And you know that how? Because its not actually true.
I know this is a long a wooly thread, but somehow it seems were are losing the point of it all.
So what exactly is "the point of it all"?
I think that this discussion has gotten more interesting not less. The "Ariel" has been discussed sufficiently that I think we might need to get back to the basics of what actually matters. There is a lot of misconceptions going around here and I, for one, think that it would be a good idea to clear them up. One will never create a great speaker using false ideas and folklore.
Thanks, that's just another piece of "folklore" that needed correction.
I don't understand what you have in mind. I'm a straight person
I was talking about finding the truth. If the question is whether something is audible or not, it should be easy to do. No need to stop at useless conclusion consisting of "by chance" or "belief" or "random". We can have better conclusion than that.
Like in most cases, we need thresholds. I don't remember or know about your standard/threshold regarding this nonlinearity.
I agree with you that for the majority, and also from looking at the most commercial speaker designs, the THD doesn't even enter the big picture...
But people have different preference and limitation. I consider my needs for low distortion system as LIMITATION, not preference. I don't need to listen very hard to know that I feel disturbed or not. Even if I don't want to listen to it, it will come to me by force.
And from where I stand, I think nonlinear distortion of drivers are the most important thing and is a critical bottleneck for me.
Your threshold might be "any drivers using flux modulation ring" or...?
Thing is, Jay, concentrating on the new subset that got it right is not needed and is a waste of time. Unless you have got some very non-normal distribution curve, like maybe a double-header, then it would warrant another look, but anything like a normal distribution is simply evidence that good and bad scores are happening randomly.
If you really want to question the first result, then repeat the entire experiment with the same entire population. For no good reason, as the initial experiment should have involved sufficient population and repetition to permit its stated confidence level. Anyway, the ongoing nature of research will mean that some validation experimentation will be conducted as flow-on experiments are designed. If that process turns up a surprise result, non-validation, then the big questions are asked, re-tested, and resolved. A lot of the science of audio perception has been through this many times, it's routine.
No, no, it is not like that. The objective question was "is it audible?" not "can the majority hear it?".
To answer "is it audible?" you only need to find one ear who can hear it then the conclusion is it IS audible. How it will affect the majority, that's another question [that should be answered with different method than DBLT].
If I know A is deft, I will not include him in my blind test. I will not repeat the entire experiment with the same entire population. I will only repeat the entire experiment with those who have bigger chance to succeed. Because this limited respondent will allow deeper or more controlled and accurate test.
But it depends on the question that you are trying to answer.
Earl, been out all day. I think that we agree fairly well on most of the things that are said but may say them differently. I was just asking earlier where you draw the line at audible and inaudible linear distortion.
In a DBLT the subjects should be able to duplicate their choices fairly consistently or I would question random selection and a poor statistical result. Test setup and test results are fairly well understood in statistical analysis. I won't say there won't be outliers in any testing but they can be incorporated into the analysis.
On that subject of distortion differences between a dome tweeter and a compression driver that has been studied to no end. The dome tweeter will consistently have higher distortion compared to a CD. Just because you have a tilted FR in a compression driver does not equate to distortion or loss of fine details. Dome tweeters have many known issues that can be see in any issue of Malcolm Colloms High Performance Loudspeaker book.
I don't see any way you can look at a CD without a horn and get realistic results. The only other way to test a CD is mounted to a properly made terminated tube that i know of.
In a DBLT the subjects should be able to duplicate their choices fairly consistently or I would question random selection and a poor statistical result. Test setup and test results are fairly well understood in statistical analysis. I won't say there won't be outliers in any testing but they can be incorporated into the analysis.
On that subject of distortion differences between a dome tweeter and a compression driver that has been studied to no end. The dome tweeter will consistently have higher distortion compared to a CD. Just because you have a tilted FR in a compression driver does not equate to distortion or loss of fine details. Dome tweeters have many known issues that can be see in any issue of Malcolm Colloms High Performance Loudspeaker book.
I don't see any way you can look at a CD without a horn and get realistic results. The only other way to test a CD is mounted to a properly made terminated tube that i know of.
So this concludes that basically [in general or at high end level where we are interested in] a CD has lower nonlinear distortion than dome. But when implemented, the horn has a tendency to increase the linear distortion of CD system. So when properly implemented, a CD has a better chance to have lower distortion, hence accurate sound details...
To validate the above, it is logical to compare the best dome implementation versus the best CD implementation, at least in a preliminary effort to answer a question "is perfect implementation of horn possible?". This has been asked by boldname before.
