Lynn Olson said:
Lynn
The VC temp problem is difficult to analyze because there is a great deal of complexity. The heat up and cool down both depend on the ambient (mostly motor structure), which can vary considerably. But from what my analysis says, the modulation of the VC resistance is pretty much independent of the ambient temperature, the steady state value of the resistance changes with the ambient, but the modulation doesn't. In other words the VC Re varies about a different static Re with ambient changes, but the variation is strictly signal dependent.
If modulation has a large effect on the sound then its the basic structure of the VC that matters because you can't reduce the modulation with cooling techniques. This has a huge implication. Basically this would imply that the more copper that one has in the VC the lower the thermal modulation. Most high efficiency drivers get their efficiency from more voice coil. A compression driver has a lot more copper than a dome tweeter. A HP pro 15" woofer has a lot more copper than a 6.5" Hi-Fi woofer.
I've been meaning to do some simulations of this effect, but there always seems to be something else to do - like making a living!! Oh to be independently wealthy!!
soongsc said:
To me if it doesn't sound accurate (real) it doesn't sound good. And the almost total lack of cost effective accurate loudspeaker systems is also what drove my designs.
What I am finding is that the high cost of accuarte designs is driven by the market and not by the cost of doing a good design. So few people actually want accuracy that you have to charge a lot to stay in business because there is no economy of scale. The market has gone to people who buy "enjoyable" loudspeakers leaving the "accurate" ones to a very expensive niche.
soongsc said:
What if "accurate" and "enjoyable" are opposing attributes? What then?
To ellaborate a little more - how do you market accuracy in a marketplace that is dominated by flashy. When people listen to speakers that are outrageously bright and boomy and are told that "its what you like that counts", they will always find accurate as "dull and lifeless".
When someone in another thread refered to my speakers as "dull and lifeless" I was tempted to say " Gee Thanks". I mean if he had called them "bright and colorful" I would have been insulted. As Patrick Bateman - whom I respect a lot - stated "Waveguides (or Summas) are a hard sell" because they simply do not grab you. They sound "flat", "colorless", etc. and are almost never received positively on first listening. When people first audition speakers they tend to "listen for something" and when there is nothing there the first impression is "dull and lifeless". But "dull and lifeless" is a very good thing for a speaker to be in the long run. It adds nothing of its own character to the sound because it doesn't have any. But it will never sell in todays marketplace with todays emphasis on the dramatic.
gedlee said:
gedlee said:
I take issue with these two statements, not because I particularly disagree with the idea that some people may prefer systems with certain distortions, or that I would argue with what level of distortion is perceptible, but because it is fundamentally saying that those who prefer some level of distortion wouldn't like the live performance of the same piece in the first place. "I like Brahms' 2nd, but only when played on a system with 3% 2nd order distortion. Played live, not so much."
Maybe we should all have a knob on our system labeled "Distortion" so we could adjust it to our liking?
tinitus said:But "dull and lifeless" is a very good thing for a speaker to be in the long run.
Truely better than too bright, but its only a step on the way and not a goal in itself, I hope
Then what is the goal?
Totally transparent sound reproduction with nothing added by the speaker is my goal and "dull and lifeless" sounds like a good description of that to me. My point is that these adjectives all imply the same thing - that the speaker should have some sort of personality. To me a really great speaker has NO personality. If the sound is dull and lifeless thats because the recording is dull and lifeless, the speaker can't ADD dull and lifeless. Any positive attribute for loudspeaker is IMO wrong.
john k... said:
I don't think that I am saying that exactly, but I would say that many, if not most, would say that the Brahms recording played on a speaker with 3% 2nd order distortion sounded better than the speaker without (assuming that the 3% were audible, which I doubt, so lets say 15% or 25% or whatever it takes). The reference to "live" is problematic IMO. I have gone over that before.
And as absurd (to me) as your last statement is, if you did that and called it the "warmth" control, it would be a huge commercial success.
