Bass and subwoofers

[goat76]

Maybe what's left to do for the doubters is to set up stereo bass in their room, and give it a serious investigation on their own?

 

[Tom C]

I am retiring from this thread for good. I don't see sufficient support from the elder members, and based on the stats, people don't seem to like my posts.

The question here is entirely different as you seem to frame it. To cut a long story short, I do have 4 subs in what is largely considered the best placement - 2 fronts two backs. They are pretty much flawless in every respect when supported with ART. This stereo bass thing is really making them sound bad. I can understand to some extent that people actually don't really understand what's really good. I also enjoyed terrible bass over decades.

Best luck to all of you guys.

Sorry I missed the backs.

 

[Chromatischism]

To all those doubters who are sceptical that this phenomenon exists, I point you towards publications of peer-reviewed research that PROVES it exists.

I don't think anyone was doubting that it exists. At least, if we are talking about the same thing--the spacious bass effect described by David Griesinger.

It just doesn't seem worth compromising your system's optimization for only a small percent of content.

Now for "stereo bass"...this is mostly not possible due to the length of bass waves. You need huge spaces to make that work, and bass sources spread some 20-30 feet apart to cover you down to 40-50 Hz. And again you'd be optimizing for a small percent of content, while giving up the modal smoothing benefits of multi-sub at all other times.

 

[Sokel]

It's worth repeating, we have the best tool on the planet for null analysis, DeltaWave, here:

DeltaWave Audio Null Comparator | DeltaWave documentation

deltaw.org

(free)

One can just compare L to R to a song and see what comes out down low, be it amplitude, phase, etc, it's that simple.

Edit: as an example it looks like this, it's the same song, the time it came out had a mono version too:



Stereo



Mono

For such an old work (David Bowie, Changes One) we can see what Lund says for artists and mains monitors.
Note the mono's perfect agreement if in doubt for the quality of files.

 

[KenMasters]

I've experimented with stereo bass in my setup. As you can see from my thumbnail, I have a sub to the outside of each front. The effect is noticeable, it sounds more "open" and initial experiments had me quite taken by the effect (as made possible by the Standard/Directional option on Denon AVRs).

Bouncing back and forth between a traditional mono bass setup with a high crossover of 150Hz, while the summed response did have a more closed in, oppressive feel, it was also more impactful/visceral and smooth. I eventually decided to stick with the summed response.

Perhaps with more subs to either side, with which I could achieve uniform bass for both left and right, along with the greater impact, I might have stuck with stereo.

Since then I have moved on from strictly manual EQ to Dirac ART and an additional two subs. I find ART provides a good mix of both attributes. It has a similarly open sound but bass remains even and impactful in the same manner as the summed response.

While ART does not provide discreet left and right oriented bass, it is uncorrelated in the sense that the filters applied to the subs for left and right differ from one another - which might explain why it sounds more open than the traditional mono approach.

 

[goat76]

It just doesn't seem worth compromising your system's optimization for only a small percent of content.

I don't know why you keep saying that, when many of us have reported that most of the music tracks we have analyzed do indeed contain stereo information all the way down to the lowest bass frequencies.

Even the post by Keith, which you quoted, had information about that, but you left that part out.

 

[3ll3d00d]

The real kicker here is that they did not supply dimensions of the test environment.

There's a pic of the IPL on P3 of

https://www.acoustics.asn.au/conference_proceedings/AASNZ2016/papers/p120.pdf

And another here https://reporter.mcgill.ca/these-are-our-brains-on-music/

It reportedly contains a 32.4 array so it's going to pretty big (I guess 5-7m across so maybe 35-50m2 range?)

It seems reasonable to assume the small HT is smaller so 4-5m wide max given that context

 

[audiofooled]

I don't know what more can be said

I would like to add a very important note that @Thomas Lund said about AE dimension having vastly varying degrees that can be identified. Highly complex and not at all black and white, as in either this or that (stereo/mono).

 

[3ll3d00d]

How do you assess whether the difference between the two channels at low frequency is a perceptible difference? Just showing that there is a difference doesn't mean it's a material difference after all.

 

[janbth]

How do you assess whether the difference between the two channels at low frequency is a perceptible difference? Just showing that there is a difference doesn't mean it's a material difference after all.

