Um, you're talking to a guy who does color for a living. The above sentences are basically nonsense to anyone versed in the craft. Sorry.
Funny, I got that information from a guy that does color, HD and Ultra HD projection design, and is basically a visual Guru in A/V circles. So who should I believe as a sound guy(visual are not my expertise), you or him?
Perhaps I should say sorry to you as well.
Oh, yeah? What color crayon? Do you have a reference?
I'm fine with that. Seems we got off topic quite a few posts back. Although I don't think we are that far off topic, you need to know your goals for speaker design.
As for the original topic and post - I'd love to hear a report of what happen when this is implemented. I tried something similar last year out of curiosity and implemented a reverse Fletcher-Munson curve as my target response. It sounded awful. Would be interesting to know if someone else had better results with a similar approach.
I'm fine with that. Seems we got off topic quite a few posts back. Although I don't think we are that far off topic, you need to know your goals for speaker design.
As for the original topic and post - I'd love to hear a report of what happen when this is implemented. I tried something similar last year out of curiosity and implemented a reverse Fletcher-Munson curve as my target response. It sounded awful.
Would be interesting to know if someone else had better results with a similar approach.I tried something similar last year out of curiosity and implemented a reverse Fletcher-Munson curve as my target response. It sounded awful.
Why reverse? You would need to make the EQ curve follow the Fletcher-Munson curve not its inverse.
This is what Audyssey Dynamic EQ does depending on master volume:
To "make up for" the ear's frequency response. What I applied was basically one of the curves you show. I just called it reverse because it's the inverse of the ear's response, if you see that plotted. Inverse is the correct term.
IIRC, I tried to flatten for the mid 80s dB SPL. I sure did not like the way it sounded. Possible that I did it wrong, that can happen.
That is of course the point of the debate. Is the proper reference the acoustic event in a real space or the sound of whichever studio got the contract for the mixdown? If as some here frantically claim the former is an impossibility in principle then fill your boots; Genelec, JBL 4430, NS10 or whatever flavor tickles your fancy.
Oh, yeah? What color crayon? Do you have a reference?
I'm seeing a photo in my mind of a blue crayon, but its only 12bit colour (Canon 5d classic) and I haven't colour corrected my memories with my Datacolor Spyder lately
Seems we are back on track - talking about correction for our ears
Hi Pano, all
The reason compensating our level dependent hearing response or it’s other “flaws” isn’t very helpful is that unlike any measurement system which compares to “perfect” or some ideal, our imperfect hearing IS our only frame of reference, the only thing we have ever known.
Dick Heyser wrote of this regarding measurements too saying we assign frames of reference to make our job easier but that relationship may not exist in nature.
Indeed our hearing is our only direct frame of reference and it consists of many things beyond sensing pressure at two points in space (and why two microphones only scratch the surface capturing something our hearing system recognizes as real.
Unlike a response curve, this is a dynamic map that includes our ears response at any level within a large span.
Unlike a response curve this is a multi-dimensional map that includes all the things your outer ears do to the signal that we interpret as up/ down left / right and close / far.
It was those directional outer ear alterations my friend Doug had identified in the 80’s and used to make that LEDR directional hearing test recordings I had linked to.
If you have not heard this before, do check out LEDR.
These aural q’s are a routine part of TV stereo now days.
Online LEDR Sound Test | Listening Environment Diagnostic Recording Test
Our low frequency response is rolled off by the leakage of our Eustachian tubes which progressively equalize the differential pressure (just like a vented box does if you were using it as a microphone) and it’s a good thing because if it didn’t changes in weather or altitude would deafen us. It is that which requires 20Hz to be about 100 million times louder than 4KHz just to sound about the same to us.
Here is another case of a distorted frame of reference too. In the early days , Dolby labs (I think two guys Benjamin and Fielder?) did an extensive study on subwoofer distortion. The found that due to the shape of the ears response curve, that at 20Hz, something like a 7% 3rd harmonic had the same apparent loudness as the 20Hz fundamental and that theme was born out in testing as well. Bad news for subwoofer makers??
Well no, due to perceptual masking as well as our frame of reference of only knowing audible distortion and with music that it is mostly harmonic in structure anyway (leaving the odd harmonics that sound bad), we routinely hear very high levels of lf distortion (compared to any electronics) and we are unaware of it. In some musical forms VERY high levels of distortion are what gives some “car bass music” that unique sound.
Air itself rapidly absorbs hf energy so one can see why our hf response rolls off leaving a couple octave wide area of maximum sensitivity and acuity. It is also why Earl Geddes concluded we really don’t ear hf distortion very easily as well. How loud would the 3rd harmonic of 5KHz have to be, to have an equal apparent loudness to the fundamental?
Lastly, a rarely discussed but powerful factor is that we not only hear what a loudspeaker radiates BUT also in some ways HOW a loudspeaker radiates.
I don’t know how to demonstrate this other than hearing it, while live first hand is powerful, the effect is captured on a camcorder too.
Here is a video a customer took and it posed the question, since these speakers have essentially the same response and sound very similar what makes the spatial part so different?.
One is a single point in time and space radiation system (2way synergy horn), the other is a multi-way source of individual radiators. Neither of these are hifi speakers and the SM-80 is not one of our most revealing speakers but even so, the difference in the source radiation is very clear. Try this video he sent with headphones.
https://www.youtube.com/watch?v=iJfH0pZve48
Consider that what one hears here is caused by small differences that reach the two (closely spaced) microphones built into the video recorder.
How a speaker radiates (even anechoic) can have a powerful effect on how it sounds, not just the frequency response or phase and so on.
