I would bet they were reasonably accurate up to a point where the cone starts to break-up. After that it's total choas so nothing would be able to accurately predict what happens. Another issue of course is the data set used. The ones in LEAP are more for cabinet interaction and baffle placement. But I have not really dug into it. I mostly use Crossover shop so I am not as familiar with Enclosure Shop. I use primarily Bass Box Pro for my box program.
I am not sure, I think the program assumes only pistonic diaphragms. You can change diaphragm shape as well. Another thing to check, I think they only have a canned on axis response for the woofers and derive the polar from that and your cabinet.
Rob🙂
I'd bet you are right prior to break up. But the break up character ends up being critical to crossover design unless you are using broad dispersion tweeters crossed well below the woofer's break up.
Dan
Dan
In the frequyency region before breakup there are significant polar response effects due to cone shape and a non-rigid cone that is flexing but not really breaking up. Both of these effects strongly influence the polar pattern as the pattern begins to narrow, but well before it "breaks-up". This is clearly where the sims are having trouble when compared to real loudspeakers. Most likely the sims are assuming a flat rigid piston, because its so simple. The more complex polar patterens seen in reality can get quite complicated.
Let's look at the filtered CSD addressed here first:
Metal Cone
Poly Cone
Note that the initial part of the CSD of the metal cone drops faster than the poly cone. What kind of sonic difference this makes, one has to implement and listen for him/her self. Normally I would also look at the 0.4ms total scale (1 sample in SE) for better comparison, but it seems to me the metal driver is going to have a cleaner sound that reveals more detail.
However, we now look at the metal driver without the filter.
note that the initial part of the CSD drops much faster? I think the filter like that is not of best design.
I will try to address the other parts of your post a bit later.
I don't have polar plots for this one...Take a look at this tweeter and its measurements courtesy of Zaph Audio:
Nuance TW5-120LR 1/2" Mylar Tweeter | Parts-Express.com
...
Dan
Beston RT003C Round Ribbon Tweeter | Parts-Express.com
But here is a CSD in comparison. I use it above 10KHz.
An externally hosted image should be here but it was not working when we last tested it.
Again, what is the sonic difference? Gradually you might be able to find a relationship after have gone through the process enough.😉
Accuracy of sims depend on how it's modeled. I have posted various sims and measruements compared in the Geddes waveguide thread.
The point I was making about the CSDs is that there is little difference in the CSD post crossover. Little difference on something difficult to hear has to be pretty far down on the list of what to measure.
Dan
Dan
With this view in mind is anyone up to the challenge to prove that keeping all drivers in phase with each other is not the most important thing?
This is a "prove me wrong or otherwise I am right view." In science we call that theory (hypothesis) and not results. Yes I know of lots of work in the area of directivity control likewise there is lots of work suggesting keeping drivers and signal in a single phase (to some positive phase) is the most important. I have seen and tested not much which suggest directivity control is not one of the top issues and I can say exactly the same thing about keeping all transducers in (sometimes taken as positive) phase. One does not exclude the other as being in the top five. To claim one of these two is all vastly the most important seems to me to declare likewise all else is not nearly so. Even I have made this error as phase and directivity control are both very important. Both without excluding either.
Is flat frequency response important and belong in the top 5 most important issues? What many say is octave to octave balance and apparent bandwidth center is more important than flatness. I agree.
Designs here offer directivity control, octave to octave balance (based on Bark frequency intervals), and uniform positive phase response between all drivers. This keeps the main lobe of radiation always on the same axis which Linkwitz believes is very important in his landmark early papers.
Correctness is not a process of exclusion, it is a process of inclusion. Different aspects should be included which is why there is not a Unified Field Theory in physics. Russian's and Westerners cannot even agree on the fundamental forces at work with Russian's claiming electric charge and Westerner's claiming gravity. Is audio really very different from this?
As far as Earl Geddes is concerned, he is just another voice with his writing style not to my pleasure and his post not once useful. I have learned a lot of views and information on DIYaudio and not one bit (sadly) came from him. I came here to learn and with very infomative post from the hard work of dantheman and soonsgc and many many others it truly has been a learning experience in both viewpoints and aspects of audio importance and methods not previously considered and certainly not so deeply as some of these hard workers have been so kind as to contribute to these threads. Reading old post is mostly what I do here now as there is so much!
