The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Your opinion about room correctional software Virtual Instrument Plugins
Old 14th June 2010
  #61
Here for the gear
 

My hope was that a thread like this could be about what to do and how to best combine DRC and acoustic products to achieve certain effects.
Old 14th June 2010
  #62
Gear Addict
 

For those who mention that room correction can be the icing on the cake once a room has a reasonable degree of baseline treatment (such that the correction software needn't work way too hard): I'd be curious to know what information folks would look at and then say, "yup, correction software might add something here." e.g., frequency response within +/- X db, or whatever the case may be in terms of criteria they have in mind. Would be useful to have some benchmarks in mind......
Old 14th June 2010
  #63
Lives for gear
 
Bjorn Omholt's Avatar
 

What you guys don't know is that Midgard Audio has a good deal of experience with room correction and has set up some very very good sounding systems with especially DEQX. So when this guy expresses pitfalls and weaknesses with sound correction, it has a lot of credibility. At least to me.
Old 14th June 2010
  #64
Lives for gear
 

I feel like pro-physical-treatment people who say, "and then of course you can use room correction software" are just throwing the pro-electronic-room-correction people a bone. If room correction software can't be used by itself to correct the response of a room, saying that it can be used once the room is well-treated physically seems to be saying, "it can be used once it's no longer needed."
Are there actually any room response problems that are left over from physical room treatment that [only] software can correct?
It seems that it's possible to create a room that is acoustically perfect (e.g., completely nonreflective to within some tight tolerance from 20-20k, if that's what you're going for). I can't think of any reason software correction would be needed in an acoustically perfect room.
Old 14th June 2010
  #65
Lives for gear
 
Weasel9992's Avatar
 

Quote:
Originally Posted by Brainchild View Post
I feel like pro-physical-treatment people who say, "and then of course you can use room correction software" are just throwing the pro-electronic-room-correction people a bone. If room correction software can't be used by itself to correct the response of a room, saying that it can be used once the room is well-treated physically seems to be saying, "it can be used once it's no longer needed."
Are there actually any room response problems that are left over from physical room treatment that [only] software can correct?
It seems that it's possible to create a room that is acoustically perfect (e.g., completely nonreflective to within some tight tolerance from 20-20k, if that's what you're going for). I can't think of any reason software correction would be needed in an acoustically perfect room.
Yes it can be helpful, but only when the software doesn't have to "work" as hard to effect the solution. It's icing on the cake, as it were. I mean, of course it wouldn't be useful in an acoustically perfect room.

Frank
Old 14th June 2010
  #66
Lives for gear
 

Well, that's what I keep reading, but there's a huge gap between that general claim and any specifics anyone has posted or linked to. Here are some quotes from this thread and linked articles:

"Bottom line is no mystery: if you put a great speaker in the proper location, maximize every possibility in room geometry and acoustical finishes; then such DRC (digital room corrections) can yield a worthy (even very worthy) benefit."
-Jeffrey Hedback

"...there are many things that modern electronic techniques can do to make a good system sound even better, when the room's acoustical issues have first been properly addressed."
-Russ Berger

"...I always go back to physics and balance, you need to treat the room first passively (acoustic materials) then correct the remaining set up and room anomalies electronically (active room correction)."
-Jeff Smith

[re: Audessey]
"Now will it make a poor room design sound great? NO! Will it give a good room that little bit of icing on the cake? Yes!"
-Mulmany

"As a starter, let me say that I don't regard digital sound correction as an alternative to acoustic treatment. You need both to achieve the best possible sound quality. Digital sound correction is not an alternative to good room acoustics when the best possible sound quality is desired. And good room acoustics isn't an alternative to good DSP either. You need both to maximise the sound quality."
-juicehifi


As far as I can tell, the only "correction" anyone is willing to claim with certainty will actually solve a problem is speaker response correction. One other possible point of agreement is the electronic removal or significant reduction of low frequencies that are very unevenly reproduced in the room. Apart from those things, is there anything that electronic room correction even does in a "pretty well treated" room? It sounds like the reality is that physical treatment is not only an alternative to good DSP; there's no such thing as good DSP for what physical treatment does.
Old 14th June 2010
  #67


Think of it like this:

The room is a road and the DSP is improved, computer controlled suspension (EQ would be standard suspension...). The best thing to do is fix the road so it is smooth. Once the road is pretty good, a DSP can help smooth out the ride.

Of course, fixing the road is harder and may be beyond your control.



