I was far more interested in hearing what people measured and what that person could see with the measurement and why that person felt the measurement was important than all this argument about how the measurement was not as good as some other measurement or otherwise less meaningful.
Some measurements certainly cause various characteristics to stand out which can often be a handy feature of a particular measurement. Continuously belittling peoples measurement because they are not the measurement someone else finds important kind of misses the point and leads to a lot of useless argument.
Let people tell there story and ask questions instead of insulting there measurement methods. Ask questions and not tell others their own answers. Some may learn something, nice!
No one here is in the position as an absolute audio authority, no one.
Some measurements certainly cause various characteristics to stand out which can often be a handy feature of a particular measurement. Continuously belittling peoples measurement because they are not the measurement someone else finds important kind of misses the point and leads to a lot of useless argument.
Let people tell there story and ask questions instead of insulting there measurement methods. Ask questions and not tell others their own answers. Some may learn something, nice!
No one here is in the position as an absolute audio authority, no one.
Me too! A multicell should be coming... maybe. Supposed to be less than 3dB of ripple in the main beam lobe which should be good. The one in the picture is 27"x44" at the mouth...big!
Last edited:
Sumadioguy, great posts..I agree and Im sorry.
What about "ETC decay test with shaped sine burst testing "??
I was reading about that somewhere else.
What about "ETC decay test with shaped sine burst testing "??
I was reading about that somewhere else.
MTF is, as far as I know, purely an optics measurement. It was evolved as a better view of lens sharpness than resolution alone. It is a measure of contrast versus spatial frequency. If you had a test target with black/white/black transistions of continuously finer spacing you would find that the contrast from black to white would reduce in contrast as the lne spacing got finer. As such it is the optical equivalent to frequency response, much like measuring the high frequency rolloff of a tweeter.
Early optical meausrements centered purely on resolution: the ability to resolve two close together stars, for example. As such resolution is an end limit of MTF. If I recall correctly, when contrast drops to 5 or 6% we can discriminate no difference (our test target bars have run together into a nondescript gray). Still, it was found that some lenses that had so-so resolution gave "sharper" images. Further testing showed that, while ultimate resolution wasn't the highest with these lenses, contrast at lesser spacial frequencies was higher. From this the measurement of contrast-versus-frequency, or MTF, evolved.
As another analog, you can get MTF from the Fourier transform of the point spread function. No different than FFT of an audio system impulse response.
I know of no special audio application of MTF, since that is already covered by frequency response measurements.
David
MLSSA did have some for of this capability, DRA labs had some AES papers on this, the Newman and Howell book on loudspeakers talked about it when comparing the quality of low frequency performance. I think it can be explored further.
Correct, MTF was first developed for optics where it is a atandard measure of optical quality, but it can be applied to acoustics. It has been used in architectural acoustics in reverberant spaces to determine a form of inteligability. I do not see how it could be applied to a loudspeaker, per see, only a loudspeaker/room combination. It has lost it's favor recently and is mostly one of those legacy attempts at something that just never panned out. Not unlike some of the discussion here.
It's very hard for me to even think of a guess as I don't know enough about the testing condition to even make a guess based on anything. Interesting data none the less and if you'd like to go more into it I'd be interested. I wish the pictures were a bit larger as it's hard for me to see the details.
Did you say previously that softer cones don't perform as well in respect to initial decay and thus a loss of lower level detail? I'm a bit confused as the CSDs I've looked at, the softer cones seem to do better, but you've looked at many more. I'm curious as to how to interpret this data in a way that will be meaningful, but I'm not there yet. I mean I get what's going on with the decay, but the loss of low level detail in poly cones has me perplexed when I look at the available CSDs.
This is very true. Prior to doing loudspeaker measurements, I didn't have much of a vocabulary to describe what I heard--just a lot of descriptive phrases that didn't mean anything in reality. Doing it yourself is the best teacher, but w/o guidance it is easy to go wrong.
Cool approach. It seems you are looking to extract every detail from the recording. That used to be my major concerns as well, but I was never able to get it to load the room correctly and thus spectral balance was always off. That lead to some recordings sounding incredible and others sounding not so hot. Thing that puzzles me now is since my spectral balance is better is improved (read much better polar response), it seems I am able to extract more detail from the recording that previously realized. The old favorites still sound great--even better than before, but things that were unbearable and upsetting now sound better. This was counter-intuitive for me. I thought a horn would kill the magic just like my old Klipsch did. It should obscure detail judging by wavelets, so there must be room for improvement, but will we be able to tell or will that improvement go unnoticed?