Thank you all hardworking contributors
for this vast amount of information and experience outside the 
ivory (and sometimes inside) the ivory tower league. Keep it up!
Would anyone hazard a list of the top five most important things and not label them most important to least important? I have name three I believe belong on the list- help and name more.
=SUM
"You can lead a horse ... to water, but you can't make him drink."
People should read Toole and Olives work which shows frequency and polar response to be the vast majority of the subjective impression. Phase doesn't even come into the picture. Some seem to claim a knowledge of AES and yet they are completely ignorant of its significant and pertinent publications.
See, now that was a "cheap-shot". 😀 Rude, Impolite, etc.. and yet it was innocuous.
You might consider that the next time you take exception to others: i.e."... was not polite and I simply will not deal with people who are not polite. " 😉 (..referenced from a recent post in the waveguide thread with Dennis/catapult.)
BTW, at the very least your posts have a polarizing effect that helps with "spurring-on" others to produce more "palatable" information.
Also, what is useful or not is relative to that user.
To keep up the analogy..
Perhaps some "horses have been to the water, have had their fill, and are looking for greener pastures". 😛
I have yet to read a post that says polar measurements are not important. Up to 45 degrees is important and has an extensive following from anyone building waveguides these days then after that its up to each speaker designer to worry what matters. Again, when building a speaker you can measure up to 180deg if you choose too! No one is telling you not too.
There is no challenge here because there is no arguement about polar measurements and its a silly challenge to start with. The importance of them is just a function of the specific build, near field monitors need on axis accuracy, off axis not so important. Full range driver/speaker builds, have complete crap off axis sound so are you going to tell them their speakers suck because they do not make polar plots their top priority??
Your claim actually is that 60 deg or greater off axis response measurements on woofers is extremely important. You use one woofer as your example. If you want to act like a science guy then you should maybe shows 10 or 100 woofers with 60 degree issues that do not show up first in 45 degree measurements. Then you will actually have a case for 60 degrees mattering to most speaker builders. Better yet, show me the succesful speaker builds that had flaws at 60 degrees because they ignore the 60deg off axis measurement.
Thousands and Thousands of speakers are designed yearly comercially and in the DIY world. THere are many, many, many success stories. Many ways to design great speakers too. Measurements are for speaker builders, J6P will never give a crap about themeasurements that matter. They will never want to educate themselves on it either. Only those in DIY forums or audio science forums where there is an audio science focus will be able to educate the members and still only a small percentage of them will be objective instead of still being completely subjective.
In then the discussion about accuracy, measurements, etc mattering to the end users is really about 1% of the audio purchasing world.
Obviously we should just stick to measurements that matter to the speaker builder and the measurement priorities will always be based on the goals of each individual project.
Expert builders like Jed, Curt C., Zaph, Jon Marsh, etc aren't jumping out of their seats over this off axis banter. Do you really think they really settling for lower quality designs?
Heck, is there even a person posting in this thread that is publishing complete speaker designs with crossovers for others to use??
Last edited:
Since it seems you have read this work, may I ask a quick question?
Do they include room spectrum decay type of information in the study?
The reason why I would ask is that in any report that either does not reveal audibility of certain things, or tests seem possible that other factors might effect outcome, I like to ask more detail about it.
Actually I think Dan was stating that it is the *most* important:
"No one up to the challenge? That what that was meant to be--a challenge to all who don't believe that the polar response is the MOST important metric in speaker design. Show something to the contrary."
(..all-caps emphasis added.)
..And sure I'm up for the challenge. 😀
There are plenty of successful well-regarded speakers that have retched vertical polars, and pretty poor horizontal polars. Many speaker designers before the 90's didn't do polars, also had poor polar responses, and yet were still considered quite good. Alternatively many loudspeakers today were designed with polar measurements, have good polar responses, and yet don't sound that good.
This isn't to say that polar measurements aren't good, or even excellent, for designing loudspeakers, but rather that they don't rate as being the "most" important in the design process. (..and in fact I don't think anything rates as "the most important" with respect to measuring, rather it *can* all be important.)
"No one up to the challenge? That what that was meant to be--a challenge to all who don't believe that the polar response is the MOST important metric in speaker design. Show something to the contrary."