You absolutely have to seperate preference from perception as they are not the same things, but are continuously linked together. I study perception not preference and I ASSUME (correct or not) that IF an aberation is perceptable then it is wrong. But this aberation could very easily be perceptable AND preferable. I see this all the time and it is the source of endless pointless arguments.
A great speaker that is close to being transparent (none are really transparent other than with some specific program material in special set ups..) can be described as;
"lively and full of colors"
as well as..
"dull and boring".
Both of those expressions are subjective and has personal "values" or "bias" (sorru can't describe my thought better right now).
A speaker (a bad one) CAN add "dull and lifeless", many speakers "add" that to the signal.
And a great speaker does not per definition sound dull and lifeless on first impression. It's not like that at all. If you have a good set up you can easily knock the socks off the listener with a good recording. If the recording is dull and lifeless that is what the listener may experience and if the recording is clear, dynamic and full of life that is what the listener will hear.
edit: small change in parantesis no1
/Peter
"lively and full of colors"
as well as..
"dull and boring".
Both of those expressions are subjective and has personal "values" or "bias" (sorru can't describe my thought better right now).
A speaker (a bad one) CAN add "dull and lifeless", many speakers "add" that to the signal.
And a great speaker does not per definition sound dull and lifeless on first impression. It's not like that at all. If you have a good set up you can easily knock the socks off the listener with a good recording. If the recording is dull and lifeless that is what the listener may experience and if the recording is clear, dynamic and full of life that is what the listener will hear.
edit: small change in parantesis no1
/Peter
gedlee,
What kind of levels of distortion do you consider audible if we are talking class A circuits with dominantly low and even order HD?
I know you have mentioned distortion from electronics being more harsh to the ear due to high order harmonics and crossover distortion (which obviously results in the former) but if these are removed so the transfer curve becomes similar to that of a speaker.. what levels of HD start to audibly color the signal?
/Peter
What kind of levels of distortion do you consider audible if we are talking class A circuits with dominantly low and even order HD?
I know you have mentioned distortion from electronics being more harsh to the ear due to high order harmonics and crossover distortion (which obviously results in the former) but if these are removed so the transfer curve becomes similar to that of a speaker.. what levels of HD start to audibly color the signal?
/Peter
I might be wrong, but what I tend to try is to deliver the best I can in a specific application range, because people are going to look for what fits a need. So right now I'm focussing on smaller speakers because the places I plan to target are places where big houses are probably out of their reach.gedlee said:
I have listened to various horns and wave quides, and the sound varies significantly. The last time I listened to a pair of Avantgards, the equipment driving them changed the sound significantly. What is dry and dull for one person might be expressed as clean sound by another, so it's hard to tell whom will like it unless they listen to them. Also it depends on cultural background. Someone once described my speakers as "a bit boxy", and I really tried to get a feeling what they hear that I might not have been aware of, this let to a much cleaner and accurate sound. At another point in time a person wondered why one set of speakers using cheaper components sounded bor detailed than the that used better components, and that led to internal wire selection process. I continue to find things that I think needs improvement, and always look forward to making it happen. I also visited another also into speaker development, he kept bugging me to comment any difficiencies in his design, I went and pulled out a specific CD and played a specific segment to explain what I thought. There is also another audiophile here that askes every person to identify 5 difficiencies in his system when they first listen to it, I think this is a good method of getting more specific description of what they feel.
The absolute polarity test is one of my favorits, I know that you have expressed opinion that you just don't go there, but have you ever though that this is also preventing yourself from new discoveries? I can say that if you can find a consistent polarity preference for each CD that you listen to, the your speakers might possibly be very clean and live and realistic, because live unamplified performances are never dull and lifeless to me. I know that regardless whose system or which CD is being played, 80% of the time when using my little speakers, if the sound is funny, I swithc the polarity, and whomever is listening with me agree that the new selection is the more accurate reproduction.