-tINY

Old 14th June 2010
  #68
Lives for gear
 
medearis's Avatar
 

Quote:
Originally Posted by tINY View Post

Of course, fixing the road is harder and may be beyond your control.
I never have a "solid room" to record in, as I'm a 'remotester', you can say.
Just about every room I've ever been into has had some weird kinks, and these are rooms of all types. Some, even in recording studios I've been to (rather rare case though)!

Since, in the remotester's world, control over the room is usually VERY limited... I'd love to have some reliable software just to find some not-so-obvious trouble freqs. There's always the obvious sounds, and with savvy EQ'ing, you can eliminate 'em. I believe that something to point out a few areas your ears over-looked is never a bad thing...

I definitely see the use of the tool, but my use is definitely not as in-depth as some of yours, I'm sure!

Old 14th June 2010
  #69
SAC
Registered User
 

This thread has outlived its usefulness - so much so that we have come back full circle to the point where, despite the fact that "room correction" software cannot fix the room, that while some acknowledge that they cannot treat the room, they still desire a bit of software that can identify the problems and would:

"love to have some reliable software just to find some not-so-obvious trouble freqs. There's always the obvious sounds, and with savvy EQ'ing, you can eliminate 'em. I believe that something to point out a few areas your ears over-looked is never a bad thing.."

Restating the desire over and over will not change the fundamental underlying issues and limitations of the tools.

And yeah, we DO have such tools in acoustical analysis.

And despite the fundamentals of acoustical physics being discussed ad nauseum - apparently without much positive result - you cannot eliminate non-minimum phase issues caused by direct and reflected signals via "savvy EQing"or un-savvy EQ! ...With software or without! And wishful thinking does nothing to change the fact.

It's time for this notion that regenerates like a severed Hydra's head, to finally die!

The existence of tools indicating the problem does not change the underlying physics which defines the nature of the problem and what is required to remediate them!

If only folks would spend half the time they expend looking for shortcuts on simply learning and understanding the fundamentals. But then, in modern science, unlike the plethora of other 'well understood' scientific disciplines, folks still want to attribute to audio a mystical nature that renders science unable to understand the fundamental behavior of the underlying physics and psychoacoustics. But folks, its not the science that is having difficulty grasping the fundamental behavior of acoustics - but rather those who look for solutions without fully understanding the problems.


But gee, if we only had a software tool that would magically fix the problems without our having to understand them or to do anything...

Personally I would be ecstatic if there was only a 'tool' to rectify such unproductive notions!
Old 15th June 2010
  #70
Here for the gear
 

Thanks BWO, now I am curious about who you are


It seems to me as if there are two different "discussions" here. Some people tends to be talking about room correction as full impulse response correction with FIR-filtering. Others seems to be including both FIR and IIR (EQ and phase correction only) into the term "room correction". Personally, I would include both and I would also claim that if the room is good, there is no obvious need for FIR filtering at all. The remaining ambience should blend naturally into the sound.

I also would like to point out that despite the name "room correction", it is not given that the problem(s) it handles is solely connected to correcting room related issues. I like to say that if the tonal balance is far off, it is no need to focus on any other problem until that is solved. A room correction could be very effective for this matter as well.

If a loudspeakers dispersion pattern is made to fit a certain type of room, and the acoustic treatment makes the room perfect, it is still not given that the acoustic treatment fits the speakers energy response, even if the frequency response is within 1-2-3dB.

Another issue is the listening distance. When listening at around 3 meters, most speakers behave pretty well, but speakers do have different dispersion patterns, and when you move away from the speaker, it can and pretty much allways will affect the tonal balance of the speaker. ADAM has a knob on many models to correct for this.

Another loudspeaker related issue is the room gain. Both placement and room size does not automatially fit all speakers tonal balance. This is not necessarily something you fix with room acoustics. However, it is a partially room related issue.

One of the most interesting issues in this field (IMO) is the loudspeaker dispersion pattern. Those with live sound experience would probably know the effect of a cardioid pattern subwoofer. Also the difference between a line array and a normal speaker, and, within hifi and monitoring, the differences between horn, normal dispersion, dipole and things like that.

I would say the ideal situation for optimum acoustics is to include:
- DRC (some kind of processing to correct room and equipment related issues)
- Dispersion pattern
- Physical room acoustics
...and make them work together.
Old 15th June 2010
  #71
SAC
Registered User
 

We have mentioned speaker issues that you mention many times, as they are VERY important in regards to how the speaker functions within the space!

