Does 'envelopment' exist? Can it be measured?

[goat76]

So we are in the realm of shuffling around subs for anecdotal, subjective, sighted listening without direct comparison in search for a "goosebump factor"?
The way I understand Toole's research, the chances of getting anything reliable and viable out of this are rather slim.

I don't have any other suggestion/approach on how others in the thread can test this for themselves, if they want to decide if they prefer stereo bass or not. As it doesn't seem to be an easy task to do this at home in a more scientific way, unless someone sets up two separate bass systems, where one is optimized for stereo bass, and the other is optimized for mono bass, which then can be blindly switched for an A/B comparison, I don't know what else can be done.

 

[Fidji]

2.2? How? LFE channel is one signal channel - mono. If you think .2 is stereo, no wonder it doesn't work for you. I admit was fooled for a while too. DTS:x pro I think has two individual LFE channel. The LFE channel is a potential problem if you do it wrong.

But yeah, not many can enjoy flush mounted systems, but 50 years? Genelec produces monitors for flush mounting still, so... has it gone bad since, or is it still the optimal solution?


I guess we don't have the same preferences, which should be OK. I have both, great stereo bass AND great multichannel bass. Technically there are shortcomings of multichannel systems, and particularly with the LFE channel. I'd rather it wasn't there, but it is, so we need to deal with it.


I agree, the understanding on the production side is limited, and some use LFE as "boom channel", just to accomodate people with small speakers and a subwoofer. I'd stay away from that practice, and rather let users distribute LF as they see fit from a 5.0 recording. But even from the late 50s there are great live-in-studio recordings that take on a new character when not summed on full range systems.

View attachment 528026
L+R+2x subs average @MLP. Not too shabby without room correction? I'm guessing the ton of bricks I have in mind will fix the 38Hz mode and 90Hz null .

Here's an interesting track, from @amirm playlist

Is it the same in mono, you think? Upmixed? I doubt anyone will say so. The channel play is an artistic expression, which I for one, would not like to destroy.

My bad - I used 2.2 as some people do here. I am fully aware that my 9.”8”.7 system is in reality just bass managed 9.1.7. You are correct we are talking about 2.0 systems, technically, just wanted to show a difference to system with mono-bass.

Measurements - no idea, if good or bad. I was not thinking about this type of measurements ;-). You can probably agree, that you could get the same fr. response from L-R+summed mono bass. Waterfalls or other charts comparing setup with the same FR and different bass routing would be more interesting and somebody smarter than me can come with some additional ideas as what could be “best charts to show AEV superiority”

it might be a bit of a misunderstanding - I also do have good stereo bass, if I want, and can sum up to mono at any LF. I just think that crossing over to Waveforming summed mono bass at 60Hz for 2CH listening session is the best option. And it has been tested …….. - on Billie Eilish ;-). Big diff on X0 at 100Hz vs “stereo-bass”. still nicely audible 80Hz but, no meaningful “affective touch” difference at below 60Hz, although I would probably wager to get DBX right at 50Hz too in my room [and probably loosing ;-) ] This seems to be in line with existing research, that shows, that everything that people seem to hear in their reverberant room at frequencies around and below 50HZ is “golden ears syndrome” [or result of colorations in bass area]

 

[Fidji]

As we're talking theoretically, there's other options as well, passive treatment all the way down, soffit-mounting, a ground-up construction of room, etc.

I don't include cost or hardship to those, but they are the best options out there, no matter the set-up of the bass.

That's about convenience or cost vs best possible, nothing more.

