or Connect
AVS › AVS Forum › Display Devices › Ultra Hi-End HT Gear ($20,000+) › Burn-in: Real Or Imagined???
New Posts  All Forums:Forum Nav:

Burn-in: Real Or Imagined??? - Page 20

post #571 of 665
Hmmmm, I read part of this paper and I think I'm out on this one. I've never heard bass inside my head, any more than I've heard a 'reversal'. Interesting ideas and all, but I'm not sure how valuable it is in the real world.
post #572 of 665
Quote:
Originally Posted by sdurani View Post

Lexicon surround processors (consumer, not pro), from the mid-1990s to present. On the MC-12, for example, all the bass filtered from the left channels (front, side, rear) gets sent to the left subwoofer. Ditto the right side.

Their (optional) Bass Enhance processing keeps a 90 degree shift/lag between the two subs, which they recommend you place as far away from the centre line as possible (the listener's +/- 90 degree points) in order to maximize the effect. The reason for the low frequency decorrelation processing is because, as you and others have mentioned, there are very few recordings with true decorrelated bass.

There are valid reasons for keeping recorded bass mono. But it's nice to have a workaround (just as one would use Neural if their favourite recording is not available in multi-channel). And, like any processing, you always have the option to turn it Off if you don't like the 'bass externalization' effect.

Sanjay


Thanks. This isn't the kind of processing I would personally think of. The 90 degree phase shift is going to have a hang-time of truly astonishing proportions...
post #573 of 665
"I've never heard bass inside my head, any more than I've heard a 'reversal'."

Doesn't mean others haven't heard it....

Shawn
post #574 of 665
Quote:
Originally Posted by sfogg View Post

"I've never heard bass inside my head, any more than I've heard a 'reversal'."

Doesn't mean others haven't heard it....

Shawn

True, just like Bigfoot Just not a big deal IME, plenty of other fish to fry.
post #575 of 665
This has to be one of the more interesting discussions I have seen in the AVS.. I will write a little more about it later for now.. I am watching the Inauguration .. An Historical even that seems possible only in this great US of A...

To be back on topic... Mono bass is a compromise not an ideal...or a necessity. We now have the technology to do away with it and the results can only be better... more later...
post #576 of 665
All,

This is a most interesting thread . The particular relevence (as I see it, take it or leave it) is that consumer listening environments are a challenge. Small room acoustics (home or auto) dominate the consumer landscape.

As a room gets smaller, the modal density drops. This is usually a concern because of inadequate LF reinforcement throughout the room, that is, more bassy spots and more thin spots. There is no good solution for a large listening area in a small room as the modal density is not high enough to allow for small changes in position without a drastic change in perceived LF spectrum.

Another concern is the perceptual exposure of discrete echoes (and the comb fitering that goes with it). This guarantees drastic changes in spectrum in the mid and HF as a function of position.

The statement that one can not alter the actual modal distribuion of a room artificially is true, however, there are possible de-correlation methods that can "spread" the spectrum in such a way that the most egregious specrum errors are "dispersed". This results in an improved experience for the consumer.

This method would also promote symmetry about zero...much as a multiorder allpass filter, reducing cone excursion and required amplifier peak capacity.

The perceived spectrum is preserved at the cost of the envelope, ameliorating the Haas effect and allowing more consistent audition within small environments with listeners next to the loudspeakers (think: car).

De-correlation is a priceless tool that we've just begun to understand. As it was pointed out by Sanjay, nobody wants "swishy/flangey" and as such, there is still more work to be done. That being said, let's keep an open mind to the technology. I've personally heard it under lab conditions and it's pretty amazing.

rob r. (old f*rt that still talks to much )
post #577 of 665
Quote:
Originally Posted by jj_0001 View Post

This isn't the kind of processing I would personally think of.

Sure. But when introduced a decade ago, it's not like there was a choice of competing bass externalization processing technologies available for us consumers to pick from (not that there are now either). It may not be optimal, but it works. If you can ever get a demo, take a listen. I'm sure if someone wanted to pursue similar processing nowadays, the last ten years of research and modern DSP horsepower would yield many improvements.

Sanjay
post #578 of 665
Quote:
Originally Posted by rob r. View Post

The statement that one can not alter the actual modal distribuion of a room artificially is true, however,



Well, you certainly can't REMOVE nodes...
post #579 of 665
Quote:
Originally Posted by jj_0001 View Post

Well, you certainly can't REMOVE nodes...