(..all-caps emphasis added.)
..And sure I'm up for the challenge. 😀
There are plenty of successful well-regarded speakers that have retched vertical polars, and pretty poor horizontal polars. Many speaker designers before the 90's didn't do polars, also had poor polar responses, and yet were still considered quite good. Alternatively many loudspeakers today were designed with polar measurements, have good polar responses, and yet don't sound that good.
This isn't to say that polar measurements aren't good, or even excellent, for designing loudspeakers, but rather that they don't rate as being the "most" important in the design process. (..and in fact I don't think anything rates as "the most important" with respect to measuring, rather it *can* all be important.)
Note that all these measurements from Zaph's site and from me are done using SoundEasy. Also note that the difference in the initial drop of my data? Well, looking at this data in a smaller time frame, like 0.37ms. If you can get the initial drop difference somthing like 10db or more, then maybe you will start to get the feeling of what I hear. Also bear in mind that if we do enough little things to provide an accumulation of improvements together, then the results will also be more audible. We can never determine what each individual will hear, the only thing we can do technically is to try and pick the right combination, make the implementation cost effective, and hope that others like it.
I think the polar response is useful for two things:
Determine whether you have a larger sweet spot, and help determine how you want to treat a room. It has very little use to determine how it sounds because there are so many other more dominating issues. For example, if the final system has one outstanding breakup mode visible in the CSD, then the sound is going to be VERY annoying at certain parts of the music. However, when you start having enough of these modes in a system, it will not sound as annoying, and sometimes may be quite pleasant. This is very similar to exciting room modes like the Geddes multi sub approach.
Note that I use the word "pleasant" rather than "realistic". So sometimes this is a choice the user and designer have to make.
George
Floyd Tooles book covers all of this, and I happen to know that its available in China on line, bootlegged! I saw it there myself. Or there is the ethical route - Amazon.com. All that I can say is that if you want to consider yourself "well read" in this field, then you have to read this book. There simply is no substitute. If you don't care what people who have studied this stuff every day of their entire life have to say, then skip it.
That said, I don't agree with it all, but the small extent that I don't agree with isn't worth discussing if you aren't on board with the main points. And his main point is that frequency response in multiple directions is just about all that matters, and, that measurements do tell the story. There is no magic, its all there in the measurements.
PS. He, like me, doesn't even measure phase, let alone claim it's a significant factor in sound quality.
Earl,
I never consider myself "well read", but I do find the information I need to decide what to work on for further improvement. Although it seems that many views of mine are different from others, but come to think of it, all the real achievers are like that. 😉 Like Warren Buffett's views on investment, which I truly admire how he finds his own methods integrating views from others. It's also made openly known that I started down a path to study wave guides integrating view from you and Jean Michel plus some influence from Howell and Newmans book.
In order to avoid misunderstanding, does Floyd Toole have the room spectrum decay characteristics singled out, when he talks about frequency response in all directions of the speaker? You know, if the answer is yes, and I get the book and can't find it, I'm going to ask what page it's on. 🙂I have already ready lots of Floyd Tooles AES papers, and have never seen that information.
I never consider myself "well read", but I do find the information I need to decide what to work on for further improvement. Although it seems that many views of mine are different from others, but come to think of it, all the real achievers are like that. 😉 Like Warren Buffett's views on investment, which I truly admire how he finds his own methods integrating views from others. It's also made openly known that I started down a path to study wave guides integrating view from you and Jean Michel plus some influence from Howell and Newmans book.
In order to avoid misunderstanding, does Floyd Toole have the room spectrum decay characteristics singled out, when he talks about frequency response in all directions of the speaker? You know, if the answer is yes, and I get the book and can't find it, I'm going to ask what page it's on. 🙂I have already ready lots of Floyd Tooles AES papers, and have never seen that information.
Last edited:
The problem is Doug, Soongsc, and everyone else reading or posting the thread, that in blind studies, on and off axis stuff has been demonstrated to be the most useful metric--the whole thing, not any little part. Far more than just the plots that I or Dr. Geddes do. The other fact is that these plots that we do (more his than mine--think what you will of the man, but please by all means demonstrate he is wrong or just lower your pride and admit he is right) tell us the most we can and reasonably do without doing what Harmon is capable of. I have now actually done further, but they have not shown me anything I didn't already understand--which is nice to know. Less is definitely less however. Read my posts earlier in this thread and it is still obvious why. Look at Dr. Geddes's demo on his site and see more reasons why.