And there seems to be an even greater resistance to the gaining of traction in this area than in the role of the room itself!

Heck, simply look at how many are worrying about bass traps and first order sidewall and ceiling absorbers who STILL have the speakers positioned on the bridge of their gloriously large work surface thus creating diffraction and surface reflections!

In fact, to take the issue a step further, the very same measurement techniques that we have advocated for room analysis (primarily the ETC and variations within the time perspective) provide incredible insight into the evaluation of speakers and - often - the ability to correct many anomalies that SHOULD have been incorporated into a properly designed (or selected) speakers!

In this respect, I cannot agree more! In fact, this is a corollary to the proper design and analysis of an acoustical space, and is fundamental to the design of such acoustical response models as the LEDE concept. The proper selection of the speaker system is as important as the design of the surrounding space!

DSP ( and simple signal alignment with respect to time) has many applications of value to offer the direct processing of the signal! This IS a proper and productive area for DSP and associated technologies, as it is at this level that we can actually effect the alignment of the various component signals into a minimum phase signal where various processing can be effectively performed in a constructive manner.

As as much as we are struggling to make many aware of the basic relationships of the direct to reflected signals within a bounded space, we can, and do need to drill down to examine the 'lower' order of magnitude response of the source itself. (Each level of focus can be seen as an order of magnitude - much in an analogous fashion as the nature of the Russian Matryoshka nested dolls; in that we can start at the top and look at the direct and reflected signals: - the direct signal of the loudspeaker assumed to be a coherent source;(in a related situation, this is also an issue between multiple distributed loudspeakers within a room); and then we can drill down to look at the relationship of the signals as they relate and interact between the various drivers and component aspects within the loudspeaker.

As one understands the basis time based relationships and the implications of the superposition between the complex behavior (meaning both magnitude and phase, coupled with Q and spatial dispersion, thus rendering the energy propagation as a complex volumetric vector doodad) of the propagation of sound, the basic physics remains remarkably consistent - simply reapplied at each level in the relationship hierarchy - to the various 'orders of magnitude' to which we have mentioned.

I will also quickly mention that viewing such behavior from a time perspective, as opposed to from a frequency perspective - using primarily the ETC response - provides the actual insight into what is actually happening between the 'component' signals at each level of analysis that is simply not apparent or visible from the frequency perspective.

Thus we see an almost identical relationship model reproduced over and over again in scale. And while we tend to focus starting at the 'largest'/highest level first, each level deserves attention as we refine the model to its optimal manifestation.

But then, while there is much that can be achieved, we still have to get folks beyond the seemingly pervasive myth that "all reflections are bad" misconception that pervades this and other forums! And to do so requires that more than a few stop and actually read a text such as Sound System Engineering by Davis (& Davis, 2nd ed.; or Partronis, 3rd Ed) to begin to understand the nature of the relationships at the various levels, be they environmental. large or small acoustical space, and direct signal/speaker level.

Unfortunately, without one taking the time to understand the basic relationships and behavior expressed in the text (and no, math is not required to grasp these functional relationships!), we are going to be limited in the effectiveness of raising the group awareness of the basic issues.

But folks running to embrace "room correction" software is NOT the answer -and it only reinforces the fact that many simply do not understand the basic signal relationships - even before the application of signal processing.
Old 15th June 2010
  #72
Here for the gear
 

SAC: I have gotten quite used to repeat these issues over and over again, and it does not tend to sink in as people are often reading these threads looking to find a conclusion to what DRC-system to choose or something else, and these advices do not answer that question.

I made a full range dispersion controlled speaker system i 2007 (~16-40k, ~130dB and dispersion controll all the way), it opened up some eyes to what these ideas can actually do to the result. They are now located in a pretty large living room (600sqft) with no acoustic treatment at all. The sound could be improved with some acoustic treatment, but not by far as much as with normal dispersion speakers. Anyway, this system is a proof of the importance of dispersion issues. However, people still tend to look at this gigantic speaker with 6 visible waveguides in front and still forget it has dispersion control.

The next step in this direction is to include room adaptive dispersion control that can be set up with different dispersion patterns according to the actual room the speakers are placed in. This if, off course, very difficult to do fully passive so it will be done with a DSP. Especially in the bass region, the DSP can give a frequency dependent dispersion pattern that can make the speaker adapt to a room almost avoiding LF room resonances completely.

