Yes of course it has to be there (and it is), what I was talking about is the
sensitivity to different stimuli.
What are you trying to say ? Should we cut off our ears and implant microphones ? ;-)
For microphones vs. ears try this: Take a walk with someone nearby a brook
or river and have a conversation. You will have no problems to understand each other. Record this and play it at home; will it be as easy to follow
the played back conversation (if you can at all) ?
Reproduction (and recording) systems are not perfect. If they were
a real violin and a recorded one would be indistinguishable.
I hope we can agree on that (and no one requires a DBT to "proove" that.)
To improve this we just could take the "brute force" approach and try
to optimize _all_ parameters we know of beyond the sensitivity of the
ear brain combination. This has not been sucessfull so far, especially
because it´s not easy to maximize everything at the same time.
Another approach is trying to understand how this ear brain combination
works and what is the _important_ information. Not to loose the most
important part is the what has to be assured first.
To get this right (within todays technical prospects), is something
John is good at, it appears, this is why I read this thread.
a real violin and a recorded one would be indistinguishable.
I hope we can agree on that (and no one requires a DBT to "proove" that.)
To improve this we just could take the "brute force" approach and try
to optimize _all_ parameters we know of beyond the sensitivity of the
ear brain combination. This has not been sucessfull so far, especially
because it´s not easy to maximize everything at the same time.
Another approach is trying to understand how this ear brain combination
works and what is the _important_ information. Not to loose the most
important part is the what has to be assured first.
To get this right (within todays technical prospects), is something
John is good at, it appears, this is why I read this thread.
Huh, the recording part is transparent nowdays, as it been for a while.
Sound producers and speaker-room system aren't.
Headphones get close as long as recording method resembles the reproduction method (binaural) and doesn't gets slaughered by producers (unprocessed recording)... that's why the dsd thing sounds nice sometimes - there are less tools to ruin it.
Try it yourself - two wm61 mikes on your head, average recorder, above average headphone system - sounds rather real to me.
Sound producers and speaker-room system aren't.
Headphones get close as long as recording method resembles the reproduction method (binaural) and doesn't gets slaughered by producers (unprocessed recording)... that's why the dsd thing sounds nice sometimes - there are less tools to ruin it.
Try it yourself - two wm61 mikes on your head, average recorder, above average headphone system - sounds rather real to me.
Some of us appear to believe that an auditory experience draws input from sources other than acoustic vibrations in air, so when these sources are blocked off the auditory pathway is degraded. These extra sources include: the equipment price, the designer's name, the metal/insulator used for connections.mikelm said:
Some of us believe that better measurements would be possible if only we had better data on hearing, rather than anecdotes which disappear under test. Note that the point of a test should be to test the equipment, not the hearers. It is only when the hearers put their reputation (and income?) on the line that 'stress' comes into it.
Some of us believe that some people prefer their sound with certain known distortions but mostly choose not to admit this so therefore have to postulate unknown distortions in sound they don't like.
Some of us are puzzled that equipment with clear design errors creating known distortions gets praised to the sky by reviewers. This appears to confirm the previous statement.
I'll 50% agree. Getting the electrical signal unmolested from the mike to the loudspeaker is a solved problem (the fact that most engineers and producers don't want the signal to be unmolested is a question of commerce, not technological capability). But as long as you can't find two mikes that sound the same, that end of the process cannot be deemed "transparent." Likewise, if we have two perfect microphones but their polar patterns are different, the sound will be different. Which is "right"? Or is neither "right"?
Kindhornman
A waterfall is a compilation from consecutive discrete FR sweeps.
I have never measured a speaker with a waterfall worsening with time.
True, warerfall helps a lot in visualizing the performance.
I would say that a resonance does show up on a FR curve as well as on an impedance sweep, especially when plotted as a Nyquist plot. There, the area of the circle is analogous to the energy of the resonant system, the diameter facing the real axis is analogous to the dissipative element and the diameter facing the imaginary axis is analogous to the non-dissipative element.
A single logarithmic freq sweep lasts quite long (3s for 128k, 5.9s for 256k ect) for resonance build-up, even with massive resonance systems (the excitation level may be of importance though).
George
I was thinking of a waterfall plot of frequency response over amplitude not time. The old standard was 1 watt now it is 2.83 volts. (One of my contributions!)
The other issue is that the distortion products of a reproduction device are also distorted by the preamplifier such that the third order products become ninth etc. Then add the amplifier's distortion and the loudspeakers And just from second and third order distortion a you will have harmonics out to the twenty-seventh! Yes they are much lower but the energy in music is at the LF end and the ear is maximally sensitive at upper midrange frequencies. Throw in the ability to perceive signal below noise and you end up with detection thresholds approaching -100db!
Wasn´t this Sennheiser ("Kunstkopf") ? Stax (the headphone maker not
the record label) had a demo cd made this way.
For whatever reason this never worked for me no matter which headphones
I tried. I still had "in head" localisation and all I noticed was a strange
coloration to the sound.
Anyway, what I meant was not the recreation of spatial clues but
what PMA said in post #39168
Attachments
Visuals play a large role in auditory comprehension. Try listening to several conversations at a dinner table. If you try to listen to the one you are not looking at you may well not be able to understand a word. Look at the talkers and its pretty easy. I don't think its a simple as some kind of beamforming. There is some serious research on this.
I did note it, and while certainly true, this has been true
for analog recordings too.
30 years ago:
I have a CBS recording (1983, Vinyl) of Mahlers 5th and 6th which proudly
states on the cover "Recorded digitally on the Sony PCM 1610 system".
I would say there is nothing wrong with this record as such but it has
some tonality normally attributed to first generation CD players to it.
I wouldn't be surprised. A lot of recording engineers and producers in that era didn't really know how to use digital recording properly (you can even see clipping on many of the recordings). And a lot of analog mikes and preamps had an edginess to them which worked perfectly for tape, but was anathema to a transparent recording medium. Yes, there were certainly lots of bad recordings, analog AND digital, foisted on the public over the last 30 years.
In experiments from that time which inserted an A-D/D-A process into an analog chain- with levels set correctly- even digital critics were unable to hear the difference.
In experiments from that time which inserted an A-D/D-A process into an analog chain- with levels set correctly- even digital critics were unable to hear the difference.
- Status
- Not open for further replies.