BTW, fine details are nothing else but a very slight almost inaudible sound. If one ear perform badly in a DBLT, can he hear this "detail"??? If I'm not mistaken, Lynn prefers the smooth sound of tweeters without WG. Can all people hear the differences???
Last edited:
But don't waveguides like Earl's OS shapes provide only minor loading to the driver? (I'm going by memory here, Earl correct me if I'm misremembering your words). If the loading isn't a factor, why wouldn't measuring on a baffle be accurate? Not suggesting it must be, just can't see a reason why not.
Has anyone seriously tried to use a CD without any type of horn attached?
The only time I have seen a CD used without a horn was when it was turned around and used as a dome midrange type of tweeter. The throat area was loaded as a rear chamber and the diaphragm was exposed.
But turn the driver around and test what you get when the only horn lens is the short throat section in front of the phase plug and tell me what kind of FR you think you are going to measure.
The only time I have seen a CD used without a horn was when it was turned around and used as a dome midrange type of tweeter. The throat area was loaded as a rear chamber and the diaphragm was exposed.
But turn the driver around and test what you get when the only horn lens is the short throat section in front of the phase plug and tell me what kind of FR you think you are going to measure.
"Minor" is subjective. I remember a hot argument between Earl and somebody who claimed to have heard the Summa [in another country] and heard it's weakness. I have never heard the Summa but I'm not surprised if I can hear the same thing. But the most important thing is that I believe there are much more goodies in the Summa than that little weakness. I have heard similar speakers like the Summa and I found the tradeoff was more than adequate, and Earl is the expert for such speaker so his speaker should be better than that example.
So it is back to the truth in Earl's statement: Things don't even enter the big picture of what is required to make a great speaker...
To build a pair of speakers that has more headroom than a cone dome speaker a certain person likes and retain the quality of "use with this special triode amp" speaker with certain music in a certain room for a certain person. Maybe even some certain people will copy it because they have the certain triode amp and like the same kind of music and think they want the same certain sound as the certain person. I think that's certain.
POOH, since you're not a fan of this "certain person" or the speakers he designs, why not start your own thread and tell the readership about your speaker?
No snark, I'm serious here, and I'm certainly sure you have a good story to tell. If the complete POOH system has been double-blind tested to Dr. Geddes' certain satisfaction, that would be even better.
Last edited:
You surely don't take my humor well.
I feel it's best to build a system like you do. A certain speaker for a certain sound for a certain room. To share those journeys to people that will never need my certain speaker in their certain room with their certain music tastes and amplifier fetishes seems counter productive. I started a thread a while back but after I posted it I felt about it as I do now. Dr Geddes is welcome to bring his contraption over to my room and compare ot with my comtraption anytime. I'll even provide the blind fold and the beer.
You surely don't take my humor well.
I'm afraid the humor doesn't communicate all that well through the screen, I'm sorry to say.
I appreciate that Dr. Geddes' takes a rigorous scientific approach to his audio designs, and that he's written a number of papers on the subject. I'm pretty much in the opposite corner, and aiming for a subjective goal using truly ancient electronics (dating back the mid-Thirties).
It's a minority approach, and almost certainly won't appeal to most DIYer's. The appeal is about as broad as the readership of the departed "Sound Practices" magazine. If the reader is not a fan of the SP approach, there are lots of other speakers (and amplifiers) to choose from.
Man I miss that magazine, even if issues were "sporadic".
-I even liked the look and feel of it (..which of course was determined rigorously through double-blind tests).
To be frank, what interest me in this thread is people are going through the same problems a few of us went through years ago trying to get hifi out of these massive drivers and mega phones. My interjections are honest and are based upon mine and other peoples journeys. I certainly don't have or pretend to have all the answers but I am honest about my experiences. Have you actually heard the Geddes speakers with your amp in your room? I once sent him an email asking if I could listen but got no reply.
bwaslo,
How do you propose to attach the CD to a flat baffle without a flare to transition to the flat baffle? Just poking a hole in the baffle and having the end of the throat flat with the baffle would have a sharp edge causing all kinds of diffraction at the exit of the driver. I don't think that would work at all well in any sense.
How do you propose to attach the CD to a flat baffle without a flare to transition to the flat baffle? Just poking a hole in the baffle and having the end of the throat flat with the baffle would have a sharp edge causing all kinds of diffraction at the exit of the driver. I don't think that would work at all well in any sense.