I hope this much smaller actively controlled version can open more eyes.

Like BWO writes, I have set up several systems with DEQX, but also with other DSP-products. The dispersion control is always an issue that tends to limit the possibilities when working on a system.



What you (SAC) write about reflections is very interesting. The company that developed the MF/LF-lenses for Bang & Olufsens top models has an interesting approach to this. I don't remember their name at the moment, but if I find it, I'll post some links.
Old 15th June 2010
  #73
SAC
Registered User
 

Q and spatial pattern control are important.
But i would disagree that they resolve the need to address what happens to the energy once it encounters boundaries.

Perhaps the most researched acoustic model is that of the LEDE model (and its logical extension in the 'ambechoic' model)- which is in fact a FAR cry from what most erroneously seem to think it is! (In fact, it is not at all what most envision it to be based upon so many utterly fascinating and erroneous accounts available all over the web! And it is ironic that what most think of as an RFZ is but a SMALL and INCOMPLETE PORTION of an approach utilizing room geometry to moderate and direct early reflections in the LEDE of which the later arriving diffuse soundfield is integral! Peter D'Antonnio was NOT about creating an anechoic reflection free room!)

In fact, the concept of the Initial signal delay gap, the effective termination of said, and the establishment of the later arriving early reflections in the form of a predominately laterally arriving well behaved exponentially decaying diffuse semi-reverberant soundfield based upon the use of diffusion in corporates BOTH the well controlled loudspeaker AS well as a properly designed bounded system in a manner to effectively address both the energy dispersal as well as the related psychoacoustics of the space.

But as far as using DSP to resolve all environmentally sourced issues without room treatment is simply to ignore what we know about acoustics and psychoacoustics. FIR filters are great for manipulating and controlling the freq response and more significantly, the phase response of a single single bandpass or cell in an array direct signal. But they have no place in manipulating secondary non-minimum phase reflections a speaker. And laterally arriving diffuse soundfields do not arrive from the direct soundfield without the interaction of boundaries and diffusive elements. Likewise energy does not simply stop at a given point short of interacting with bounded surfaces. Thus to posit that the response within a room can be determined by the manipulation of the direct sound field without addressing he superposition of specular energy or of modal action is to posit that which has no basis in reality in a small acoustical space.

What many are not aware, is that such technologies are not new. And audio is not a leading field of development! Almost all of the technologies that have filtered down to acoustics have been sourced in optically based research. In fact, one might want to simply investigate the Schmidt CASI scatterometer rigs used for the analysis of optical diffraction of subatomic level structures to get an idea of what folks are just now approaching in the audible wavelengths with the exploration of absorptive, diffractive and diffusive scattering in locations such as NWAA. And one might wonder why Ron Sauro is among the first to identify and quantify diffractive edge behavior and losses in such structures as diffusors and absorbers! What is radical and 'paradigm' upsetting in acoustics is rather old hat in areas where such behavior is assumed and well understood! In fact, it is rather disappointing in one respect to see that our advances in acoustics lag so far behind the introduction of such approaches that have become routine in optics and sub-atomic research.

Heck, just witness here the active resistance to the acceptance of measurements from the time perspective that have been the de facto norm in so many other realms of physics research!

And even the latest research in boundary interactive technologies has not progressed to point where we can do much beyond very limited surface areas with extremely short wavelengths utilizing active systems - far outside of the the audible range! And passive systems are much less capable and far more limited in scope and abilities. In fact, having dealt quite extensively with current cutting edge imagining research and living for too long in the world of Mueller calculus and Mueller matrices and using elliptically polarized sources in a valiant attempt to minimize various sources of environmental interaction, I do not anticipate any substantial practical applications in terms of boundary mitigation in the audible wavelengths. (And to provide FAR too much information, how else would one reasonably know that the head of the octopus sperm is nature's 'left handed' helix!? And I DON'T want to know who or how they discovered that! But where else (indeed) do you find such a structure for reference??? ) In fact, we are having trouble effectively addressing such issues in a much more greatly constrained environments in the x-ray and infrared spectral regions - not to mentions application areas of radar, sonar, and imaging systems! And what success we have enjoyed has involved the necessary use of elliptically polarized coherent light sources - and I'll leave it at that. In other words - 'it ain't happening in the audible frequencies anytime soon'!