Well, it's been a long time since I had to do the math, but can't you change the geometry to change the nodal distribution.
Ahh, the room is fixed you say. True, but doesn't going from a point source, to a line array, to a planer source change the acoustics. Additionally, doesn't the simple addition of point sources change the spacial distribution of nodes?
post #580 of 665
Sanjay,

Are you sure Bass Enhance is just a 90 degree shift?I thought it did a little more then that. I think there was a post about this from Jim Mueller in the old forums but they are offline now.

I could try and measure the phase changes of bass enhance with a transfer function through my FFT but I think the latency through the DSP really confuses the FFT.

Shawn
post #581 of 665
Quote:
Originally Posted by sfogg View Post

Are you sure Bass Enhance is just a 90 degree shift? I thought it did a little more then that.

I'm not sure about the exact amount of phase shift; I was just using the number (90 degrees) that Griesinger mentioned in the paper I linked to earlier: (page 27) "In general externalization can be improved by shifting the phase of the low frequencies by 90 degrees for the left and right sides of the listener." It's possible that the actual implementation was more than that and/or may have had other things going on.

I know you don't use it in your current theatre, since your subs are up front. But what did you think of it in your previous set-ups?

Sanjay
post #582 of 665
Sanjay,

I think Jim had mentioned the effect wasn't a fixed phase change but that it sort of varied the phase between the L/R bass.

"I know you don't use it in your current theatre, since your subs are up front. But what did you think of it in your previous set-ups?"

I tend to leave it turned on for music in the theater but the effect is harder to hear down there because of my frontal L/R subwoofer setup.

In my last two system when I had stereo subs directly to the listeners sides the effect was more obvious on more material.

As mentioned turning on bass enhance resulted in the bass not sounding the same as it does when wearing headphones, it sort of joined the rest of the instruments on stage. That was the externalization factor.

The other thing it tended to do is make it feel as if my side walls were sort of melting away or at least moved much further away. It changed the feeling of how I perceived the acoustic space to give the impression of a larger space.

In a nutshell it made bass sound closer to what I hear live in a hall. IME work in this area is worthwhile.

Shawn
post #583 of 665
Okay, maybe I'm not out on this just yet.

What is the threshold audibility/perceptibility of bass directionality in an anechoic environment?

Because if you look at the idea that a 90Hz wave is about 12', well, that pretty much eliminates your ability to hear/fee/ an entire waveform before you are beginning to hear/feel the reflection. Doesn't bode well for being able to sense direction.

Which, brings me back to the idea of, rather than phase tomfoolery, delayed inversion of the reflected sound - use a multi-channel setup to run full range signals to every speaker, but also send a targeted, delayed and contoured inverse signal to the other subs to null out the room additions from those areas. This would substantially tighten up the bass, reduce hang time and, of course, make everything sound overly lean, so you'd have to create a new contour. And you could create a new 'big room' overlay to an extent, by adding back in some bass reflections from the other rooms with more delay. So each full range speaker and/or sub would have several duties - create a bass note, cancel notes coming from other speakers, and then, optionally, create a new simulated room interaction over top of the old. Of course, this would require 4-7 subwoofers to pull off well, but I've always wanted to experiment with this type of design to see what it would do.
post #584 of 665
I just started reading this thread and for the life of me I can't figure out what phase shift and decorrelation have to do with Burn-in: Real Or Imagined???. Does burn-in cause phase shift or decorrelation? Or is that what you mean by "phase tomfoolery" Alimentall?
post #585 of 665
It's called thread shift, which is a well-documented audio related phenomena
post #586 of 665
Quote:
Originally Posted by Alimentall View Post

It's called thread shift, which is a well-documented audio related phenomena

Are you sure it is not just your imagination caused by the placebo effect? By the way, I was really disappointed they not only shut down but deleted the Black Box thread. Someone must have really not liked my last post on voodoophilia. Or perhaps it was one of the responses to it that I didn't get a chance to see.
post #587 of 665
Quote:
Originally Posted by faberryman View Post

Are you sure it is not just your imagination caused by the placebo effect? By the way, I was really disappointed they not only shut down but deleted the Black Box thread. Someone must have really not liked my last post on voodoophilia. Or perhaps it was one of the responses to it that I didn't get a chance to see.

I wouldn't worry too much, everyone always blames me anyway. Reminds me of my college days when I'd come back from a weekend at home to find myself mysteriously in trouble with the dean
post #588 of 665
Quote:
Originally Posted by Alimentall View Post

I wouldn't worry too much...