Still talking theoretically ;-)

Once I would painstakingly and with great cost finish my “Most affectively touching and enveloping stereo system in the world” I would continue to give the same treatment to surrounds and heights and then use LARES emulator to get ultimate enveloping rendition of stereo recording of Billie EIlish. [or maybe something else finally]

 

[dagfinn]

Measurements - no idea, if good or bad. I was not thinking about this type of measurements ;-). You can probably agree, that you could get the same fr. response from L-R+summed mono bass. Waterfalls or other charts comparing setup with the same FR and different bass routing would be more interesting and somebody smarter than me can come with some additional ideas as what could be “best charts to show AEV superiority”

I could certainly achieve a flatter line in several ways, but the price is too high, and I don't hear any improvements. And as you know, move the mic and things change. I'd rather fix the room, that's where the 1,2 tons of bricks come in . Of course I could post more graphs, RT60 and waterfalls, whatever, but I'm not out to prove my stuff is better than yours, just that it doesn't have to be very complex to have something decent, and without summing to mono. It just has to have some power, I think, and two full range channels. If you listened to the track I posted, the LF L/R pulses are way below 60Hz, and when I switch to mono by pushing the button, the track changes character, much more than a little uneven FR would do. You may think it's ok - I don't. To me, the L/R play is earcandy (and body sensation when played loud), and it goes very flat when summed. Try the track in headset, then compare to summed.

 

[olieb]

As it doesn't seem to be an easy task to do this at home in a more scientific way, unless someone sets up two separate bass systems, where one is optimized for stereo bass, and the other is optimized for mono bass, which then can be blindly switched for an A/B comparison, I don't know what else can be done.

Hm, some posts above it seems you were not asking for so much optimisation to have a comparison.

It's not even important if the mixing engineer heard it or not in his studio speaker setup situation, as long as he just leaves it without summing the lower bass to mono.

But notwithstanding, an experiment like the one you are proposing has been done at Harman, some time ago.
This is the quote from Toole about the experience and result in https://www.audiosciencereview.com/...ith-major-audio-luminaries.62951/post-2294071

Years ago David Griensinger mounted a test of what might be possible in a "stereo bass" configuration in a relatively dead listening room at Harman. A group of highly experienced and opinionated listeners experienced a protracted experiment of variables, including digitally concocted signals and many musical selections. We heard some spatial effects, but all were in the subtle end of the scale and there was no benefit from bass management crossover frequencies below about 80 Hz. David appeared to be more sensitive than some of the others, and this might suggest a learning curve. He has written that the effect is audible in concert halls down to 60 Hz. You, Thomas, have said 50 Hz, I believe. So it is agreed that there is a low frequency limit, as is likely associated with the spacing of the ears. Obviously, a real test should begin outdoors - most anechoic chambers would not qualify. Once a baseline for audibility is established, add reverberation and vary the crossover frequency to see at what points the desired auditory envelopment is degraded.

Audibility of the goosebumps seems not that clear cut, at least not the part that is evoked from LF below 80Hz in small rooms.
The problem of all these sighted, enthusiastic tests is, that one all too easily falls prey to all sorts of biases. The brain just hears all too well what it wants to hear, if you give it the opportunity.

 

[Fidji]

Correct, this issue has been that way on forums for a long time

My bad, honestly, I am not combative here. I usually run my english posts on forums via Gemini with prompt “is this statement comprehensible for native speaker in upper quartile of verbal reasoning skills and adequate knowledge of topics discussed?”. I am fully aware that my written English is not perfect. This time I did not, as I am enjoying sun overlooking spring in Tuscany and it makes me complacent ;-).

So let’s start again, as it would be probably ineffective to get through the whole discussion one more time:

Statement:
“In absence of controlled testing of “AEV” phenomenon in reverberant small acoustic spaces on real musical musical material [as opposed to test tones], all the statements from either hobbyist or industry insiders are anecdotal at this stage and irrelevant for scientific purposes”

Trying to make it as binary as possible and easily disprovable by demonstrating existence of relevant research.

 

[goat76]

Hm, some posts above it seems you were not asking for so much optimisation to have a comparison.

Hey, I'm just trying to figure out a more reliable way for you who is asking for it.

I'm not too concerned about it, as I personally find it good enough to just listen specifically for the spatial changes that occur when shifting between mono and stereo bass, using a mono-maker in my DAW. Using my isolated stereo bass test signal of a deep playing bass guitar, it is easy to hear that the bass surrounds me when listening to it in stereo, while the bass sound shrinks down to the area between the subwoofers, which are positioned right on the outside of each main loudspeaker.