Best,
Tom Danley
Physics, not Phads
The reason compensating our level dependent hearing response or it’s other “flaws” isn’t very helpful is that unlike any measurement system which compares to “perfect” or some ideal, our imperfect hearing IS our only frame of reference, the only thing we have ever known.
Dick Heyser wrote of this regarding measurements too saying we assign frames of reference to make our job easier but that relationship may not exist in nature.
Indeed our hearing is our only direct frame of reference and it consists of many things beyond sensing pressure at two points in space (and why two microphones only scratch the surface capturing something our hearing system recognizes as real.
Unlike a response curve, this is a dynamic map that includes our ears response at any level within a large span.
Unlike a response curve this is a multi-dimensional map that includes all the things your outer ears do to the signal that we interpret as up/ down left / right and close / far.
It was those directional outer ear alterations my friend Doug had identified in the 80’s and used to make that LEDR directional hearing test recordings I had linked to.
If you have not heard this before, do check out LEDR.
These aural q’s are a routine part of TV stereo now days.
Online LEDR Sound Test | Listening Environment Diagnostic Recording Test
Our low frequency response is rolled off by the leakage of our Eustachian tubes which progressively equalize the differential pressure (just like a vented box does if you were using it as a microphone) and it’s a good thing because if it didn’t changes in weather or altitude would deafen us. It is that which requires 20Hz to be about 100 million times louder than 4KHz just to sound about the same to us.
Here is another case of a distorted frame of reference too. In the early days , Dolby labs (I think two guys Benjamin and Fielder?) did an extensive study on subwoofer distortion. The found that due to the shape of the ears response curve, that at 20Hz, something like a 7% 3rd harmonic had the same apparent loudness as the 20Hz fundamental and that theme was born out in testing as well. Bad news for subwoofer makers??
Well no, due to perceptual masking as well as our frame of reference of only knowing audible distortion and with music that it is mostly harmonic in structure anyway (leaving the odd harmonics that sound bad), we routinely hear very high levels of lf distortion (compared to any electronics) and we are unaware of it. In some musical forms VERY high levels of distortion are what gives some “car bass music” that unique sound.
Air itself rapidly absorbs hf energy so one can see why our hf response rolls off leaving a couple octave wide area of maximum sensitivity and acuity. It is also why Earl Geddes concluded we really don’t ear hf distortion very easily as well. How loud would the 3rd harmonic of 5KHz have to be, to have an equal apparent loudness to the fundamental?
Lastly, a rarely discussed but powerful factor is that we not only hear what a loudspeaker radiates BUT also in some ways HOW a loudspeaker radiates.
I don’t know how to demonstrate this other than hearing it, while live first hand is powerful, the effect is captured on a camcorder too.
Here is a video a customer took and it posed the question, since these speakers have essentially the same response and sound very similar what makes the spatial part so different?.
One is a single point in time and space radiation system (2way synergy horn), the other is a multi-way source of individual radiators. Neither of these are hifi speakers and the SM-80 is not one of our most revealing speakers but even so, the difference in the source radiation is very clear. Try this video he sent with headphones.
https://www.youtube.com/watch?v=iJfH0pZve48
Consider that what one hears here is caused by small differences that reach the two (closely spaced) microphones built into the video recorder.
How a speaker radiates (even anechoic) can have a powerful effect on how it sounds, not just the frequency response or phase and so on.
Best,
Tom Danley
Physics, not Phads
That makes immediate sense to me, say no more. Thanks for putting it in so few words.
Um, you're talking to a guy who does color for a living. The above sentences are basically nonsense to anyone versed in the craft. Sorry.
Indeed. I used to be able to tell exactly what kind of moving light was used in a show at a glance by minute differences in colour temperature and rendering. It's a learned skill. Now I can't, I'm out of practise, so it can be unlearned as well.
I came to the same conclusion independently.for listening to binaural via speakers.
Why it works is speculation, since HRTF received wisdom is a bit iffy IMO.
I don't know how my multimiked dummy head works either, although I know why I did it that way.
My recent bbird recordings have an extra two mics compared to the earlier ones, making a total of 'lots of mics'.
You forgot to include the Basilar Membrane of which Mr. Knight wrote about here earlier and was laughed off the board with his claimed solution to solving the natural dip in our hearing response.
http://www.diyaudio.com/forums/multi-way/229122-basilar-membrane-crossover-distortion.html
The Cereni Holophonic binaural demonstration is the most eerily believable recording I've ever heard. It includes a very solid reproduction of a shaken matchbox moved from slightly behind the ear down to the floor and back overhead. Not data but certainly a convincing example that it's possible.
Not exactly what we were discussing. Please see http://www.diyaudio.com/forums/mult...future-loudspeaker-design-15.html#post3498181
Apart from that, yes, the recording you mentioned did work pretty well over headphones. Normally I don't get very good results with HRTFs other than mine.
Ah! I have been using the wrong sort of matches all along.
Freesound.org - "matchbox dropped in bin.flac" by dwareing
Now I need an italian match shop.. in Wales.
Not to mention that the problem of "how we hear" is so much more complex than just frequency response. Frequency response at what level? What about differences in recruitment for different individuals? these things are all taken into consideration in hearing aids, but that goal is a different thing.
Why not make all loudspeakers flat and let hearing aids correct the failures in the individuals hearing? That makes more sense to me and works in practice. I know because I have seen people with very bad hearing fitted with hearing aids and they say that music listening has improved immensely?
Why modify the speaker for the individual? Makes no sense.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.