BTW, I don't believe him b/c he states things in a way that many find abrasive.
I believe him b/c he has a lot of studies to back up his position. If you have none, I have a lot of trouble believing you. If you have one well conducted one relative to listening to any recording other than test tones, I'll concede you have a point.
Successful, well regarded speakers are a dime a dozen. Know this and digest it. Almost any speaker is well regarded by someone who sees it while it plays--especially the audio press. The truth is that universally respected loudspeaker speakers in DBTs (a very telling method) all share the same characteristic--smooth polar plots. Hmmmm😕
Designing for one axis is of limited use even for near field. Every movement of the head may well cause significant differences. You won't know if you don't measure. Get headphones if near field is what you are looking to do and move all you want. The fact that I've shown that every woofer I have has a changing response with every axis shows that predicting a response isn't easy and that I've demonstrated that it can change the handoff region radically on any axis as well (which should be obvious) shows that these measurements are necessary. They all do worse beyond 45 degrees. The statistics of 8 now are not that significant, but they do indicate a trend to the savvy or moderately observant. The fact that diffraction is more readily observed well off axis also indicates a need to measure well off axis.
Nothing stated has even suggested to measure anything else for any useful purpose--though I do believe that more is nice to know for the uber speaker. My belief is no better than yours(whoever you are). The evidence, well I haven't seen any contradictory to my point. The fact is, most do not design to the competent standard. Don't tell me you have the uber speaker if it doesn't do the basics. A Yugo with a V8 will not compete with a straight six Porsche. It's just that more measurements of less use do not mean a better design. If they are ignoring the most important demonstrated factor, what is the point? Why go on if the basics are not correct. I have to say it seems silly and I have no intent of insulting anyone. Can anyone site one thing that refutes my position? I'm still asking earnestly.
No one has yet to show one study of more significance, or of any significance other than that of the polar response. So why not start there?
This is becoming clearer as those who believe otherwise how no proof. Please some one show me otherwise.
We're still at this and the horses are dry as a camel hind end in a sand storm.
Implications: I had to through that out there. It is not meant to make anyone unhappy, upset, or in anyway emotional other than possibly stimulating a sensation or a cognitive experience that may excite a slight up-curling of the lateral aspect of the medial anterior upper os, chuckle, or outright eye rolling. 😀
It's getting heavy in here. No one wants to sight any studies? You just refute what has been demonstrated without any meaningful refutation? I believe is not meaningful. Neither is someone else believes. Has everyone read Dr. Toole's book? Raise your hand if you have. It has a lot of research in it. Much cheaper than buying all the papers sited in it from the AES.
I do appreciate SUMs argument though I wish he hadn't tried to make that post personal w/ Dr. Geddes. SUM, read Dr. Toole's book and you'll know your argument doesn't work, site meaningful to the point at hand (audibility with recorded material other than test tones) references for your position, or conform to my position.(That is a joke as I'm sure you know but others may not. To read some peoples comments, it may seem I'm unreasonable or perhaps even hostile🙄) Stating conditions for all statements seems silly but it has to be done or audiophile emotions will flare like hemorrhoids on cold pavement. I don't even want to get into the replies to the personal replies about personal replies.... Oy Vey. I must have missed a point that caused some one to get upset somewhere, but try not to make it personal. Man made the audio, not the other way around. I believe we are all well above that--even though I say that without supporting evidence at this point. Hopefully no one tries to prove me wrong on this one.
If you can't site references or there are no references for your belief, please do some studies to prove your belief or find the research that supports your belief. There is no other way that it is reasonable for anyone to believe your point. The other option is to refine your belief system to conform to what has been demonstrated. The only other option I can think of is to live in your own world that is not based in any known reality. Be careful who you let know if this is your decision. Padded rooms still exist(again a joke).
All humor has a bit of truth in it,
Dan
Do you guys want me to stop kicking the dead horse yet? My leg is tired and my foot is sore.(again I'm joking)
As far as experts not understanding this--I never called them experts. Neither should you.
BTW, I don't believe him b/c he states things in a way that many find abrasive.