Again, DSP and basic signal alignment coupled with good design can indeed assist in the design of well behaved controlled Q speakers. But they nevertheless do not resolve the environmental issues imparted by the incidence of sound energy on boundary surfaces. And room treatment is necessary to achieve the level of completeness and accuracy to which we aspire. DSP has significant problems with multi-sourced non-minimum phase signals at a single finite "point", let alone over a distributed time variant field.

(One might note: Both Dick Heyser and Peter D'Antonio (and others integral to the exploration of acoustic behavior such as Ron Sauro, etc.) also came out of similar academic military directed research backgrounds which lead them to subsequently bring the respective derivative application technology to the measurement and analysis of reflective, diffractive and diffusive behavior in audible wavelengths.)

Thus signal processing technologies, while useful tools, as simply one small part of a larger systematic approach to the design of a total acoustical response created within a bounded small acoustical space. And the irony is that their primary applications are in the design and modification of the direct signal sources (e.g. speakers!), while most (here) are looking to attempt to apply them to distributed multi-sourced time variant fields where they are neither effective nor efficient.
Old 15th June 2010
  #74
Lives for gear
 
Bjorn Omholt's Avatar
 

Quote:
Originally Posted by Midgard Audio View Post
The next step in this direction is to include room adaptive dispersion control that can be set up with different dispersion patterns according to the actual room the speakers are placed in. This if, off course, very difficult to do fully passive so it will be done with a DSP. Especially in the bass region, the DSP can give a frequency dependent dispersion pattern that can make the speaker adapt to a room almost avoiding LF room resonances completely.
Resonances and room nodes are one thing, but the importance of delay (ETC) is another. And I cannot see how you can achieve an excellent result in the low frequencies without heavy use of bass trapping.
Old 15th June 2010
  #75
Lives for gear
 
Nordenstam's Avatar
 

Hi!

Glad to see eminent contributors from my part of the world joining the thread.

Quote:
Originally Posted by Midgard Audio View Post
My hope was that a thread like this could be about what to do and how to best combine DRC and acoustic products to achieve certain effects.
+1

So what can one practically do without any potential for harm?

Correcting minimum phase errors in the speakers is pretty obvious. How does one go about ensuring something is minimum phase in the first place? Have a look at the phase plot and avoid areas that goes beyond the unit circle boundaries? That would take care of the speaker itself. As I've gathered, this is the relatively easy part of the task.

The next subject in line that I believe can be corrected with some certainty is the static reflections from speaker to boundary and back to speaker again. These alter the acoustic impedance as the driver sees it. They do not change with measurement position. It's also a straightforward process to calculate and experimentally verify the frequencies involved. As far as I've gathered, the boosts this cause can be corrected without much problem. Am I right in assuming this?

Energizing strong room resonances/standing waves less than a flat curve gives seems to be another place where correction can do good without much potential pitfalls. Except that the direct sound will be off. Probably not the biggest problem when it's only in the very low end. Right?

Then.. What?


I know this isn't much of a post in itself. Hope to get the ball rolling in the right direction!


Best regards,

Andreas Nordenstam
Old 15th June 2010
  #76
Lives for gear
 
avare's Avatar
 

Quote:
Originally Posted by Midgard Audio View Post
Some people tends to be talking about room correction as full impulse response correction with FIR-filtering. Others seems to be including both FIR and IIR (EQ and phase correction only) into the term "room correction". Personally, I would include both and I would also claim that if the room is good, there is no obvious need for FIR filtering at all. The remaining ambience should blend naturally into the sound.
+1. I agree with the last sentence quoted completely.

Andre
Old 15th June 2010
  #77
Lives for gear
 

Quote:
Originally Posted by Lupo View Post
The next subject in line that I believe can be corrected with some certainty is the static reflections from speaker to boundary and back to speaker again. These alter the acoustic impedance as the driver sees it. They do not change with measurement position. It's also a straightforward process to calculate and experimentally verify the frequencies involved. As far as I've gathered, the boosts this cause can be corrected without much problem. Am I right in assuming this?
I would also like to know more about this. The same issue that causes variable impedance at the speaker is simultaneously causing variable pressure around the room. If the speaker is putting out a linear response at the speaker, there still remains the issue of the nonlinearities at any given point away from the speaker.
Old 15th June 2010
  #78
SAC
Registered User
 

Originally Posted by Midgard Audio
Some people tends to be talking about room correction as full impulse response correction with FIR-filtering. Others seems to be including both FIR and IIR (EQ and phase correction only) into the term "room correction". Personally, I would include both and I would also claim that if the room is good, there is no obvious need for FIR filtering at all. The remaining ambience should blend naturally into the sound.


