Edit/Addition:
Having an interval of N = 4 cycles to be analysed, the pathlength of the sound
( Pathlength = N x Wavelength, Pathlength = N x SpeedOfSound / Frequency, D = N x c/f )
will be approximately:
400Hz....3,4m
800Hz....1,7m
1600Hz...0,9m
This means at 1600Hz in many room steups just bottom reflection and a reflection from a close sidewall might come into play and about >3KHz in many cases there will be no more reflections from "structures away/outside the speaker" having influence. So using this technique and N=4 we are arrived at "pretty anechoic" conditons about >3Khz then and what we do in the highs is mainly "speaker correction".
Still way too high a frequency to go to "speaker correction only" to my mind .... and still we have to talk about "which kind of a speaker correction"(*).
But i'd like to remind posters and readers kindly: This is not the room/speaker correction method proposed by the OP.
___________________
(*) There are also things that are not useful to "correct like resonance" even when measuring/correcting "speaker only":
One should be able at least to distinguish interference artefacts from resonance artefacts, to decide for useful compensation in a speaker.
Last edited:
Don't skip this post LineArray: http://www.diyaudio.com/forums/full-range/284916-room-correction-speaker-correction-what-can-we-do-dsp-power-now-availabl-10.html#post4594289
I still feel funny typing your name, knowing that you've swayed away from that concept and me starting to use it
.
Starting being the wrong word here though as it started for me over 4 years ago.
I still feel funny typing your name, knowing that you've swayed away from that concept and me starting to use it
.Starting being the wrong word here though as it started for me over 4 years ago.
I agree. The question is which compromises make sense and which don't. Only someone with a naive understanding of sound recording, reproduction and perception would make dogmatic statements like "phase is more important than frequency response".
The topic can not be completely understood by excluding the unique properties of the stereo sound field which is very much part of the transmission channel. Making the speaker behave more like the recorded signal ("needs to be able to reproduce a square wave") isn't necessarily improving stereo perception in each and every scenario.
Last edited:
Wesayso said:
You know quite well, there is a difference being made between ipsilateral and contralateral side reflections (briefly):
- Earl Geddes votes for avoiding the ipsilateral ones but maintaining the contralateral ones.
- Floyd Toole states that even the ipsilateral ones may contribute to spceousness and enjoyfull listening.
So your above statement was oversimplified, which is no good thing in room acoustics:
Merging both standpoints ("Geddes vs. Toole") means, a loudpseaker design and setup has to be able to balance ipsilateral vs. contralateral side reflections. You might point out, how this could be done using a speaker having no controlled dispersion (not to talk about "adjustable dispersion pattern", which seems like a "red rag" to you, i am sorry).
But if you use e.g. (monopolar non-dipole) line arrays in a living room you have highly correlated and strong side reflections - ipslateral and contralateral as well - unless you care for serious room treatment of the side walls ... which could be some combination of diffusion an absorption.
My strategy is usually not causing acoustical problems with a setup, to have something to "correct" afterwards, but creating a beneficial soundfield from the start as far as possible.
Kind Regards
Last edited:
Like I said: when choosing between phase and frequency it is frequency for me. But that didn't stop me from trying to get both as right as I could but it has it's own set of compromises. I would love to try and do the same quest with a synergy horn setup. I couldn't sell that concept to my other half, yet she accepted those huge contraptions with an a4 size footprint.
So where's your APL_TDA measurement again?
You know what I am talking about and so do I. I already admitted to use room treatment to reach my goal, didn't I. Look at the thread from jim1961, following the lead to his thread on gearslutz and see what he did to test the above theories. It's not just a well damped room. Every reflection and diffusion reaching the listeners ears has been carefully planned.
That's dedication. DIY dedication. Not just talking about theories from others.
I still agree with Geddes here after making up my own mind. It's not just DSP power as I've said before and you can even read about it.
What I did was planned for 1.5 years before the actual build and did involve the theories of others but mainly to test my own. Mine is on here for all to read. We still have to guess about anything you did.
So what is your point exactly?
Just a note here: The "red rag" you bring up is not about dispersion. I know perfectly well why I chose my set of compromises and how I wanted to deal with that.
I have no issue with that at all. Why would you think that? I never even mentioned it, but you keep bringing it up as if you really want to tell us what it is you are doing, but can't. I cannot help you there. Without spilling your beans we can only guess about it. I'm not willing to do that, I might guess wrong. So prove your point, show some measurements... I certainly did my share of those.
I just can't follow what it is you're trying to do in this thread. To me I saw common ground in the views of Raimonds and decided to test his software. I will probably try his suite at some point. Just out of curiosity.
I asked about your own DSP experiences, no answer there... just more quotes from Toole. So what do you want? If you don't like the subject of DSP, why bother to spill al this stuff on us? Without telling us what it is you want to tell us. That much is clear. You hint at microphones and directivity. I know why you would want directivity, I'm no fool. But why the secrecy if you're trying to make a point?
You want proof from us. Proof that has been posted before if you look for posts by member "Mitchba" using Acourate as his poison of choice.