But notwithstanding, an experiment like the one you are proposing has been done at Harman, some time ago.
This is the quote from Toole about the experience and result in https://www.audiosciencereview.com/...ith-major-audio-luminaries.62951/post-2294071

I have also read about that test, but I have no idea why some of them could barely hear it in that particular situation. I suggest people set their subwoofers up in a stereo configuration in their own listening room and try for themselves.

Audibility of the goosebumps seems not that clear cut, at least not the part that is evoked from LF below 80Hz in small rooms.
The problem of all these sighted, enthusiastic tests is, that one all too easily falls prey to all sorts of biases. The brain just hears all too well what it wants to hear, if you give it the opportunity.

It was someone else who mentioned "goosebumps". I only use that word as a half-joke, and I thought everyone would get that.

No, I don't get goosebumps that easily. I hear the increase in envelopment, as in the sensation of being surrounded by the sound, a diffuse type of sensation that enlarges the sense of space around me in the listening position. When I tried setting up my subwoofers in a stereo configuration the first time five years ago, I didn't have any expectations at all. The thing I was testing was trying to get the subwoofers to play physically in phase with each of the main speakers, to hear if that made a difference, and the enhanced sense of space just happened to come with the package.

Can you tell me about your own tests of setting up your system for stereo bass reproduction?

 

[olieb]

I'm not too concerned about it, as I personally find it good enough to just listen specifically for the spatial changes that occur when shifting between mono and stereo bass, using a mono-maker in my DAW. Using my isolated stereo bass test signal of a deep playing bass guitar, it is easy to hear that the bass surrounds me when listening to it in stereo, while the bass sound shrinks down to the area between the subwoofers, which are positioned right on the outside of each main loudspeaker.

But how do you know that this comes from "auditory envelopment"? And not from something else? It might just be a room mode being excited more strongly from a single sub that invokes that feeling. In that case it would be an effect of (uneven) FR.

It was someone else who mentioned "goosebumps". I only use that word as a half-joke, and I thought everyone would get that.

Well, you used it, I used it. And there seems a notable lack of anything tangible about what "AEV" is or how to recognise it. And no, "You will know, when it is there" doesn't cut it. So, actually I kind of used it as a full-joke.

Can you tell me about your own tests of setting up your system for stereo bass reproduction?

I do my critical listening mainly binaural with APLVIrtuoso [binaural room simulation] or in the concert hall. In bass it is certainly better than anything I could ever achieve in a real room.
I can switch between mono and stereo bass in the DAW too, but with close to no difference in FR.
If I use a (bass-mono) crossover at 80 Hz (or lower), I have to search for quite a while before I find passages in music where I even get a hunch of hearing a difference. I did not yet bother to set up a blind test, chances are that the hunch will go away in that situation. And of course I am listening with "stereo bass" as switching to mono is an artificial add on. I did this only for checking for myself the strong claims about "AEV" from stereo bass in small rooms, that are being made.
But maybe my hearing is broken.

 

[MayI]

“In absence of controlled testing of “AEV” phenomenon in reverberant small acoustic spaces on real musical musical material "

As JJ and Lund have repeatedly pointed out, the vast majority of researchers are oblivious to the effect at LF. There has been next to zero research.
One of the few currently accessible due to not being on broken AES site https://audioroundtable.com/misc/icad2005F96.pdf

The cabinets housing this 3-way loudspeaker system, which were never moved between conditions, were placed on the floor, with the low-frequency
driver (15-in diameter) at a distance of 200 cm from the listening position. These subwoofers, which were never moved between conditions, were placed on the cement floor, with the driver (15-in diameter) at a distance of 200 cm from the listening position.