A good question and a good point.

An analysis I'd like to see is mixing the L and R channels into midband L+R (i.e. the mono part) and sideband L-R, and compare the relative signal levels of these.

 

[Sokel]

How do you assess whether the difference between the two channels at low frequency is a perceptible difference? Just showing that there is a difference doesn't mean it's a material difference after all.

You can filter the delta result at 80Hz or so at DeltaWave and listen to it (also included at the app) .

 

[Fidji]

To all those doubters who are sceptical that this phenomenon exists, I point you towards publications of peer-reviewed research that PROVES it exists. All these resources are available earlier in this thread.

First, a nice introduction to the concept is Matt Poes' article in Audioholics - Is stereo bass myth or reality?

Earlier in this thread we had Thomas Lund's study. This link takes you to an earlier post which links to a full text PDF as well as my summary. In essence: @Thomas Lund conducted an experiment with correlated and uncorrelated pink noise, and the experiment was conducted in two phases. One with headphones, and another in the free-field (speakers positioned around a listener at the top of a hill). Untrained listeners were used with a very wide age range, some elderly and maybe with hearing loss. Result: uncorrelated bass was audible.

That still leaves the question - is uncorrelated bass still audible in a listening room? @MayI posted another study, which I commented on here. Answer: YES it IS audible in a listening room. The only problem is that the authors did not specify the dimensions of the listening room (such a daft omission!) so we don't know how generalizable the findings are. But it DID show that our ability to discriminate uncorrelated bass deteriorated as the wavelengths grew longer, as you would expect.

At this point, we have multiple leading investigators - Griesinger, Lund, JJ, that McGill study, etc. with independent peer-reviewed publications showing the same thing.

The last remaining defence is that "but recordings don't contain stereo bass!". Wrong again. I examined some recordings in my collection using Audacity to split a stereo track into mono, then inverted and summed the second mono track to see if it nulls. I then examined many recordings in my collection - 100% of them had stereo bass. This experiment was repeated by @Sokel using a slightly different method, and he found the same.

All this is still rather theoretical. Maybe you're not convinced by the peer-reviewed publications. And we still don't know what dimensions the listening room has to be - it would be VERY NICE if someone did another study comparing listening room sizes with audibility thresholds (but that experiment would be extremely difficult to conduct!). Well, JJ posted some test signals so you can listen with your own system. Sorry I can't find the link at the moment, maybe he can post it again.

I don't know what more can be said.

Nice summary, thanks a lot, as scope of threads started to drift into lot of theorizing and anecdotes.

But as they say in McKinsey - "So What?"

While we have meanwhile decades of observations, lot of implications, both practical and theoretical - nothing really happened. We understand all the underlying theory, we have all the tools and the solution does not seem to be financially prohibitive. So in theory - take couple of cheap subs, place them properly, take MiniDsp, do couple of sweeps - and voila - your own personal Royal Albert Hall in 3 min. If all what you say re stereo bas information - what is the reason we do not see super simple A/B demonstrations on audio shows or AES meetings? Why all we have, are tests in unrealistic conditions?

It is even more baffling, as we are talking about highly marketable product - “envelopment” - why is nobody selling 2subs+DSP+proprietary filtering as “Griesinger Spatial Reference” to audiophiles with hefty margins?

So, yes, theoretically yes, it is proven with enough rigor to accept the phenomena really exists. Similar to nuclear fusion or superconductivity.

To move the discussion forward - we need somebody super smart like Thomas to come with white paper - “how to achieve AEV in real life conditons” - basically describing system, that will bring all the theoretical benefits into reality. Conditions can be tough - e.g. minimum room size, acoustic properties, any constraints an limitations, it can be even not very practical, you can limit usable recordings etc - but we need to see standardized working system. Until then it is just nice theory with lot of anecdotical evidence.

I am very much volunteering - dedicated room, acoustically treated, have 8 freely movable/steerable/DSPable subs, can add another 4 if it would help anything.

I will still make a prediction - small room acoustics will prevent most of the “easy” setups, and at the end of the day, after lot of effort and expense - there will be 2CH system with impeccable stereo bass. Simple A/B comparison will show, that envelopment is much worse than on upmixed material on cheap surround system. And visit to Vienna and MusikVerein [or Berlin, Amsterdam, Boston etc] concert hall will break all the illusions re how good this “stereo bass” is in terms of envelopment and how much reality it brings to your home system.

