actually mostly we have an easier time making measurements, processing massive amounts of data and cleverly displaying the numbers
but quasi linear systems identification and modeling have been major themes for quite a while now
Frequency/Time response duality, scope cameras were "known to the ancients"
storage 'scopes date to 1964 - over 50 years ago
I have seen Volterra Series in an 1950's Mech E Control Theory Handbook
SMTE IMD was a standard >65 years ago:
32. Read, G.W. and R. R. Scoville - "An Improved Intermodulafion Measuring System" Journal of the 45. Waddington, D.E.O'N- "Intermodulafion Distortion
SMPTE,February 1948.
33. Roys, H.E. - "Intermodulation Distortion Analysis as Applied to Disk Recording and Reproducing Equipment" - Proceedings of the IRE, October 1947.
Black's Negative Feedback patent was motivated by wanting in the 1920's to series hundreds of amplifiers in cross country long distance telephony, frequency multiplexing required low distortion to avoid wiping out S/N across dozens of voice bands packed together
Belcher's BBC Noise Fill Test, AP's FastTest are directly descended from Noise Power Ratio testing in Telcom and Radio Communications
Nyquist and others pretty much set down digital signal processing fundamentals by the end of the 1950's
if you are going to invoke "information capacity" you should be aware of (or have already been influenced by) Shannon-Hartley Channel Capacity Theorem
actually a good question would be what fundamental audio measurement today isn't based in 50+ year old theory, even anticipated in practice – perhaps in EE at large if not already audio standards
perhaps some data presentation like waterfall plots? Laser Doppler Velocimetry?
but quasi linear systems identification and modeling have been major themes for quite a while now
Frequency/Time response duality, scope cameras were "known to the ancients"
storage 'scopes date to 1964 - over 50 years ago
I have seen Volterra Series in an 1950's Mech E Control Theory Handbook
SMTE IMD was a standard >65 years ago:
32. Read, G.W. and R. R. Scoville - "An Improved Intermodulafion Measuring System" Journal of the 45. Waddington, D.E.O'N- "Intermodulafion Distortion
SMPTE,February 1948.
33. Roys, H.E. - "Intermodulation Distortion Analysis as Applied to Disk Recording and Reproducing Equipment" - Proceedings of the IRE, October 1947.
Black's Negative Feedback patent was motivated by wanting in the 1920's to series hundreds of amplifiers in cross country long distance telephony, frequency multiplexing required low distortion to avoid wiping out S/N across dozens of voice bands packed together
Belcher's BBC Noise Fill Test, AP's FastTest are directly descended from Noise Power Ratio testing in Telcom and Radio Communications
Nyquist and others pretty much set down digital signal processing fundamentals by the end of the 1950's
if you are going to invoke "information capacity" you should be aware of (or have already been influenced by) Shannon-Hartley Channel Capacity Theorem
actually a good question would be what fundamental audio measurement today isn't based in 50+ year old theory, even anticipated in practice – perhaps in EE at large if not already audio standards
perhaps some data presentation like waterfall plots? Laser Doppler Velocimetry?
Last edited:
I don't understand why you think all recordings would try to recreate a live event. Most recordings are created with artificial sounds or close miking so there is no real live event they could recreate.
Furthermore, stereo is not capable of recreating physical sound fields. The inventor of stereo, Alan Blumlein, realized early on that other mechanisms are at work.
The simplest form of a stereo recording is two microphones capturing the sound at two different locations. Each signal is then played back from one of two loudspeakers placed at ±30 degrees from the listener. Any reflections from the recorded space that would have reached a listener's ears from a multitude of angles is now projected from only two locations namely the speaker locations. Highly unnatural.
Stereo is also a highly artificial listening situation. Phantom imaging exploits an artifact of our hearing in order to make sounds appear at different locations. It is a type of signal that doesn't occur in nature.