Oh, one other thing, that last paragraph is what I meant by "not agreeing with you." I like your approach and the pains and details you go through, but what seems your basic goal hasn't well correlated with what I've heard. I am hoping to learn more from your approach. It does correlate closely with what I believed and desired prior to reading Dr. Toole's book and Dr. Geddes' posts on various sites. SLs site also had a major influence for a while. Right now I'm looking to perfect my speaker's ability to produce finer details--even if I'm unsure wether I'll be able to tell the difference. You never know until you try.
Dan
Last edited:
How much detail I would further go into depends on how much others are doing. 🙂The graph sizes seem fine to me. Perhaps you'd like to elaborate what particular part you find interesting?
Show a CSD of softer cone measurement within the 0.4ms frame with a resolution of about 0.01~0.02ms, then I can explain.
In case you did not notice, I added polar plots to the recommended measurement. This help in determining how the room would be treated.
I often hear people into musical performances complain that they cannot hear enough detail. I myself am also learning what these people are talking about as I get the chance to talk with them and better understand what they think should be there.
Out of these people that you mention, it seems that most first look at data and math to see what should be done. I think SL is more into listening for things that don't quite sound right, then find out what's going on. I'm the kind that am satisfied for no more than 3 months after every improvement that I make. So, when I don't hear a designer express what additional improvements are desireable from a listening point of view, I would start to question whether the priority is to improve design or somthing else.
I often hear people being afraid of "doing things wrong". I think when one enjoys the process of trying to develop something hopefully better than others out there, doing it wrong is part of the learning process. As long as one can understand why it's wrong, then they really learn more than those doing it right by the book.
One of the most memorable moments of my flying experience was landing with the wind at mid night in an uncontrolled airport poping up and down trying to land with the end of the runway in sight. Sure, I was wrong not to check the wind direction before approaching, but my short field landing was perfect.🙂
One of the most memorable moments of my flying experience was landing with the wind at mid night in an uncontrolled airport poping up and down trying to land with the end of the runway in sight. Sure, I was wrong not to check the wind direction before approaching, but my short field landing was perfect.🙂
That approach of measuring with as many different test methods as possible seems like a very good idea. Invent test and use other popular test. Examine all those different results. Maybe something can be determined. I use IASYS, TEF, LMS, RTAs, very slow swept sine (good for spotting buzz and rattles), and many of my own.
I'm having trouble reading the timeframe except w/the Fullrange driver. None the less I can use that to compare the others. What I find interesting is how a driver's changes over time. It would be really interesting to see time and volume and from a metal cone and Polycone. You should start a website w/ all the info you have. Have you considered it? It could be well regarded Like the Driver Vault or Zaph Audio by the DIY community.
That would be great and I'd bet there'd be no need for explanation if I had that data. Your making me want to do some measurements today.
I've thought about building treatments according to polars in the past, but did very little towards that goal. Everything I tried did not perform as well as I had hoped--or even close to. Correcting polar responses with room treatments would make for a cool demonstration if someone could pull it off. Right not I'm just preferring to not create a problem in this regard as fixing it seems very daunting. It didn't take me long to realize it was best left to someone with a lot more time on their hands than I have.
Do you mean recordings or live events?
Questioning other's motivation has always been an interest of mine. I think I'm a CSI at heart. It never ceases to amaze me how wrong others are when judging my motivation. I bet I'm seldom more correct than others.😉
Dan
George
There are limits to that approach. It makes no sense to just keep doing things that others have already done and drawn conclusions about unless there is some very good reasons to believe that their conclusions are wrong.
I'm a big fan of the slow swept sine wave and warble tones for the same reason.
Nice gear SUM!
Dan
I wish I could get HolmImpulse to work on my notebook with XP Pro, but nothing I do can do will install it. #$@&$% PCs 😀.
Anyway, I think the actual tool isn't as important as the operator and the process you use to collect that data.
I have found HOLM to be extremely stable oin everything that I have run it one and I run Win7 64 bit which seems to have the ability to cause almost anything to fail to load. (Once it loads its very stable, but installation can be a real issue.)
And here I thought that having lots and lots of software packages and taking tons of data insured good loudspeaker designs.😉
I'm basically just a smart-alec at heart - if you can't take something serious then make a joke about it. But yes, jokes often go awry on forums like this, and I shouldn't do it. My apology.
No apology necessary! I was happy to joke right back at you.
Yes, sometimes people miss the joke and sometimes it offends, but I am loath to sterilize all my postings just because some people might not read it correctly.
No one has a reasonable expectation to not be offended sometimes in their life and it is prudent to always reexamine what we read and hear to make sure we really understand what we perceive.
So, fire at will, sir! 😀
- Status
- Not open for further replies.