So you do make a lot of noise, but fail to bring something new to the table. The material you bring up is known I assume. I sure know it.
I just don't get it and I can't blame the OP for not posting while you're on this crusade. All I am asking is why?
Last edited:
Wesayso said:
Maybe there will be some dispersion patterns etc. in a different thread some time: The mid-to-high device having adjustable dispersion allows for XO above 400Hz, so not a "Full Range" topic i fear.
Furthermore it is pefereably combined with a cardioid woofer and a subwoofer system, again pretty much away from "fullrange" forum ...
But "ajdustable dispersion pattern" as a way of "room compensation", may have been recognized as being a valid approach in general in this thread, if not by yourself.
I "spill my beans" whenever or whereever i decide to do so and i will show parameters and measurements i see as being informative. This seems not an "inviting" context here ...
In the contrary: Like many other threads on "room compensation" , here is a lack of accepting basics like
- dispersion, DI
- early reflections (levels and directions)
- rooms decay time versus frequency
- ...
and so on.
Sticking to the notion of "room correction" (not "speaker correction" ...) even in the mid-to-high range is just "grasping at straws" for those, who outmaneuvered themselves by ignoring the basics.
Last edited:
Did you bother to look up the clientele of the OP? Not exactly living room people I'd say. He does what he believes in and saw a lot of talk on this forum about digital sound processing. He shared his view with us. Nothing wrong with that in my opinion.
If you did look into that you'd know the OP did not offer his software as a means to treat a room. I even think he may have been surprised by my resulting plots.
You still fail to answer my points and keep rehashing points that are known to more members on here than you alone. The following points were on my mind when I started my project:
And I chose my set of compromises to deal with each one of them in my own way. For those interested I decided to start a thread about it. Sharing my experiences with others, good or bad. My experience with using DSP has been a good one. Not to eliminate the need for room treatment. But to help me get further without turning my living room into jim1961's awesome listening room. If I had a spare room I would know what to do with it, that's for sure.
I'm sorry that I didn't do it your way. But I'm perfectly happy with what I've learned from this myself. Does it help you if I tell you that all my processing above 500 Hz is minimum phase in settings?
Meaning who, exactly...
And what do you get out of this?
If you did look into that you'd know the OP did not offer his software as a means to treat a room. I even think he may have been surprised by my resulting plots.
You still fail to answer my points and keep rehashing points that are known to more members on here than you alone. The following points were on my mind when I started my project:
- effects of phase in our listening experience
And I chose my set of compromises to deal with each one of them in my own way. For those interested I decided to start a thread about it. Sharing my experiences with others, good or bad. My experience with using DSP has been a good one. Not to eliminate the need for room treatment. But to help me get further without turning my living room into jim1961's awesome listening room. If I had a spare room I would know what to do with it, that's for sure.
I'm sorry that I didn't do it your way. But I'm perfectly happy with what I've learned from this myself. Does it help you if I tell you that all my processing above 500 Hz is minimum phase in settings?
Meaning who, exactly...
And what do you get out of this?
Last edited:
No need to patronize me.
Some tests were done with headphones. Even under such conditions phase error were largely inaudible. I doubt you will ever find a speaker that has less distortion.
Have you looked at how phase response changes when a violinist changes his position? You'll see dramatic changes. Yet that violin sounds just like the same violin. Our hearing is quite complex and not much is really understood. But what we know is that it has its limitations. Once something is beyond a certain threshold we won't be able to perceive it. Just like we can't see infrared light.
By the way, Watkinson presents just a hypothesis without any testing whether it is true or not. No conclusion can be drawn from this.
So what you're basically saying is that we don't need to take existing roads into account when designing and building a car. It is enough to build the best car with "best" being defined by an untested hypothesis.
By the way, your guru Watkinson advocates omnidirectional loudspeakers because only then reflections could be correlated to the direct sound by our hearing. Another untested hypothesis.
Derek,
The problem being there is no perfect moving coil speaker that is perfect, whether a standard curved cone, straight walled cone or even an BMR type. I have worked with someone on a flat panel type of diaphragm device and I do know what kinds of secondary problems those bring to the party. You just exchange one set of problems for something else. Nothing is perfect or even close in any sense. All we can do is make the best compromises we see fit, not say ours is perfect and everyone else has got it wrong. My expertise was in materials so I understood some things others just missed, but I do know that I can't make anything perfect, there are always physical trade-offs we have to choose.
I may not like the MBL type of speaker but I have been helping to develop one of those types of devices. I can help with materials properties that others don't understand to improve that type of device. Still doesn't mean I want to listen to that type of dispersion pattern, I don't.
The problem being there is no perfect moving coil speaker that is perfect, whether a standard curved cone, straight walled cone or even an BMR type. I have worked with someone on a flat panel type of diaphragm device and I do know what kinds of secondary problems those bring to the party. You just exchange one set of problems for something else. Nothing is perfect or even close in any sense. All we can do is make the best compromises we see fit, not say ours is perfect and everyone else has got it wrong. My expertise was in materials so I understood some things others just missed, but I do know that I can't make anything perfect, there are always physical trade-offs we have to choose.