So it's possible in terms of dimensions of source from listener in a domestic sized room, but not possible with typical domestic type room decay times.
Thus controlling/lowering decays times in the room at LF 40-90hz must be done, unlike the majority of tests that were done and very little/no effect found. Classic GIGO scenario. The impetus for such research, other than Lund, seemingly no longer exists.

 

[Duke]

@Thomas Lund, to give you an idea of where I'm coming from, whether or not envelopment exists or can exist in a home audio context is not a question I would ask, despite my inability to convince a skeptic of its existence.

Maybe you can help me understand something mentioned in the McGill University paper... and, apologies if this is something you have already explained elsewhere!

Here's the paper (not asking you to click on it; just posting the link for completeness): https://audioroundtable.com/misc/icad2005F96.pdf

Let me quote a couple of passages, bolding mine, and then I'll get to my question:

"... differences between correlated and decorrelated audio signals reproduced by laterally positioned subwoofers are maintained at the listener’s ears, though to a reduced degree, and are not obscured in the presence of reverberation. Though the signals presented at low correlation levels are effectively re-correlated at the listeners ears (by virtue of the absence of interaural level differences), the interaural phase differences that are still present for laterally positioned subwoofers enable listeners to hear the differences between the decorrelated and correlated subwoofer signals...

"... it was concluded that this discrimination was enabled by the interaural phase differences that were presented at the listener’s ears"

My own blind-but-not-thoroughly-controlled testing arrived at a similar conclusion about the ear's ability to hear interaural phase differences at low frequencies.

Anyway, here is my question:

How is this possible? The path length around the head from one ear to the other is such a tiny fraction of a wavelength at bass frequencies that, intuitively, it seems (to me at least) impossible that we could detect interaural phase differences in the bass region. Yet if I understand the paper correctly, our ability to detect interaural phase differences is more robust than our ability to detect other interaural differences at very low frequencies. So... what is the mechanism by which the ear/brain system detects these low-frequency interaural phase differences?

 

[olieb]

How is this possible? The path length around the head from one ear to the other is such a tiny fraction of a wavelength at bass frequencies that, intuitively, it seems (to me at least) impossible that we could detect interaural phase differences in the bass region. Yet if I understand the paper correctly, our ability to detect interaural phase differences is more robust than our ability to detect interaural level differences at very low frequencies. So... what is the mechanism by which the ear/brain system detects these interaural phase differences?

Maybe you find an answer here:
"Auditory localisation of low-frequency sound sources"

https://aaltodoc.aalto.fi/server/api/core/bitstreams/e43fa993-e3c9-4784-a6c2-05a292aab188/content

It is not so much the phase, but ITD, that is detected by the hearing system. And ITD is constant, independent of wavelength. It is just (a bit) harder to extract from the audio stream as zero crossings occur at longer intervals and detection might need more spl, too.

 

[Duke]

Maybe you find an answer here:
"Auditory localisation of low-frequency sound sources"

https://aaltodoc.aalto.fi/server/api/core/bitstreams/e43fa993-e3c9-4784-a6c2-05a292aab188/content

It is not so much the phase, but ITD, that is detected by the hearing system. And ITD is constant, independent of wavelength. It is just (a bit) harder to extract from the audio stream as zero crossings occur at longer intervals and detection might need more spl, too.

Thanks for the link.

If I understand correctly, under anechoic conditions interaural time delay (ITD) differences in the ballpark of 150 microseconds (.15 milliseconds) were detectable with pink noise bands down to 31.5 Hz (the lower limit of the testing). I can't help but think that's amazing. If my arithmetic is correct, this corresponds with a path length difference of about two inches!

 

[thcdru2k]

There is a little confusion about routing. If L/R (and LFE) are summed to a single mono sub signal, any interaural differences in the LF is lost. If it is distributed per channel those differences are preserved. That kind of routing isn't typically exposed in standard AVR bass management.

 

[Mort]

There is a little confusion about routing. If L/R (and LFE) are summed to a single mono sub signal, any interaural differences in the LF is lost. If it is distributed per channel those differences are preserved. That kind of routing isn't typically exposed in standard AVR bass management.

right which effectively rules out any surround system. It'd have to be 2 channel right?

