I'm not sure what you're refrerring to but audio reproduction should be like video reproduction. There's only one way to do it right. This might kill the art of creative audio marketing but it will enable the art of virtual aural spaces, music and sound.
Yes.
Grab the sticker if you can... the uSoft police would like to see it on your MacBook. But with W7 out and no support i don't know that they'd care.
dave
Hi FrankWW, thank you for posting the link.
In sung vowels being more energy towards high frequencies and
the formant structure being different to speech vowels is known
and in fact should influence the qualities to achieve in a
loudspeaker, if good voice reproduction is aimed for.
These would be at first:
- low coloration
....- polar dispersion only smoothly varying with frequency (again and again ...)
....- smooth frequency response
- low harmonic distortion
- sufficient dynamic headroom
- sufficient bandwith
To put transient response of a loudspeaker on top of a list
of properties needed for good vowel formant reproduction is
simply misleading.
Please let me explain why, it is quite simple:
The articulatory movements are rather slow.
When singing "mamama" even in say 1/3 second
the duration of the vowels would be about
1/18 seconds. Assuming 50ms for the vowel to
build up and then turning into something different,
for a formant at 8Khz you have about 400 cycles
at that formant frequency ...
It is easy to reproduce the sung envelope at that
frequency even for a tweeter with slight imperfect
decay at the top end. Thats it.
There would be other kinds of signals which would
demand good transient response up to the top
end. Concerning the human voice we have to keep
in mind that listeners even accept a 's' sound as a 't'
if you just shorten (cut way in the middle)
it using a signal editor ...
So even for the transients of the human voice i
doubt, that the decay behavior of a loudspeaker
above 8Khz is the most important thing.
It is truly nice to have, but the properties
mentioned above are by far more relevant for
voice reproduction.
Reproducing cymbals and intruments with extended
overtones may be a different kind of thing.
Kind Regards
Last edited:
Proper BMWs has 2 cyclinders and the cyclinders move horizontally and in opposition 🙂
dave
I've just departed from the Geddes/directivity love-fest where I was pilloried and threatened with a fist-fight for daring to inquire if anybody had a few pieces of EVIDENCE they could point me to. Nada (except the old bluff, "why don't you read through the 4500 posts" in that thread).
I agree completely: you certainly can show ("prove" if you like the word better) that there are no humanly detectable differences, to any degree you want. But your post has to also say that the test has to have what is called "power" sufficient to show any differences that might have been there would have become apparent. For the people who run journals, that typically means nothing other than large samples (and they ignore bad technique and operational sloppiness which also can make an experiment "fail").
Actually, a some of Toole's research is kind of casual and unilluminating and lacking in solidity (and some is very illuminating, if memory serves). And that was before he became a figment of Cosmic Audio Corp.
There is a long history of the see-saw between golden-ear hearing and physical testing. Neither has had a check-mate. The measurement folks always talk engineerese with a quaint 19th century German accent that denies there are humans in the loop. The golden-eared folks seem wholly ignorant of the most obvious psychological influences (or Human Factors, if you prefer the sound of that).
Last edited:
Ben, you certainly didn't expect to go to a thread about waveguides, tell people their evidence wasn't good enough in a snide way and expect someone would listen to your position even though you weren't up to speed on the conversation. Or did you? I think you are a bit too bright for that. Please don't bring that junk over here. No one wants it. Beyond that, I agree with your post. If you read the WG thread, you'd know it's anything but a "love-fest" though some posts do appreciate the level of expertise by many of the contributors. Many of which are here too.
OK, never mind. I'm not touching it. I am afraid of the law.
😱
Markus, I agree. The industry needs tighter standards on both sides--recording and playback. That would surely cause outrage in audiophile circles, but it would awesome if great sound could be guaranteed when you buy a recording.
Cool thing is since I've gotten into speakers with good polar responses, more of my recordings work very well. None are now annoying that I've played. Some of that is going to be b/c I've stopped playing my worst recordings, or what I thought were I'd bet. I'll never know until I try.
Dan
OK, never mind. I'm not touching it. I am afraid of the law.
😱Markus, I agree. The industry needs tighter standards on both sides--recording and playback. That would surely cause outrage in audiophile circles, but it would awesome if great sound could be guaranteed when you buy a recording.
Cool thing is since I've gotten into speakers with good polar responses, more of my recordings work very well. None are now annoying that I've played. Some of that is going to be b/c I've stopped playing my worst recordings, or what I thought were I'd bet. I'll never know until I try.Dan
Last edited:
When Dan referred to the readings already mentioned previously, he closed a second circle in the merry-go-round.
Video is quite interesting. During a conference here discussing the future of 3D video, I pointed out a few things.
1. 3D video has been around for at least 10 years. The idea never took off.
