gedlee said:
Umm, excuse me... you just stated that your methods are 2 channel and 'error" immune... but implied that perhaps Markus' data is not. So please don't chastise me about "impolite" posts. I never stated that your statistics has anything to do with a lack of error analysis. perhaps you don't unsderstand the statistics I'm talking about.
John L.
data intepretation (interpolation)
No one said Markus is trying to proove anything. One would hope we all try to use whatever tools are available to improve our systems. Markus' data in post 321 is not consistent. No attempt to explain why. Then in post349, another graph, with different results (Peak to trough @ 50 Hz varies from 15dB in one set to less than 6 dB in another, with no explanation. So why not present the data as an average, smoothed, give error bars (or whatever you wish to call them) etc. Otherwise, it lookslike hash, and makes conclusions suspect to random changes.
then again, correlate the changes with what you actually hear listening to music, and there's progress to be made.
John L.
John L.
markus76 said:
No one said Markus is trying to proove anything. One would hope we all try to use whatever tools are available to improve our systems. Markus' data in post 321 is not consistent. No attempt to explain why. Then in post349, another graph, with different results (Peak to trough @ 50 Hz varies from 15dB in one set to less than 6 dB in another, with no explanation. So why not present the data as an average, smoothed, give error bars (or whatever you wish to call them) etc. Otherwise, it lookslike hash, and makes conclusions suspect to random changes.
then again, correlate the changes with what you actually hear listening to music, and there's progress to be made.
John L.
John L.
Re: data intepretation (interpolation)
Did you read the post? I've showed 3 different positions. As this is an in-room response there have to be deviations.
This shows a completely new setup calibration. Sorry if I haven't stated that clear enough for you. But on the other hand why didn't you just ask?
I don't want to sound disrespectful but no one in this forum beside Earl has the expertise to interpret the data I'm posting because the methodology used is his and hasn't been described elsewhere on the internet. So all we can see is a more or less smoother frequency response where smoother is better and peaks are worse than dips. But that should be something commonly known?
Best, Markus
Did you read the post? I've showed 3 different positions. As this is an in-room response there have to be deviations.
This shows a completely new setup calibration. Sorry if I haven't stated that clear enough for you. But on the other hand why didn't you just ask?
I don't want to sound disrespectful but no one in this forum beside Earl has the expertise to interpret the data I'm posting because the methodology used is his and hasn't been described elsewhere on the internet. So all we can see is a more or less smoother frequency response where smoother is better and peaks are worse than dips. But that should be something commonly known?
Best, Markus
I've read that many times but have never found proof in literature. Did you?
Or does it create a low IACC which is desirable? The delay of that reflection to the direct sound is 9 ms...
Re: Re: data intepretation (interpolation)
Yes, I read the posts. I could point out all the posts by others where your data and methodology has been either questioned, refuted, or accepted as "done", but I won't. Too many "egos" invovled. I've got other things to do. I think you did a bang-up job on the nathan kit, which looks to be a great speaker for small rooms, one heck of a finish under less than ideal circumstances. I hope they meet all your expectations. Perhaps you'd be more satisifed by expending less energy trying to generate more data and sit back and enjoy the fruits of all your labor.
I prefer "homebrew" but have at times enjoyed Murphy's, Boddington's, etc.
Regards
John L.
markus76 said:
Yes, I read the posts. I could point out all the posts by others where your data and methodology has been either questioned, refuted, or accepted as "done", but I won't. Too many "egos" invovled. I've got other things to do. I think you did a bang-up job on the nathan kit, which looks to be a great speaker for small rooms, one heck of a finish under less than ideal circumstances. I hope they meet all your expectations. Perhaps you'd be more satisifed by expending less energy trying to generate more data and sit back and enjoy the fruits of all your labor.
I prefer "homebrew" but have at times enjoyed Murphy's, Boddington's, etc.
Regards
John L.
Re: Re: Re: data intepretation (interpolation)
Who said I'm not satisfied? I'm just trying to get the best out of my equipment. Words describing what only can be heard are of no use here. I try not to participate in such discussions anymore. Only measurements can help when looking for objective exchange of knowledge.
Best, Markus
Who said I'm not satisfied? I'm just trying to get the best out of my equipment. Words describing what only can be heard are of no use here. I try not to participate in such discussions anymore. Only measurements can help when looking for objective exchange of knowledge.
Best, Markus
Markus,
It looks as tho you've got your Nathans 'tuned-in', at least as a great first pass. What are your listening impressions?
Are they meeting your expectations?
It looks as tho you've got your Nathans 'tuned-in', at least as a great first pass. What are your listening impressions?
Are they meeting your expectations?
markus76 said:
I'm not sure there is such a thing as "proof" in any "paper". More like "something worth reviewing".😉 I don't however remember this from any scholarly work on the subject (..not to say that there hasn't been or that I haven't read it, just that I don't remember it). Subjects "touching" on this subject sure, but you've already discussed them in the perception thread (..was Hawksford discussed as well?) Anyway, I don't agree with several conclusions Hawksford made with respect to loudspeaker placement in room. (..and yes, I've tried the method recommended, and well before reading his paper.) I guess I should read Toole's book, but who has the time?