Quote:
Originally Posted by avare View Post
+1. I agree with the last sentence quoted completely.

Andre
OK. But the 'problem' here is that we are still dealing with a well behaved and controlled direct signal (for which DSP can be of great use) and a well tuned room such that later arriving reflections are blended in a controlled constructive fashion reinforcing the sense of space - an application for which DSP 'room correction' is unsuited.

In other words, we have complimentary systems that minimize the creation of additional destructive issues and that augment one another. Exactly what the proper selection of source and the proper tuning of the space ideally achieve.
Old 15th June 2010
  #79
SAC
Registered User
 

Originally Posted by Lupo
The next subject in line that I believe can be corrected with some certainty is the static reflections from speaker to boundary and back to speaker again. These alter the acoustic impedance as the driver sees it. They do not change with measurement position. It's also a straightforward process to calculate and experimentally verify the frequencies involved. As far as I've gathered, the boosts this cause can be corrected without much problem. Am I right in assuming this?

Quote:
Originally Posted by Brainchild View Post
I would also like to know more about this. The same issue that causes variable impedance at the speaker is simultaneously causing variable pressure around the room. If the speaker is putting out a linear response at the speaker, there still remains the issue of the nonlinearities at any given point away from the speaker.
Not a bad idea if one selects one position and focuses on one incident spot on a given boundary. But remember that reality is not quite so simple, and that we must address multiple sources and the incident point everywhere on every boundary - and suddenly the neat linear relationship becomes anything but. And the latency of subsequent processing becomes prohibitive even for just one listening spot and one source.

Now increase this scenario for all sources and increase the potential listening spots and the number of incident locations!

And lets not forget the complexities of reality that enter into the equation as all surface impedances are complex (this is a huge factor complicating issues) and that they are NOT constant and consistent.

Suddenly our ray tracing model is anything but a simple linear model. And we enter into the world where even the well defined room model (e.g. EASE and CATT-a) where auralization is employed is strained and which involves a latency that is utterly insufficient to support real time processing - even while assuming that the source is well behaved and functions as a true point source with defined Q and dispersion (well, we can dream, as this is ideal and "a consummation devoutly to be wished" - to coin a phrase!).

Hmmm...and does anyone dare to complicate the model by adding coupled spaces or any number of real world variations sure to make even a cluster of massively parallel processing systems shudder...

The fact is, we long ago left the realm of what is practically possible and entered into the realm occupied by room modeling that, for all of its amazing advances, is by no means close to real time processing and even less regarding the complex non-linearities of complex surface modeling and non-ideal sources.

The fact is, in practice we have difficulty doing this on the level equivalent to what an ideal model room mode calculator performs, let alone with the complex non-linear systems imposed by the variations of real world geometry and real world surface impedance models and real world source modeling applicable to a multi-positional time variant distributed topology that basically says 'do this extremely complex calculation once, and then an almost infinite more times for all of the various positional variations in near zero real time - assuming that there was a way to simultaneously modify the direct signal to address not only the direct signal, but to compensate for the multitude of derivative reflected signals, modified in time and character by the non-linear nature of the incidence with a variety of complex surface impedances - and let's be nice and limit this model to simple first order reflections and ignore the complexities of such surface issues as diffusion- of which 'simple' well behaved quadratic models are only now becoming available for simple modeling (e.g.EASE)!

Well...its easy to become overwhelmed in simply describing a very basic model, let alone in addressing the complexities of many real world environments. And we have already long ago exceeded the capabilities of all but the most advanced modeling programs which do not have restrictions on latency (and which still exhibit fundamental simplification of surface and source models)!

And all of this assumes for the sake of discussion that the DSP algorithms are all fully mature, functional, and accurately address the myriad issues - as we have not even addressed the challenges of the mathematical models required to construct this Holodeck that many brochures are want to convince you not only exists, but exists in a mature form on readily available processing equipment that you simply run a jumper to or, better yet, exists in a few additional ICs mounted inconspicuously in the motherboard of your home theater receiver...despite the fact that we have strained or exceeded the capabilities offered by such systems as the RS/6000SP and derivatives running PSSP -assuming that you have the software models up and running!.