2. Focus, or skew of focus in visuals is very annoying, much more so than poor image focus in audio. Since the focus of camera might not be what the viewer decides what he wants to focus on, the eyes try to focus on something different from what the camera is focused on, then the person becomes dizzy. This is basically a physical limit that probably cannot be fully solved in the next 10 years or so.
3. Based on the above, 3D video will eventually swrink back to the special purpose market such as training.
It was also noted that a great percentage of people are not 3D sensitive. So I pointed out that video will eventually take a similar route as audio as far as market trends are concerned.
Last edited:
I was more talking about color management not 3D video. But you're right, 3D is an interesting topic and it will be of interest as long as humans have two eyes. Same is true for our two ears and virtual auditory spaces. So here some OT remarks on stereoscopy.
Stereoscopy has been around since photography itself. Look up "Kaiserpanorama".
Cameramen just need to control the depth of focus. In animated movies it's a non issue. James Cameron even argued that a limited depth of field is desirable because of aesthetic reasons.
That's a pretty bold claim - do you have any data on that? Never heard that there are people incapable of perceiving the third dimension (unless they have only one eye).
Hey Dan, think twice before you tell an interested newcomer they have no right to ask for a few references to evidence in post 4500 (roughly). A great deal of huffing and puffing ensued and nobody had a useful thing to say except, "read Toole." That's it.
Would I be correct that you are strongly blasting me without having read all 4500 posts that preceded my inquiry?
And just where do you get off thinking I told anybody anything snide or critical. Except that when a person asks for evidence during a love-fest, that immediately marks him as an enemy, eh.
What conclusion would you draw about the Geddes patented speaker?
Last edited:
Yes I have read the thread and there is a lot of useful information in it. What you started an argument over has already been discussed(argued). You really should be well informed about any topic prior to arguing about it. What conclusions I draw are inconsequential. Read this thread and you'll know where I stand. I shouldn't have to reiterate it just for you. Leave that stuff over please. Better yet, leave it at home. No one should have to go through and explain everything from the very start of a conversation just so you don't have to. It takes too much time and seems rather selfish of you to think that's how it should work. I think it's clear to see where I "get off" or more accurately what sets me off. I can't stand people who want to argue w/o bringing anything to the table. It's another waste of time. No sense arguing w/o an argument.
Now please leave your baggage at the door whatever it is, You are mucking up this thread as well and I'm done responding to it.
Dan
Now please leave your baggage at the door whatever it is, You are mucking up this thread as well and I'm done responding to it.
Dan
It's obviously more rewarding to demand training instead of actually doing what's required to gain knowledge.
Hi Oliver,
Actually i think i did not quote the most relevant part of that article because i was too intrigued by the information about the second formant. I didn't do a very good job. It's the 1st formant that's critical and it's usually in the area of around 3000Hz.
Now, the point being missed here and it's relevant to Soonsc's concern, is that singer's formant is a continuous thing: it's produced with every note and it's a dominant part of the singer's total energy budget. (That is why the singer can be heard above the sound of the orchestra even when it's playing loud and it's right in the middle of the most sensitive part of our hearing range.)
In my understanding it's not vowel formation which needs be the focus of attention but the singer's formant which exists "independently".
In the case of sopranos the singer's formant remains even if the vowel can no longer be produced.
We should remember singer's sounds are sustained just like any other wind instrument. This is not the same as in ordinary human speech.
If a loudspeaker has poor decay characteristics in that range, then the colour, timbre of the singer's sound will be modified audibly.
I'm sure the sound of a great singer like Domingo would definitely be different being reproduced by the speaker of Soonsc's top example as opposed to the bottom one.
There are similar problems with reproduction of other instruments with extended overtone structures. For example violins have overtones about equal in energy content to that of the fundamental tone and linear distortions such as produced by poor loudspeaker decay times will introduce colouration to their sound. This is especially the case with solo violin because the soloist is also emphasizing a formant in a manner similar to that of a singer.
(There is German violin manufacturer which has done good research on the relation between measurement of violin sound and subjective experience:
Master Studio for Violinmaking - Martin Schleske Munich, Germany Harmonic Structure )
soonsc's example:
There are other factors in addition to transient response which are obviously important to measure and you have noted some below. I'm not making any claims about what is most important in the general scheme of things.
But I do think one thing that is important is linear distortion which is difficult to measure, although not impossible to infer its relative importance for any given situation, from things like soonsc's CSD's above, or the raw impulse response, or Geddes's foam technique for diminishing waveguides' and horns' linear distortions.
Certainly a very important thing is knowing what our speakers are doing in the area between about 1 and 6 kHz. The more good information we can gather about that, the better.
Actually i think i did not quote the most relevant part of that article because i was too intrigued by the information about the second formant. I didn't do a very good job. It's the 1st formant that's critical and it's usually in the area of around 3000Hz.