I may not like the MBL type of speaker but I have been helping to develop one of those types of devices. I can help with materials properties that others don't understand to improve that type of device. Still doesn't mean I want to listen to that type of dispersion pattern, I don't.
But what can be said is that a BMR has potentially the worst time behavior. It works by utilizing resonances which by definition smear any time information. Not saying this would create audible problems in a well designed BMR though
We're pretty far off-topic now.
Pnix,
actually I don't think that we are that far off topic as far as needing to understand how different types of speakers actually function as far as their FR and Phase, timing response. This would have a definite factor with any use of dsp corrections of every type. If your using a line array against a wall or a mini monitor out in a room or a horn loaded system each will have its own specific needs. I don't think we need to go to deep into details to understand that, but I think anyone trying to apply these tools should understand what it is they are trying to correct for.
I''l leave it at that.
And I agree that the function of any flat planner diaphragms works much differently than many would seem to believe, they definitely don't function as a rigid flat diaphragm in the least.
actually I don't think that we are that far off topic as far as needing to understand how different types of speakers actually function as far as their FR and Phase, timing response. This would have a definite factor with any use of dsp corrections of every type. If your using a line array against a wall or a mini monitor out in a room or a horn loaded system each will have its own specific needs. I don't think we need to go to deep into details to understand that, but I think anyone trying to apply these tools should understand what it is they are trying to correct for.
I''l leave it at that.
And I agree that the function of any flat planner diaphragms works much differently than many would seem to believe, they definitely don't function as a rigid flat diaphragm in the least.
But what can be said is that a BMR has potentially the worst time behavior. It works by utilizing resonances which by definition smear any time information. Not saying this would create audible problems in a well designed BMR though
We're pretty far off-topic now.
I don't understand this at all....Please explain, I am very interested in BMR 's and you obviously know a lot about them....
It is my understanding that you're working with such drivers so I don't understand why you want me to explain what the principal idea behind a balanced mode radiator is. Rest assured I've read the available literature (as I've read all the papers and books I've linked).
Last edited:
I was just looking at this thread
http://www.diyaudio.com/forums/full-range/142467-driver-behaviour-pistonic-ocillation.html
No thanks, over and out
http://www.diyaudio.com/forums/full-range/142467-driver-behaviour-pistonic-ocillation.html
No thanks, over and out
Because you can...
If you understand the BMR principal you should be ware that the time domain and power bandwidth are the the strengths of the driver.....The exact opposite of your conclusions.....Have you really understood the raison d'etre ?
Thanks so much for your assurance.....Here was me all worried & ready to buy a parts express passive two way ported kit.....Phew!
Please post some examples of your conclusions......? CSD plots are the defacto indicator of time domain performance....
Have a look at one of my drivers CSD plots.....From 300Hz up its clean ....Under 1 Ms settling time. No Eq applied
Of course as its power response is equally impressive, the drivers frequency response can then be Eq'd to flat or whatever response one desires.
The second attachment should help explain why time domain is so important.
Enjoy the attachments I do hope they help
If you understand the BMR principal you should be ware that the time domain and power bandwidth are the the strengths of the driver.....The exact opposite of your conclusions.....Have you really understood the raison d'etre ?
Thanks so much for your assurance.....Here was me all worried & ready to buy a parts express passive two way ported kit.....Phew!
Please post some examples of your conclusions......? CSD plots are the defacto indicator of time domain performance....
Have a look at one of my drivers CSD plots.....From 300Hz up its clean ....Under 1 Ms settling time. No Eq applied
Of course as its power response is equally impressive, the drivers frequency response can then be Eq'd to flat or whatever response one desires.
The second attachment should help explain why time domain is so important.
Enjoy the attachments I do hope they help
If you understand the BMR principal you should be ware that the time domain and power bandwidth are the the strengths of the driver.....The exact opposite of your conclusions.....Have you really understood the raison d'etre ?
Thanks so much for your assurance.....Here was me all worried & ready to buy a parts express passive two way ported kit.....Phew!
Please post some examples of your conclusions......? CSD plots are the defacto indicator of time domain performance....
Have a look at one of my drivers CSD plots.....From 300Hz up its clean ....Under 1 Ms settling time. No Eq applied
Of course as its power response is equally impressive, the drivers frequency response can then be Eq'd to flat or whatever response one desires.
The second attachment should help explain why time domain is so important.
Enjoy the attachments I do hope they help
This is like witnessing an accident. You want to move on but you can't.
Now things get a bit absurd. You're posting a "comparison" without any data that could be compared? The single plot you show doesn't look that spectacular. Just did a quick comparison to a Tymphany 3".
Having better drivers is great but you can't cheat physics. Distortion, bandwidth and dispersion will always be a trade-off. Creating a driver that can reproduce square waves is a dead end. Our hearing has limited bandwidth but a perfect square wave has infinite bandwidth. It's an exercise in futility.
First you need to understand how stereo really works (inter-channel intensity differences create phase differences at the ears) then you can decide what the goal really is. Again, read Blumlein's stereo patent. Understanding the speaker-room interface is key.