Actually, when the speaker cannot reproduce enough low level detail, the it becomes dull and lifeless. I have a friend that has a pair of speakers like that. He has already been asking me if he can choose a pair of speakers of mine that he likes if he can here any difference between different units.gedlee said:
soongsc said:
Have you ever thought that maybe I have done those tests? Why do you assume that because I don't share your opinion I haven't done the work?
Consider a nonlinear system with only a very large 2nd order nonlinearity on one side. Now consider a source, such as Tracy Chapman, which has an extremely high Kurtosis (very high probability of one side of the signal over the other - its not symmetric). Now if I play this through the nonlinear system it will distort one way, but not when the phase is reversed. Could this be heard? Sometimes, it depends. Is it the absolute polarity that you are detecting - absolutely not.
Thank you for this remark. It is exactly the same argument I have used many times (without much result) in discussions about absolute phase. If you can hear it, it is a secondary result of something else (I myself could hear that difference with real speakers, but not with good ESL-headphones). Same goes for even order nonlinearities with amps. One could argue that any even order nonlinearity in the ear itself might also produce the difference (at higher SPL) but I don't know of any sources that can really prove this claim.gedlee said:
- Klaus
Pan said:
I'm going to try and be as tactfull as possible here, but it probably won't work.
Your posts appear to try and draw me into an argument. I'm not going to argue with you although I might answer your questions.
But I am disinclined to answer your posts for two reasons: You are not polite or respectful and you play trump cards. What is a trump card? Its the card that we all have that we can play at any time that stops an argument or discussion on the spot. Its the "It sounds good to me" card (or variations on this theme). When its played there can be no more rational discussion of a technical subject because this is a subjective opinion and there is no rebuttal to subjective opinions. (Well maybe "DOES NOT!!", thats what my son uses.) The conversation stops cold, no more to say.
You played that card on me a while back when talking about room modes. It worked, the conversation stoped immediately. But since you can have an infinite number of these cards, can play them at any time and have shown your willingness to do so, I am not interested in conversing on these terms.
"What kind of levels of distortion do you consider audible if we are talking class A circuits with dominantly low and even order HD?"
There is no answer to this question because there is no metric that I can use that is reliable. Well maybe the GedLee metric, in which case the number that was statistically shown to be the threshold of audibility was 2.0. But I am sure that this answer doesn't help you at all.
Sorry I can't be of more help.
KSTR said:
A big problem in audio is the tendancy to use any rational explaination of the cause of something as factual proof. It takes more than something being a possible cause to make it the real cause.
Correlation of two events does not establish cause and effect.
I have a whistle that scares aware elephants. There are no elephants in Michigan so my whistle must work pretty good.
I'm sure there can be many explanations, but we won't know for sure until we get to a higer level of fidelity. won't we.gedlee said:
If you feel that you have reached a point where no further improvment can ever be made, then I guess life is boring from here on.I would like to see some measurements of the ESL phones. Normally diaphrams of that design tends to have some trailing impulses of quite high level which will mask such effects. I really have a hard time hearing absolute polarity differences with any headphone I use. If I had one of those artificial ears from B&K, I certainly would like to find out why. My guess is that we will see more resonances listening through headphones than listening to speakers. This is usually the most probable cause of masking of many types of distortion.KSTR said:
Probably a way of (partly) circumventing the thermal problem could be to drive the speaker with constant power, not constant voltage (which has been done decades ago, IIRC in the POWRTRON tube amp by S.F.White). That is, using a generator resistance slightly above the DC coil resistance. Then we are near the top point of the power transfer parabola and stay there even if the VC heats up and has increased to, say 150% of the cold value. Effectivly we would swing around the top point and in that region power transfer is way more constant than it is when we move up and down along the steeper slopes as it is the case with true voltage or current drive.Lynn Olson said:
Will need drivers that actually can be run with higher impedance (not showing jump resonance phenomena, especially).
- Klaus
Non linearities is one cause of audible differences switching absolute polarity. However there are signals (even music signals) that gives the same results without the help of non linearity in the playback gear. One Swedish physicist looked into this at the 70's, I'll see if I can find some more.
/Peter
/Peter