Now, the point being missed here and it's relevant to Soonsc's concern, is that singer's formant is a continuous thing: it's produced with every note and it's a dominant part of the singer's total energy budget. (That is why the singer can be heard above the sound of the orchestra even when it's playing loud and it's right in the middle of the most sensitive part of our hearing range.)
In my understanding it's not vowel formation which needs be the focus of attention but the singer's formant which exists "independently".
In the case of sopranos the singer's formant remains even if the vowel can no longer be produced.
We should remember singer's sounds are sustained just like any other wind instrument. This is not the same as in ordinary human speech.
If a loudspeaker has poor decay characteristics in that range, then the colour, timbre of the singer's sound will be modified audibly.
I'm sure the sound of a great singer like Domingo would definitely be different being reproduced by the speaker of Soonsc's top example as opposed to the bottom one.
There are similar problems with reproduction of other instruments with extended overtone structures. For example violins have overtones about equal in energy content to that of the fundamental tone and linear distortions such as produced by poor loudspeaker decay times will introduce colouration to their sound. This is especially the case with solo violin because the soloist is also emphasizing a formant in a manner similar to that of a singer.
(There is German violin manufacturer which has done good research on the relation between measurement of violin sound and subjective experience:
Master Studio for Violinmaking - Martin Schleske Munich, Germany Harmonic Structure )
soonsc's example:
An externally hosted image should be here but it was not working when we last tested it.
There are other factors in addition to transient response which are obviously important to measure and you have noted some below. I'm not making any claims about what is most important in the general scheme of things.
But I do think one thing that is important is linear distortion which is difficult to measure, although not impossible to infer its relative importance for any given situation, from things like soonsc's CSD's above, or the raw impulse response, or Geddes's foam technique for diminishing waveguides' and horns' linear distortions.
Certainly a very important thing is knowing what our speakers are doing in the area between about 1 and 6 kHz. The more good information we can gather about that, the better.
I've worked on and off for a 3D movie concern for the last 10 years, doing 3D requires a whole lot more care than mono.
Care also has to be taken in animations as well, but it can be programmed in.
James Cameron's is just one approach.
IIRC the figure is something like 1 in 10 have trouble or are incapable integrating a 3D image.
dave
Interesting. Are those people just incapable of uncoupling parallax from focus or is their sense of sight incapable of perceiving a three dimensional space?
I know about stereo photography, and still have brochures of such lens adapter and viewing device. Both eyes do not operate equally, each person has a dominant eye, so it's a 50% chance. If you consider that fact that in video, there is switching between the views, ... not so simple getting the all round focus since the human tends to focus on what the viewer wants, etc.
Why do you think fighter pilots had to have perfect eye sight? Ask them what kind of eye exams the older pilots had to take? The incapability of perceiving 3D was reported by another proffessor at the conference, I did not see the need to check because humans are losing capability to perceive 3D in both eyes and ears, simply because we do not need that capability to survive anymore, it's just a luxury. I had taken some flights to experience what it takes and feels during formation flight, and I can tell you, nomatter how well I can keep proper distance during driving, it would have taken me a hundred flight hours to even be able to correctly know how close I was to the other plane and the rate at which the distance was changing. The pilot flying in front later told me that he was really nervous having me follow him.
My car, my home, my boat ... I even own some real steroscopic cameras from the 50s ... You're seeing problems where there are none and your claim still stands unproven. We're training our sense of sight and our hearing every day just by living. There's no way to unlearn spatial perception by living a "normal" life. Do you know anybody that has a hard time grabbing his cell in front of him? Which "conference"? What professor? Fighter pilots are chosen because of their special spatial perception capabilities? You know that spatial perception is limited to about 50 meters because our eyes are only about 65 mm apart?
What you talk about is quite different from 3D perception, but I don't want to get too far OT. Cover one eye and see if you can grap you cell in front of you. No need for 3D perception because 2D projection of 3D provides enough information for the brain to learn. Try to do some formation flying with someone and you can experience how bad your 3D perception is.
Last edited:
I'm one of those people. My eye doctor says it's just that some peoples' brains never trained to combine the images in both eyes together -- I in effect only see through one eye at a time, although both of them work fine. I found out about it because I found I was completely unable to see those pictures made up of dots that people focus on for a minute and an image jumps out. For me, all I got was a headache. She (the eye doc) gave me a test and told me that I have no depth perception, other than what can be inferred through distance and motion. Helps explain why I had so much trouble catching baseballs or shooting baskets (or maybe I'm just a klutz).
I wonder whether there are people who can't hear dimensionally, only hear through one ear at a time?? That I have no trouble with.
There are depth cues but that's the difference between 2D and 3D - most people that try to grab something with one eye closed, they miss. Better test: close one eye, make a fist and hold your hands about 2 ft apart in front of you with your arms bent. Point your index fingers towards each other and move them towards each other so the finger tips meet in front of you. Now try the same with both eyes open. Any difference?
- Status
- Not open for further replies.