Bottomline, those wishing to see the basic capabilities of room modeling software easily costing more than most folks computing systems and processing capabilities increased (and based upon the input of carefully defined parameters - as opposed to supposedly measuring and determining their real nature on the fly with a 'built in' microphone!) are frustrated as much by how far things have progressed as by how far they have yet to go. And they have the luxury of allowing the programs to run for days on elaborate dedicated hardware platforms. And yet glossy brochures touting real time "room correction" capabilities abound.

Dream as one might, and ignoring the fundamental kinks in the systems that form impediments to basic processing models, its time to acknowledge the fundamental limits to the systems.

And if anyone doubts this, start small and take some time and talk to the folks at AFMG or RPG regarding the difficulties they are having with basic room modeling models in both EASE and in programs as seemingly simple as Room Sizer an Room Optimizer - and ask them when the later two will allow input of parameters other than for simple ideal rectangular shapes, or heaven forbid, surface impedance - or to generate some real laughs, sources modeled as other than ideal point sources addressed in the far field!
Old 15th June 2010
  #80
Here for the gear
 

Quote:
Originally Posted by bwo View Post
Resonances and room nodes are one thing, but the importance of delay (ETC) is another. And I cannot see how you can achieve an excellent result in the low frequencies without heavy use of bass trapping.
I am not quite sure what kind of delay source you mean. Delay occurs normally when the room is a major contributor to the energy. Actually, delay is pretty much a result of Q-factors and the resulting phase distortion, and this is closely connected to resonances. The ears are not very sensitive to delay smaller than 1/2 wavelength if the transition is smooth, so for this to occur in the speaker or as a result of the near boundaries is quite unlikely if you avoid resonances in the first place.
Old 15th June 2010
  #81
Here for the gear
 

Quote:
Originally Posted by Lupo View Post
The next subject in line that I believe can be corrected with some certainty is the static reflections from speaker to boundary and back to speaker again. These alter the acoustic impedance as the driver sees it. They do not change with measurement position. It's also a straightforward process to calculate and experimentally verify the frequencies involved. As far as I've gathered, the boosts this cause can be corrected without much problem. Am I right in assuming this?
Hi Andreas!

I am not sure I understand you correctly, but do you asume that the cone motion will be affected by the sound reflected from a boundary close to the speaker in such a way that it is audible as a reduction in SPL?

If the speakers play at 90dB on a distance of 1 meter, and the boundary is 1m away from the speaker, the reflection will typically be 84dB. About 30mm from the speaker cone, the combined SPL from the diaphragm is ~120dB. If the excursion at this level is 0.1mm, the attenaution will result in a cone excursion reduction of 0.0016mm. If the cone excursion were reduced by 0.016mm, the attenuation would be around 0.7dB. This is 10 times less...

Since cone drivers emit sound from such small surfaces, the mechanical forces involved are so significant, the room impedance has to be considered extremely high compared to the "output impedance of the speaker".
Old 15th June 2010
  #82
Lives for gear
 
avare's Avatar
 

Quote:
Originally Posted by SAC View Post
Exactly what the proper selection of source and the proper tuning of the space ideally achieve.
You summed it up quite nicely. IF the room and speakers are properly designed as an acoustic system, then furhter processing is not required.

Andre
Old 15th June 2010
  #83
Gear Guru
 
Ethan Winer's Avatar
 

Lightbulb

Quote:
Originally Posted by mrhudson View Post
I'd be curious to know what information folks would look at and then say, "yup, correction software might add something here."
From my perspective the recipe is very simple: You put in as many bass traps as possible, then use EQ to reduce the one or two remaining modal (not positional) peaks below 50 or 60 Hz. Reducing LF modal peaks is all room EQ is good for anyway, and most bass traps don't work very well below 50 to 70 Hz. This is what I did in my living room home theater system. Even with a lot of bass traps I had a small (3 dB) peak around 40 Hz. So I used the one-band cut-only EQ that's built into my SVS subwoofer to reduce that one peak by 3 dB. Subjectively the sound is better, without the extra "weight" the peak added.

--Ethan

________________
The Acoustic Treatment Experts
Old 16th June 2010
  #84
Lives for gear
 
Weasel9992's Avatar
 

Quote:
Originally Posted by Ethan Winer View Post
From my perspective the recipe is very simple: You put in as many bass traps as possible, then use EQ to reduce the one or two remaining modal (not positional) peaks below 50 or 60 Hz.
Exactly. I totally agree.

