

Build the EQ into the speaker itself, don't hand it off to an equalizer with discrete control used after the fact. Instead, use tone controls that keep the overall frequency response curve intact to adjust for recording errors.

When it comes to room acoustics you can improve your room. A better room is better than a bad room, just like it is with human voices.

The question then for the audio system DIYer (beyond just DIYing the loudspeaker) is whether it is better to address the room in the loudspeaker than in the room itself.
For example, bad bass response in certain room cannot be addressed with the speaker design itself.

Are you aware that (electronic) equalizer is not the main tool for acoustically treating the room?
 
That would depend on your DSP. In the music player JRiver, it's easy to mix any channels you want. I've used it to make a center mix. You can also set to to decode multichannel from 2 channel. I suppose Equalizer APO could do the same.

On my DCX2496 there is also a sum function usually used for subs, but it could be whatever you want. I have also used it for a mono center.

Thx......don’t want to go too far off topic, when I do start putting it together I’ll start a new thread.
 
I will 75% disagree with Toole.


Why? Because I have heard good old two channel stereo sound astonishingly real, so real that it shifts your perceptions and ideas of what recordings and playback can be. Real enough to make you - and other people in the room - say "How is this possible?" The recordings are not the problem, and it's shocking to learn that.

The 25% where I agree with Toole is that most people will never hear this, and it's near impossible to achieve the illusion in the average domestic listening space. When I've heard the illusion it has always been in large rooms with good to great acoustics. If you are in a smaller space with poor acoustics, it's easier to achieve a sense of realism by throwing more tech at it. More speakers, more channels, more gadgets, etc. That's the terrain where most listening is done. But it still isn't as real as great speakers in a great space.

IME, stereo recording is not inherently the problem, the room and speaker are.

I agree with you. And this not the first time that bradleypnw has tried to convince us that we all need to abandon stereo and put in multichannel or else we are being left in the dark ages. We are audio Neanderthal, so to speak.

It’s no secret that Toole has been a long time advocate of multichannel. It’s pretty clear from his book. But it’s also no secret that he has a large well suited listening room with very, very expensive equipment. I would guess the total cost of his system is well over $50,000.

So how does that help most of us here? Frankly, it doesn’t. Very few, if any, here have that kind of budget. Nor do many, particularly our European and Asian friends, have rooms of that size.

Bradley is getting all of this from the AVS Forum website where the real emphasis in on Home Theater, and not necessarily the best audio sound quality. Their concern is more the total integrated experience of audio and visual. And when you include the visual element the true quality of the audio becomes less important and compromises are acceptable.

In other words, for the same amount of money you can get five mediocre speakers and the amplifier channels needed to drive them instead of two high quality speakers and a matching quality stereo amp. For HT this is considered a reasonable compromise. For audiophile listening it’s not.

So he really isn’t doing us any favors by repeatedly pushing multichannel over stereo to this community. Most people here are striving for outstanding sound quality and proper integration in their listening rooms. Not always an easy task. And once they achieve it they are extremely well satisfied and continue to enjoy listening to music on their stereo systems.

Not many who achieve that level of performance feel they are missing anything by not having multichannel sound. And for most there is a very good chance, particularly in a small to medium size room, that adding more speakers would just screw up what they already have perfected and are fully enjoying.

So thank you, Bradley. But your claim that stereo is a 'weak link' is certainly not shared at all in the minds of most of those here. In fact, many believe stereo can be outstanding in its current form, and is not in need of any change at all.
 
Last edited:
For example, bad bass response in certain room cannot be addressed with the speaker design itself.

Earlier in the thread it was stated that you use room curves to equalize low frequencies but not above 300-500Hz, depends on where you transition frequency occurs. Room curve is probably the same thing as a house curve but I'm not sure because I've never heard of a house curve.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.