I'm not worried, just disappointed. The moderators sure have a heavy hand around here. I didn't shout fire in a crowded theater or anything.
post #589 of 665
Quote:
Originally Posted by Swampfox View Post

Well, it's been a long time since I had to do the math, but can't you change the geometry to change the nodal distribution.
Ahh, the room is fixed you say. True, but doesn't going from a point source, to a line array, to a planer source change the acoustics. Additionally, doesn't the simple addition of point sources change the spacial distribution of nodes?

You can't REMOVE nodes. You can damp them, etc, which is what Earl's stuff does among other things if I understand it right.

But since doubling (r more) time delay doubles (r more) the mode density, nothing keeps you from ADDING modes with simple processing that involves delays.

This is where jaxter, todd, and beezelbub seem to go astray. Nobody (I hope) is suggesting that you can disappear the natural modes in a small room. The room WILL store energy, and that's all there is to it.

A good venue will store a lot more energy, over a lot more time, with a lot more reflections and effective "distances", and get a nice, rich set of modes from that. With recording something that contains some velocity, you can capture some of that information. By using digital delays, you can quite trivially also simulate something like that process.

I vote for capturing the actual venue when such a thing exists. For synthetic venues (i.e. studio production) some digital means can do a very nice job, but of course of some "imagined" venue, rather than a real one. To date, I am not aware of any simulation that sounds as good as simply capturing 5 or 7 (for 5.0 or 7.0 rendering, the "up" and "down" parts of the mic array have been shown to be way too easy to pollute with venue noise, even though they are otherwise quite useful) channels in an appropriate fashion, which is neither coincident nor widely separated. I'm sure there are other ways, the whole issue of methods of this sort has not been even remotely explored in a systematic fashion.
post #590 of 665
Quote:
Originally Posted by Alimentall View Post

That sounds contradictory! So, you're saying that you should have only one subwoofer for all channels, right? (I can't agree with that), But then you don't want that sub in front of you because if the bass hits your ears at the same time, it comes across as mono?

To hear the mono bass effect best, use only one bass speaker, placed in front of you. Of course, no one sets up a system like this, but that's the best way to learn how to hear it.

Multiple mono subs (ie. multiple subs all playing back the same signal as opposed to decorrelated signals), if not placed correctly, will cause all sorts of frequency response problems due to interference.

Quote:
Originally Posted by sfogg View Post

I think Jim had mentioned the effect wasn't a fixed phase change but that it sort of varied the phase between the L/R bass.

My recollection is that the phase shift happens only when the program material needs it --- the bass enhance algorithm figures out if it's seeing mono bass, and effects the phase shift if necessary. I also wouldn't imagine the filter to be on/off --- there's probably some kind of smooth, IIR-like progression from 0 to 90 across the frequency range.

Quote:
Originally Posted by Alimentall View Post

What is the threshold audibility/perceptibility of bass directionality in an anechoic environment?

First things first: we're not talking about directionality, but envelopment and spaciousness. Second, I would guess perceptible since the headphone experiment described a few pages ago makes the source of the effect sound pretty obvious.

Quote:


Because if you look at the idea that a 90Hz wave is about 12', well, that pretty much eliminates your ability to hear/fee/ an entire waveform before you are beginning to hear/feel the reflection. Doesn't bode well for being able to sense direction.

Not necessarily. We are talking about interaural time difference (ITD) cues here: the difference in arrival times of a soundwave between the ears. You don't need to hear the entire thing to hear this. Further, if you believe Griesinger's theories, it's fluctuations in the ITD that actually cause this effect, so it's really two steps removed from hearing the whole waveform.

This is why the goal of a small-room multichannel bass is not to reproduce the actual modal structure of a large venue --- that is physically impossible as has been pointed out by many people already. Instead, we are seeking to reproduce the perceptually relevant effects of those modes in the small room. JJ will cringe at this, but think of it like making an MP3 of the acoustic field: let's capture and reproduce those portions of the soundfield that are perceptually relevant.

Some people mentioned 5.1 and how that .1 channel represents mono bass, but remember that the other 5 (or all 8 in most discrete 7.1 systems) channels are full range! Therefore, we have had multichannel delivery systems that can work for multichannel, discrete bass for a long time already. The .1 channel is not a sub channel --- it is a bass headroom channel, though it's been severely abused as a sub channel for many things already.