A sound mixed to the center is played back from both loudspeakers. The left ear also hears the signal coming from the right speaker and the right ears also hears the signal coming from the left speaker. This is called interaural crosstalk. Two speakers radiating the same signal create an interference pattern. Small head movements cause level and time differences between the ears which leads to image shift and localization ambiguity. It also creates an uncorrectable dip at around 2kHz.
'The bottom limit of the dynamic range'. What about the upper limit of the dynamic range? Shouldn't it be as high as possible otherwise the bottom limit couldn't be far away. Full range drivers are particularly bad at this.
My point was that you need to be able to measure something in order to correlate it to perception. 'DDR' lacks that correlation.
Last edited:
bingo we have a winner.
that's what 'DDR' does. it's basic Marketing 101, eg takes a technical disadvantage and turns it upside down. Changes the whole conversation doesn't it.
I didn't say 'if it cannot be measured it doesn't exist'. I've said if it can't be measured then 'the term is of no value'.
+1. An XY pattern stereo mic is essentially at same location but produces great stereo image. That is what my Zoom H4 recorder uses.
Could DDR be totally subjective? Some drivers are known to have 'tone'... but you must hear it to appreciate it. If someone can't hear it then, for that person, it doesn't exist.
Maybe there's no measurement for tone or DDR.
Could impulse response measuremens be an indication of how well a driver tracks impulses making it more likely to be more accurate (when combined with frequecy respose, of course).
Maybe there's no measurement for tone or DDR.
Could impulse response measuremens be an indication of how well a driver tracks impulses making it more likely to be more accurate (when combined with frequecy respose, of course).
Should be as high as needed to make sure that the source is not compressed. When people talk about dynamic range it is often this end they are actually talking about.
dave
Not yet having a correlation does not make the phenomenom any less real. There is far too little real scientific research that correlates what we hear to what is measured.
Ironically this last makes the use of uncorrelated measures a subjective choice and conclusions similarily subjective.
The whole point of coining the term DDR was to take all the varied terms used to try to describe this into something more technically correct, less audiophile reviewer flowery, and to lead to more measurements that would put an objective spot-light on it.
dave
No. Measurements that focus on quantitatively evaluate this would be a big boon to designers.And i am sure measures are possible, or even exist, They certainly do in electonics as the (similar purposed) term used in the ADC/DAC field shows.
dave
you are smart enough to understand the question
do psychoacoustic codecs reduce the audible/subjective whatever you are calling DDR
if so is it always, or when do the compression algorithms start to impact this DDR
which is interesting given the vague formulation of DDR when you consider that psychoacoustic codecs only use ~6-7 bit per critical band
formal DBT testing drove the codecs to a degree not previously seen in audio development
do psychoacoustic codecs reduce the audible/subjective whatever you are calling DDR
if so is it always, or when do the compression algorithms start to impact this DDR
which is interesting given the vague formulation of DDR when you consider that psychoacoustic codecs only use ~6-7 bit per critical band
formal DBT testing drove the codecs to a degree not previously seen in audio development
Last edited:
Since one of the assumptions used when devising these is to throw away stuff that should be masked, i would say yes, a lossy compression algorithim would almost invariably reduce the DDR of the source material.
I expect they always impact the DDR, but depending on the resolution of the following chain out to the transducer it may still be higher than the device with the least DDR -- probably the transducer in most cases as long as the electroncs power supplies are not compromisesd orpushed past their limits, althou crossover distortion in Class B amplifiers could play a role.
The subject of lossy compression is probably one of the largest sources of real studies of human auditory perception.
dave
I didn't say what you wrote above.
My point is to really determine the accuracy of a loudspeaker we need a known reference. A studio created recording is an unknown because it doesn't exist outside of its electronically stored format in a way that it can be referred to outside of the reproduction chain.
Live music exists acoustically before it is stored, and can be used as the ultimate reference. So a subjective assessment of a loudspeaker against exposure to live music is the definitive test.
Anyway as I said previously there are two different world views doing battle here, and it doesn't seem possible for them to reconcile in a usefully informative way.