 

[Sokel]

Nice summary, thanks a lot, as scope of threads started to drift into lot of theorizing and anecdotes.

But as they say in McKinsey - "So What?"

While we have meanwhile decades of observations, lot of implications, both practical and theoretical - nothing really happened. We understand all the underlying theory, we have all the tools and the solution does not seem to be financially prohibitive. So in theory - take couple of cheap subs, place them properly, take MiniDsp, do couple of sweeps - and voila - your own personal Royal Albert Hall in 3 min. If all what you say re stereo bas information - what is the reason we do not see super simple A/B demonstrations on audio shows or AES meetings? Why all we have, are tests in unrealistic conditions?

It is even more baffling, as we are talking about highly marketable product - “envelopment” - why is nobody selling 2subs+DSP+proprietary filtering as “Griesinger Spatial Reference” to audiophiles with hefty margins?

So, yes, theoretically yes, it is proven with enough rigor to accept the phenomena really exists. Similar to nuclear fusion or superconductivity.

To move the discussion forward - we need somebody super smart like Thomas to come with white paper - “how to achieve AEV in real life conditons” - basically describing system, that will bring all the theoretical benefits into reality. Conditions can be tough - e.g. minimum room size, acoustic properties, any constraints an limitations, it can be even not very practical, you can limit usable recordings etc - but we need to see standardized working system. Until then it is just nice theory with lot of anecdotical evidence.

I am very much volunteering - dedicated room, acoustically treated, have 8 freely movable/steerable/DSPable subs, can add another 4 if it would help anything.

I will still make a prediction - small room acoustics will prevent most of the “easy” setups, and at the end of the day, after lot of effort and expense - there will be 2CH system with impeccable stereo bass. Simple A/B comparison will show, that envelopment is much worse than on upmixed material on cheap surround system. And visit to Vienna and MusikVerein [or Berlin, Amsterdam, Boston etc] concert hall will break all the illusions re how good this “stereo bass” is in terms of envelopment and how much reality it brings to your home system.

About conditions:

Lund have hinted such, old-good control rooms with flush mounted big main monitors. They certainly fall under small room acoustics, I've seen some smaller than my dedicated 80 m² room.


I quote:

Thanks to big, flush-mount monitors, recording artists have been able to hear AEV for 70 years now. Listen to even the earliest Bruce Swedien recordings of Count Basie, Oscar Peterson etc. Bruce heard it, or he wouldn’t have been able to preserve AEV so faithfully.

 

[MayI]

I don't think anyone was doubting that it exists. At least, if we are talking about the same thing--the spacious bass effect described by David Griesinger.

Most here deny it exists and like yourself, bizarrely strongly oppose the idea itself. A strange allergy to "new" evidence....to them of course. Read on..

It just doesn't seem worth compromising your system's optimization for only a small percent of content.

Evidence of compromise as you claim? Evidence of "small percentage of content"? Published papers like the McGill one would suffice, thanks.

Now for "stereo bass"...this is mostly not possible due to the length of bass waves. You need huge spaces to make that work, and bass sources spread some 20-30 feet apart to cover you down to 40-50 Hz. And again you'd be optimizing for a small percent of content, while giving up the modal smoothing benefits of multi-sub at all other times.

It's clear you either didn't read or understand the McGill paper and others above, including what JJ is saying. It's not the size of the room per se, it's the decays times, the smaller rooms/typical home construction will have higher times, but there is nothing preventing anyone from addressing that issue.

 

[Thomas Lund]

Agreed, though the link and references at bottom, shows there was interest at least 20yrs ago!

LF noise in immediate judgment of spaciousness has been practiced for more than 30 years, e.g. by Jens Blauert and David Griesinger. We use noise to specifically study basic sense of space in humans, i.e. auditory envelopment (AEV).

More than just being repeatable, LF noise is a neutral stimulus, compared to music or atmosphere, and it comes with built-in maximum inter-aural movement velocity (change rate). All relevant qualities when studying elementary sensing. To rely on brainstem physiology rather than cortical.