 

[thcdru2k]

right which effectively rules out any surround system. It'd have to be 2 channel right?

I think you would still preserve those interaural differences if that is how your setup is routed. You would need an separate decoder with pre-outs where LFE is disabled. Than you would need a way to route the channels to desired subwoofer. However it also would depend on how bass is routed when stereo music is upmixed so that I am not certain.

 

[Mort]

You would need an separate decoder with pre-outs where LFE is disabled.

Does that really work? Don't surround recordings often come with a mono LFE track? I'm admittedly not strong in this area...

 

[thcdru2k]

Does that really work? Don't surround recordings often come with a mono LFE track? I'm admittedly not strong in this area...

Sorry I worded that poorly i mean the AVR subwoofer output not the LFE channel itself. My concern is that when speakers are set to small in a typical AVR, the low frequency content from the left and right channel is redirected into the subwoofer output, where it is summed with the LFE channel.

The left, right, and even surround/atmos channels need to remain full range and than crossed externally into separate bass paths.

 

[Oddball]

Not as many as you'd like to insinuate. Apple Music, the biggest supplier of streamed atmos,

"Apple Music focuses on native, object-based Dolby Atmos mixes rather than user-side "upmixed" stereo, requiring creators to submit dedicated Atmos files. While Apple policy prohibits artificial upmixing from stereo for submission, some user reports suggest varied quality in how spatial audio renders, and others suggest some "upmixed" content might pass through."


Me too. I thought about hanging 4 more JBL 3600s from my ceiling but fear the roof structure wouldn't support them.. I really would like to hear what some have done with a swarm of Genelecs in the studios like Steven Wilsons. As I've said here often, it's more important than some would like you to believe to have excellently matched speaker for all channels, most specially for the bed channels. Expensive yes, but still crucial none the less.


Amen

Time to change the roof structure. Adds to the property value as well. Although going too big with Atmos is a bit of a waste as not much in there based on the verified analysis over at AVS Forum.

And world is not going to change overnight. L and R get love, center is a bastard and other channels are unwanted children. Obviously pretty wrong view but that's how it works for the most part.

 

[Oddball]

I could certainly achieve a flatter line in several ways, but the price is too high, and I don't hear any improvements. And as you know, move the mic and things change. I'd rather fix the room, that's where the 1,2 tons of bricks come in . Of course I could post more graphs, RT60 and waterfalls, whatever, but I'm not out to prove my stuff is better than yours, just that it doesn't have to be very complex to have something decent, and without summing to mono. It just has to have some power, I think, and two full range channels. If you listened to the track I posted, the LF L/R pulses are way below 60Hz, and when I switch to mono by pushing the button, the track changes character, much more than a little uneven FR would do. You may think it's ok - I don't. To me, the L/R play is earcandy (and body sensation when played loud), and it goes very flat when summed. Try the track in headset, then compare to summed.

I'd be really interested if you posted more graphs. You should prove that your stuff is better otherwise it is assumed that it is not. Your definition of flat might not be in line with what others perceive as flat.

What changes the character and what you hear from it is unfortunately the only thing that "supports" the sacrifise. With all respect to all the great minds involved, this whole concept is very difficult to implement in practical situations - which I pointed out several times in the thread, but no one seems to care at all.

 

[john61ct]

right which effectively rules out any surround system. It'd have to be 2 channel right?

Wut?

2-channel source material yes

can be upsampled to 5 or 7 full-range channels

Then the low frequency content can be split out

either within the stereo source

or within the output channels.

My co-located stereo subs are driven from the Front L/R

while Centre channel has its own co-located sub

plus mono-fed subs more freely located to tackle room modes

These can be fed by the mono-LFE signal or from a mono-summed version from any of the stereo pair outputs

or from a blending of those two mono LF sources.

None of the above necessarily having anything to do with any proprietary AVP / AVR or bass management "products"