As for that "corner" reflection, I'm NOT concerned with the delay (and if it reides in the correct time-window), but rather the amplitude in relation to direct sound and its "channel" in respect to its position and you.
Really all of the above "proof" or not - doesn't mean squat if it doesn't work for you. Best thing I can recommend is to try multiple suggestions until you "hit" on the one that seems best (..provided of course its livable). It shouldn't take a great deal of effort, and the time spent may well be less than half the time you've spent on just this thread.🙂
Oh, I'll also note that its a pretty easy experiment to do for yourself. Some powered computer speakers with good "imaging" characteristics in a bathroom very close to most of the walls - still provides good "imaging" and "soundstage" effects (with uniform placement).
Some (hopefully) interesting data on reflections at the listening position.
You see a very early and loud first reflection from the coffe table from the left side. There's none for the right speaker because of the sofas armrest. The strange influence of the reflection from the left: the stereo image is shifted to the RIGHT. Any ideas how to explain that? When removing the coffee table the image is centered again.
Left speaker:
Right speaker:
Left speaker without the coffee table:
Right speaker without the coffee table:
Last diagram shows what happens when I remove the absorber at the backwall behind the listening position (only right speaker shown - left one looks more or less the same):
Best, Markus
You see a very early and loud first reflection from the coffe table from the left side. There's none for the right speaker because of the sofas armrest. The strange influence of the reflection from the left: the stereo image is shifted to the RIGHT. Any ideas how to explain that? When removing the coffee table the image is centered again.
Left speaker:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker without the coffee table:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker without the coffee table:
An externally hosted image should be here but it was not working when we last tested it.
Last diagram shows what happens when I remove the absorber at the backwall behind the listening position (only right speaker shown - left one looks more or less the same):
An externally hosted image should be here but it was not working when we last tested it.
Best, Markus
markus76 said:
It would be interesting to see the time for that reflection, certainly the amplitude is high.
What was the subjective effect of removing the rear wall absorber?
Markus
Personally I am most interested in the first 10 ms. Any chance of data from 0 to 20 ms. (after travel time has been removed - I have no use for dead space.)
Personally I am most interested in the first 10 ms. Any chance of data from 0 to 20 ms. (after travel time has been removed - I have no use for dead space.)
Here we go - first 20 ms:
Left speaker:
Right speaker:
Left speaker without the coffee table:
Right speaker without the coffee table:
Left speaker without the coffee table and backwall absorber:
Right speaker without the coffee table and backwall absorber:
Left speaker:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker without the coffee table:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker without the coffee table:
An externally hosted image should be here but it was not working when we last tested it.
Left speaker without the coffee table and backwall absorber:
An externally hosted image should be here but it was not working when we last tested it.
Right speaker without the coffee table and backwall absorber:
An externally hosted image should be here but it was not working when we last tested it.
A cubic armchair made of thick foam seems to be a very effective absorber for floor reflections. Talking about something like this:
An externally hosted image should be here but it was not working when we last tested it.
Except for the coffee table problem those are quite decent. Obviously you have to have the backwall absorber in place.
I had a client who had a 4 ft by 4 ft solid walnut coffee table between his speakers and the listener. It was so beutifull that he hardly ever had anything on it. In all the measurements all I ever got was the coffee table reflection which swamped everything else. I couldn't get him to move it. Oh well, we all have our priorities.
I had a client who had a 4 ft by 4 ft solid walnut coffee table between his speakers and the listener. It was so beutifull that he hardly ever had anything on it. In all the measurements all I ever got was the coffee table reflection which swamped everything else. I couldn't get him to move it. Oh well, we all have our priorities.
Actually it would be possible to build an absorber that functions as a coffee table but putting anything on it would shoot down functionality...
markus76 said:
I think it might be fun to build a coffee table out of a binary difraction grating on top of a wood 2D quadratic residue or primitive root diffuser.
The lower frequency effectiveness of something like a binary difraction grating is set by the depth of the wells I believe, so just like an absorber they would have to be very deep to do anything to the bass.
Lose the coffee table and put drinks on side stands next to the couch.
edit: oops, I'm thinking of a quadratic residue diffuser. Too many strings of tech words to keep track of.
Lose the coffee table and put drinks on side stands next to the couch.
edit: oops, I'm thinking of a quadratic residue diffuser. Too many strings of tech words to keep track of.
How?
Besides the fact that we don't want to have even a minimal reflection at that point in time (< 1.5 ms) a quadratic residue diffusor would have to be very deep (depth > 60 cm) to be effective from let's say 300 Hz on. Don't know how big a binary amplitude diffusor would need to be.
Looks like Angus has the answer:
"Two Dimensional Binary Amplitude Diffusers"
Paper Number: 5061 AES Convention: 107 (September 1999)
Authors: Angus, James A. S.; D'Antonio, Peter
Best, Markus
- Status
- Not open for further replies.