Frank
Old 16th June 2010
  #85
Lives for gear
 
Bjorn Omholt's Avatar
 

Even better would be to address low frequency anomalies with resonators. Walls are a fine place for panel resonators.
Old 17th June 2010
  #86
Lives for gear
 
Nordenstam's Avatar
 

Quote:
Originally Posted by Midgard Audio View Post
I am not sure I understand you correctly, but do you asume that the cone motion will be affected by the sound reflected from a boundary close to the speaker in such a way that it is audible as a reduction in SPL?

.. Since cone drivers emit sound from such small surfaces, the mechanical forces involved are so significant, the room impedance has to be considered extremely high compared to the "output impedance of the speaker".
Hi!

Perhaps impedance is the wrong idea? It's about trying to create the opposite of the pressure already existent at that time/space in the room. The speaker is trying to make the air pressure go above atmospheric while there's already a negative pressure zone in the air outside the cone(or visa versa). You probably get the idea! Also, keep in mind that a speaker typically radiates into less than 4pi space at such low frequencies.

Have seen this in several measurements where there are prominent dips at the 180' frequency cancelations from close boundaries. This is not the dip from superpositioning direct sound and reflected sound at the sweet spot location. It's a dip that does not change with measurement position, corresponding directly to the roundtrip from speaker to boundary and back to speaker. Intersecting the path with absorbent also helped to dampen the problem when it was observed.

The point at hand here is that it's a steady state effect independent on listener location. This makes me think it should be possible to tamper with the speaker stimuli without creating possibly obnoxious alterations to the direct sound.
Old 18th June 2010
  #87


Sure, it's possible. Deconvolving the effects of the nearfield impedance from the reciever (listener/mic) position effects could be quite difficult.

Then, as you said, absorbtive treatment is fairly effective (and currently cheaper).



-tINY

Old 18th June 2010
  #88
Here for the gear
 

Quote:
Originally Posted by Lupo View Post
Hi!
Perhaps impedance is the wrong idea? It's about trying to create the opposite of the pressure already existent at that time/space in the room. The speaker is trying to make the air pressure go above atmospheric while there's already a negative pressure zone in the air outside the cone(or visa versa). You probably get the idea! Also, keep in mind that a speaker typically radiates into less than 4pi space at such low frequencies.
Yes, I get the idea. However, the speaker cone is not sensing atmospheric pressure. It only applies a force that is many times stronger than what is required to move the air. I think this effect is very unlikely. If you can see it as a dip in the response curve, it should also be viewev as some 50% 3.rd harmonic distortion. The forces at play are not very far from what I described earlier. It is important to point out that you need at least 1/4 wavelength from the speaker to the nearest boundary for this effect to be able to take place, and at that point, the amplicifaction from the room is not that significant.

Quote:
Originally Posted by Lupo View Post
Have seen this in several measurements where there are prominent dips at the 180' frequency cancelations from close boundaries. This is not the dip from superpositioning direct sound and reflected sound at the sweet spot location. It's a dip that does not change with measurement position, corresponding directly to the roundtrip from speaker to boundary and back to speaker. Intersecting the path with absorbent also helped to dampen the problem when it was observed.
Isn't that typical for the rear wall (behind the speakers) reflections? They will affect the listener at most places in the room, and is a result of the blend between direct sound and rear wall reflections.
Old 18th June 2010
  #89
SAC
Registered User
 

The Heyser spiral IS the complete complex impedance.

Easily measured and displayed in both 3Space as a rotating phasor and also instantly deconvolved into each of its perspective responses!

Heck, you move the cursor along the actual response and it drops perpendicular bisectors to each axis and displays instantaneous values along each axis!





Now, if someone will tell us how DSP can not only address an infinite number of non-linear temporally and spatially distributed variations in the room (based upon the incidence of signals with non-uniform acoustical impedance boundaries and a non-ideal signal source) but simultaneously compensate for that variation in the ambient soundfield while maintaining the integrity of the direct signal - all by modifying just the direct signal!

And the latency is????? How many days do you allow? And what processing engine are you proposing to use IF such a task were even possible?

Personally, you would be much better off expending the energy to create a Holodeck which would have much greater commercial application.
Old 18th June 2010
  #90


There is no question that you can do it for one position in the room (forgetting for a moment that we have two ears).

Latency isn't an issue on playback either - only a problem if you want to play instruments live and listen through your monitors.




-tINY

Thread Tools
Search this Thread
Search this Thread:

Advanced Search

Forum Jump
Forum Jump