However, we have still just scratched the surface. More studies are needed, and with music, to connect the inter-aural dots; also when slow listening and social factors (e.g. bonding) are included.

 

[Fidji]

LF noise in immediate judgment of spaciousness has been practiced for more than 30 years, e.g. by Jens Blauert and David Griesinger. We use noise to specifically study basic sense of space in humans, i.e. auditory envelopment (AEV).

More than just being repeatable, LF noise is a neutral stimulus, compared to music or atmosphere, and it comes with built-in maximum inter-aural movement velocity (change rate). All relevant qualities when studying elementary sensing. To rely on brainstem physiology rather than cortical.

However, we have still just scratched the surface. More studies are needed, and with music, to connect the inter-aural dots; also when slow listening and social factors (e.g. bonding) are included.

Thomas, thanks a lot, really appreciate your [and others industry insiders] input here. It gives me lot of food for thoughts.

For me to digest it all [that's why I mega simplify it} - is it a correct statement:
"take two big Genelecs, flush mount them in properly treated room and you will experience AEV sensation from most of the recordings even under controlled conditions"?

or is it all much more complex and AEV is rather theoretical concept now, that still awaits real-life implementation, that is not so easy as it might seem in home conditions? And maybe this is the reason why we do not see too much controlled testing in real life on music material.

EDIT: for context - I am referring to this

About conditions:

Lund have hinted such, old-good control rooms with flush mounted big main monitors. They certainly fall under small room acoustics, I've seen some smaller than my dedicated 80 m² room.


I quote:

 

[goat76]

How do you assess whether the difference between the two channels at low frequency is a perceptible difference? Just showing that there is a difference doesn't mean it's a material difference after all.

To hear the difference in isolation, I used a DAW and set a low-pass filter at a few different crossover points, and then I used a monomaker that I could switch on and off for an instant comparison between mono and stereo.

If I remember it correctly, I used a mono recording of a bass guitar, which I duplicated and panned, one track hard left and the other hard right. A small amount of a delay was then applied to one of the channels to introduce a little bit of phase shift, to introduce a small difference between the left and right channels to mimic stereo bass.

What was very easy to hear when switching from stereo to mono, was that the bass sound collapsed to the area between the two main speakers, but opened up in the room when switching off the monomaker.

 

[Chromatischism]

I don't know why you keep saying that, when many of us have reported that most of the music tracks we have analyzed do indeed contain stereo information all the way down to the lowest bass frequencies.

Even the post by Keith, which you quoted, had information about that, but you left that part out.

Even if I did listen to a lot of music that meets that criteria, my home theater would be compromised. That's why I keep saying that

I also don't listen to the music listed here so far.

Evidence of compromise as you claim?

It's simple. If you understand the benefits of a multi sub system, using proper placement and EQ to reduce frequency response variation, you will understand what you lose when chasing this bassciousness effect.

It's clear you either didn't read or understand the McGill paper and others above, including what JJ is saying. It's not the size of the room per se, it's the decays times, the smaller rooms/typical home construction will have higher times, but there is nothing preventing anyone from addressing that issue.

It's clear that you don't understand the modal issues I'm talking about. Every frequency has a different wavelength. Study up on how that relates to distances in your room. Decay times are irrelevant to what I'm talking about.

 

[levimax]

Another way to test is take a track with "know uncorrelated bass" and "Sum bass to mono". You can do this in Audacity. You can then use Foobar2000 ABX or similar to switch back and forth level matched. See below for Audacity method courtesy of Claude AI.

 

[NTK]

Even if I did listen to a lot of music that meets that criteria, my home theater would be compromised. That's why I keep saying that

I also don't listen to the music listed here so far.


It's simple. If you understand the benefits of a multi sub system, using proper placement and EQ to reduce frequency response variation, you will understand what you lose when chasing this bassciousness effect.


It's clear that you don't understand the modal issues I'm talking about. Every frequency has a different wavelength. Study up on how that relates to distances in your room. Decay times are irrelevant to what I'm talking about.

Capably setup "stereo bass" systems should be fully adept at producing mono bass without compromise. Just a different set (or preset) of DSP parameters and signal routing. You should be able to have both easily if your system is flexible enough, just may not be both at the same time.