I believe him b/c he has a lot of studies to back up his position. If you have none, I have a lot of trouble believing you. If you have one well conducted one relative to listening to any recording other than test tones, I'll concede you have a point.Successful, well regarded speakers are a dime a dozen. Know this and digest it. Almost any speaker is well regarded by someone who sees it while it plays--especially the audio press. The truth is that universally respected loudspeaker speakers in DBTs (a very telling method) all share the same characteristic--smooth polar plots. Hmmmm😕
Designing for one axis is of limited use even for near field. Every movement of the head may well cause significant differences. You won't know if you don't measure. Get headphones if near field is what you are looking to do and move all you want. The fact that I've shown that every woofer I have has a changing response with every axis shows that predicting a response isn't easy and that I've demonstrated that it can change the handoff region radically on any axis as well (which should be obvious) shows that these measurements are necessary. They all do worse beyond 45 degrees. The statistics of 8 now are not that significant, but they do indicate a trend to the savvy or moderately observant. The fact that diffraction is more readily observed well off axis also indicates a need to measure well off axis.
Nothing stated has even suggested to measure anything else for any useful purpose--though I do believe that more is nice to know for the uber speaker. My belief is no better than yours(whoever you are). The evidence, well I haven't seen any contradictory to my point. The fact is, most do not design to the competent standard. Don't tell me you have the uber speaker if it doesn't do the basics. A Yugo with a V8 will not compete with a straight six Porsche. It's just that more measurements of less use do not mean a better design. If they are ignoring the most important demonstrated factor, what is the point? Why go on if the basics are not correct. I have to say it seems silly and I have no intent of insulting anyone. Can anyone site one thing that refutes my position? I'm still asking earnestly.
No one has yet to show one study of more significance, or of any significance other than that of the polar response. So why not start there?
This is becoming clearer as those who believe otherwise how no proof. Please some one show me otherwise.
We're still at this and the horses are dry as a camel hind end in a sand storm.
Implications: I had to through that out there. It is not meant to make anyone unhappy, upset, or in anyway emotional other than possibly stimulating a sensation or a cognitive experience that may excite a slight up-curling of the lateral aspect of the medial anterior upper os, chuckle, or outright eye rolling. 😀
It's getting heavy in here. No one wants to sight any studies? You just refute what has been demonstrated without any meaningful refutation? I believe is not meaningful. Neither is someone else believes. Has everyone read Dr. Toole's book? Raise your hand if you have. It has a lot of research in it. Much cheaper than buying all the papers sited in it from the AES.
I do appreciate SUMs argument though I wish he hadn't tried to make that post personal w/ Dr. Geddes. SUM, read Dr. Toole's book and you'll know your argument doesn't work, site meaningful to the point at hand (audibility with recorded material other than test tones) references for your position, or conform to my position.(That is a joke as I'm sure you know but others may not. To read some peoples comments, it may seem I'm unreasonable or perhaps even hostile🙄) Stating conditions for all statements seems silly but it has to be done or audiophile emotions will flare like hemorrhoids on cold pavement. I don't even want to get into the replies to the personal replies about personal replies.... Oy Vey. I must have missed a point that caused some one to get upset somewhere, but try not to make it personal. Man made the audio, not the other way around. I believe we are all well above that--even though I say that without supporting evidence at this point. Hopefully no one tries to prove me wrong on this one.
If you can't site references or there are no references for your belief, please do some studies to prove your belief or find the research that supports your belief. There is no other way that it is reasonable for anyone to believe your point. The other option is to refine your belief system to conform to what has been demonstrated. The only other option I can think of is to live in your own world that is not based in any known reality. Be careful who you let know if this is your decision. Padded rooms still exist(again a joke).
All humor has a bit of truth in it,
Dan
Do you guys want me to stop kicking the dead horse yet? My leg is tired and my foot is sore.(again I'm joking)
As far as experts not understanding this--I never called them experts. Neither should you.
Last edited:
"Blind studies"?
That's not speaker *design*.
The polar response may be critical to overall success for the listener, but *measuring* the polar response is not critical to overall success for the design. BIG difference. 😉
You do NOT need to *measure* the polar response to achieve a particular polar response.
Comprende?
- Status
- Not open for further replies.