I remember asking John Eargle (RIP) at one of the first or second Surround Pros years ago about his miking techniques, and how decorrelated his bass signals were. He didn't have a working theory of stereo bass, but he said that he always tried to make sure his channels were as decorrelated as possible (ie. capturing as much info about the venue) so that he had a good chance to do the right thing in the future should some new thing show up.

--Andre
post #591 of 665
Quote:
Originally Posted by AndreYew View Post

To hear the mono bass effect best, use only one bass speaker, placed in front of you. Of course, no one sets up a system like this, but that's the best way to learn how to hear it.

Are you suggesting that we should learn how to hear something that we don't now hear so we can come up with solutions to fix the problem we never heard?
Quote:



Multiple mono subs (ie. multiple subs all playing back the same signal as opposed to decorrelated signals), if not placed correctly, will cause all sorts of frequency response problems due to interference.

That's a different issue though!
Quote:



First things first: we're not talking about directionality, but envelopment and spaciousness. Second, I would guess perceptible since the headphone experiment described a few pages ago makes the source of the effect sound pretty obvious.

I have no interest in 'enveloping bass', just directionality. I guess I'm back out on this. Too eerily similar to the dipole rear speaker argumentation where directionality wasn't important, only the ironically nebulous term 'envelopment'.
Quote:



Not necessarily. We are talking about interaural time difference (ITD) cues here: the difference in arrival times of a soundwave between the ears. You don't need to hear the entire thing to hear this. Further, if you believe Griesinger's theories, it's fluctuations in the ITD that actually cause this effect, so it's really two steps removed from hearing the whole waveform.

I don't think those kinds of cues work that low in response, hence the need to goose it up with phase bending.
Quote:



This is why the goal of a small-room multichannel bass is not to reproduce the actual modal structure of a large venue --- that is physically impossible as has been pointed out by many people already. Instead, we are seeking to reproduce the perceptually relevant effects of those modes in the small room. JJ will cringe at this, but think of it like making an MP3 of the acoustic field: let's capture and reproduce those portions of the soundfield that are perceptually relevant.

He's not the only one that just cringed! Yeah, this is all mildly interesting and all, but I think it's another one of those attempts at solving an unnoticeable problem, so I think I'll move on unless there's something more significant to be discussed.
post #592 of 665
"Are you suggesting that we should learn how to hear something that we don't now hear so we can come up with solutions to fix the problem we never heard?"

So you can't hear the difference between bass reproduction in a hall and in your home?

Shawn
post #593 of 665
Quote:
Originally Posted by sfogg View Post

"Are you suggesting that we should learn how to hear something that we don't now hear so we can come up with solutions to fix the problem we never heard?"

So you can't hear the difference between bass reproduction in a hall and in your home?

Sure, but I don't associate that with the subject at hand.
post #594 of 665
Quote:
Originally Posted by Alimentall View Post

I have no interest in 'enveloping bass', just directionality. I guess I'm back out on this. Too eerily similar to the dipole rear speaker argumentation where directionality wasn't important, only the ironically nebulous term 'envelopment'.

Is 'envelopment' really a nebulous term? Dr. Toole discusses it -- LEV or 'listener envelopment' -- and its prerequisites at length over the course of his recent book, along with ASW (apparent source width), as two signifiers of 'good sound' to many listeners under tested conditions. I doubt a 'nebulous' effect would merit such attention.

Btw, I enthiastically second the motion to split the decorrelated bass discussion off into a new thread, with a title that might actually be informative to browsers, compared to 'Burn In'.
post #595 of 665
Quote:
Originally Posted by AndreYew View Post

My recollection is that the phase shift happens only when the program material needs it --- the bass enhance algorithm figures out if it's seeing mono bass, and effects the phase shift if necessary.

I remember something like that also. THX does something similar with their Adaptive Decorrelation processing, though only for the surround channels: the more mono the surrounds, the more decorrelation; the more stereo the surrounds, the less decorrelation applied.

BTW, I dug up the old interview Philip did with Griesinger (relevant parts below). Maybe that's where your recollection comes from?
Quote:


Bass Enhance is there because most of the people who mix recordings, of popular music anyway, mix the bass in monaural, into the center. What Bass Enhance does is route the mono LF to the sides, and put a constant phase shift between the two channels. If you have music that was recorded with stereo bass, with decorrelated low frequency reverberations for example, then Bass Enhance doesn't do much. The two channels have no correlation, there is no constant phase relationship between them, so Bass Enhance doesn't do anything. The phase shift is only there when you have a very constant monaural bass signal, and then it changes the way that sounds.

Informally, I do experiments with it in demos all the time, and I find that some people don't like it and some people really do. It has a lot to do with what your expectations are for bass. If you're a recording engineer that sits between two tower loudspeakers in a dead room 8 hours a day listening to the kick drum go boom-boom-boom hitting you in the chest, you may not like Bass Enhance because it changes the situation completely. It's a completely different sound. Those people are more likely to say "There's something wrong with the bass" and "It's phasey" And of course they're right.

In the several of the rooms I've demo'd it in, it is quite different. In some rooms it makes a much more dramatic difference than others. And it always makes bass sound less loud, because it's taking energy out of the very center where you probably put yourself and sort of spreading it more around the room. That will make it softer in the center. But more enveloping. I tend to like it, because to me it makes the bass sound more like a live performance. If that's your reference, then you'll probably like it. If it's not your reference, you may not. So it does have to do with expectations I think.

Keep in mind that this was a first step. As I mentioned to Shawn, ten years have passed since then. I don't know if there has been more recent research on the subject but, combined with processing power of modern DSP chips, somebody should be able to come up with better processing for low frequency envelopment/externalization.

Sanjay
post #596 of 665
Did the Lexicon device actually indicate whether its input has mono bass or not (e.g, does it tell the user when 'Bass Enhance' is actually in use, the way some AVRs show when LFE content is actually present in the ".1" channel)? Asking because there's a bit of controversy on one of the audio groups I read (frequented by recording engineers), over how common monoaural bass was/is in commercial recordings. Reports from Lexicon users might provide objective data.
post #597 of 665
Quote:
Originally Posted by krabapple View Post

Is 'envelopment' really a nebulous term?

A little nebulous to me. Or maybe arbitrary? Especially when you're talking about 'bass envelopment'. I'm not saying that it isn't a cool trick, but it just doesn't interest me. I only mention this because I thought we were having a somewhat different discussion at first.
post #598 of 665
Quote:
Originally Posted by krabapple View Post

Did the Lexicon device actually indicate whether its input has mono bass or not (e.g, does it tell the user when 'Bass Enhance' is actually in use, the way some AVRs show when LFE content is actually present in the ".1" channel)?

Unfortunately not, though I agree it would be useful info. Bass Enhance decorrelation only has two settings: Off and On. So the user always knows whether it's in use.

BTW, what sort of display function would you use to represent how mono or stereo the bass is? Sliding scale, from M to S? Some sort of threshold, below which it would say Mono, above which it would say Stereo? Or something else?

Sanjay
post #599 of 665
Quote:
Originally Posted by faberryman View Post

I just started reading this thread and for the life of me I can't figure out what phase shift and decorrelation have to do with Burn-in: Real Or Imagined???.

Would be great if the thread was split, from vague memory back somewhere when someone asked earle to join in?

It is interesting to me as I have been following the threads on diy discussing earles distributed subs philosophy, and new stuff related to that is coming up here.

That is great cause it is a different set of guys here so new and differing points are coming up.

I enjoy threads that meander and change, but do feel this has the potential to bring up stuff that should stand alone.

Unltimately I'd love to get enough pointers and starting points to throw a few subs in and try the enveloping bass trick Sounds interesting.

Once I mucked about with distributed subs, and at one time I recall that I played them out of phase somehow, can't remember the specigics now. But I do recall it was rather pleasant and interesting. certainly not exactly what is being discussed here, but elements of it I wonder.

Ultimately it didn't work, but now this has come up perhaps if that type of effect (would it be in the same family as decorrelation?) were limited to helper subs it could be useful. (helper meaning subs only playing low stuff but specifically to provide only the 'enhancement' effect)
post #600 of 665
Quote:
Originally Posted by AndreYew View Post

This is why the goal of a small-room multichannel bass is not to reproduce the actual modal structure of a large venue --- that is physically impossible as has been pointed out by many people already. Instead, we are seeking to reproduce the perceptually relevant effects of those modes in the small room. JJ will cringe at this, but think of it like making an MP3 of the acoustic field: let's capture and reproduce those portions of the soundfield that are perceptually relevant.

Errr? I'm not too likely to cringe, especially since that's what I'm proposing we all should do.

We'll never EVER reproduce the room analytically. Perceptually, on the other hand, is entirely feasible, but not with 1 or 2 subs.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Ultra Hi-End HT Gear ($20,000+)
AVS › AVS Forum › Display Devices › Ultra Hi-End HT Gear ($20,000+) › Burn-in: Real Or Imagined???