Stereophilus Posted October 26, 2020 Share Posted October 26, 2020 37 minutes ago, davewantsmoore said: A lot of the level errors parts is contined with research on speech intelligibility and audibility. Shape/slope.... is spread wider. There's studies on timing distortion of single channel signals.... studies on actual magnitude (ie. how much non-linear distortion is audible) .... and then studies which look at "imaging" (ie. distorting one channel), which applies to level and shape. Almost all imaging studies I've seen are to do with level ...... although I know of a lot of people who propose that non-linear distortion that is correlated in one channel is very important too. I've seen no formal studies of it, except for annecdotal evidence, and logic around why you might expect it to work like that. If you're looking for papers with "the complete answer", and/or concise numbers/guidelines, etc..... then you won't find this. What is audible depends on a lot of complexities like the content (masking), and levels. I don't keep a list around, as I only spend time on things I think I should be looking into..... not things I've decided to leave alone. Then it is not unfair to say that our knowledge in the area of audibility is quite incomplete. As you say, there are many variables, and a complete synthesis of this seems unlikely, especially when individual (person to person) variations are factored in. I did pull you up on this, not because I expected you to have the studies at your finger tips (although sometimes you do surprise me, Dave), but because we need to be careful about our assumed knowledge... As I iterated many pages back in this very thread. Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 26, 2020 Share Posted October 26, 2020 38 minutes ago, allthumbs said: It would be interesting to know how hearing aid manufacturers or the likes of Cochlea approach those problems? 19 minutes ago, Stereophilus said: Edit: For the sake of being obvious, these things have no meaningful relationship to our discussion about audibility in the context of this thread. Well, to be fair to @allthumbs, you can be hearing impaired but it doesn't mean you are necessarily neurologically so. The thresholds of audibility don't just include pressure levels you can hear - the bit your ears handle - they extend into how you process those signals. Can we a small difference in the time alignment or shape of a signal? Assuming your ears can pick it up and give your brain the relevant nervous sytem signals, how your brain processes those then takes over. Doesn't really matter whether whether the initial pressure phenomena came from the ambient or was amplified by a device. Makes for interesting possibilties though, no? A member of this house has a very different response curve (tinnitus) and can hear certain changes in the audio system with bat-like acuity. What your brain can do, compensate for and be trained to do is phenomenal. Gives credence to earlier suggestions by @frednork on the value of trained listeners. Link to comment Share on other sites More sharing options...
davewantsmoore Posted October 27, 2020 Share Posted October 27, 2020 19 hours ago, rmpfyf said: Oh? Thank you for proving my point Link to comment Share on other sites More sharing options...
davewantsmoore Posted October 27, 2020 Share Posted October 27, 2020 18 hours ago, Stereophilus said: Then it is not unfair to say that our knowledge in the area of audibility is quite incomplete. I'm not sure what point you would be trying to make by saying this. Do more experiments? Sure. 18 hours ago, Stereophilus said: I did pull you up on this, not because I expected you to have the studies at your finger tips (although sometimes you do surprise me, Dave), but because we need to be careful about our assumed knowledge Yes.... You need to assume almost nothing (in the experiment design, or analysis) Notice, that I'm making no assumptions, other than "this is what I suspect you will find". Link to comment Share on other sites More sharing options...
davewantsmoore Posted October 27, 2020 Share Posted October 27, 2020 19 hours ago, Stereophilus said: We do have a lot of experimental data on audibility. A lot of it exists at a very basic level. Not a lot of data exists about audibility of complex sound waveforms with constant variability of amplitude, frequency and harmonics. This crosses over with the debate about the subjectivity of hearing and double blind tests and test subject control. "Understood" is, in my view, a very large assumption based on inadequate data. I don't think at this stage audibility of somthing, or not, is very relevant. The challenge is to quantify the differences. Then once you know what the differences are (if there are any) then have a listening test to see if they're audible..... and also you can compare those differences to other sutdies/data to see if other people found differences like that audible. Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 47 minutes ago, davewantsmoore said: Dave, don't suffer last-word-itis. If you start from a perspective of stating 'but audibility is well-defined', it's very easy to point out where it's not. There's little point saying something 'is' because it 'mostly is'... particularly in a study of corner cases/testing at or near limits, which this may well be. There's the possibility that there are audible phenomena not easily measurable... as I wrote earlier, just admit it and exhale. There certainly very many challenges in audio that are aptly characterised by a variety of very accessible methods. This may be one of them. It may not. No need to get twisted. Link to comment Share on other sites More sharing options...
davewantsmoore Posted October 27, 2020 Share Posted October 27, 2020 43 minutes ago, rmpfyf said: Dave, don't suffer last-word-itis. I should have been more verbose. Perhaps I proved it.... but I don't understand what your point was. 43 minutes ago, rmpfyf said: There's the possibility that there are audible phenomena not easily measurable... as I wrote earlier, just admit it and exhale. "Easily" is very subjective here. I'm not wholy sure what you want me to admit. Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 12 minutes ago, davewantsmoore said: I'm not wholy sure what you want me to admit. Principally that you need the last word I ultimately don't think we disagree. Link to comment Share on other sites More sharing options...
Telecine Posted October 27, 2020 Share Posted October 27, 2020 Has anyone actually measured anything yet or at least devised a measurement method? 2 Link to comment Share on other sites More sharing options...
Ittaku Posted October 27, 2020 Share Posted October 27, 2020 (edited) 1 minute ago, Telecine said: Has anyone actually measured anything yet or at least devised a measurement method? I did some measurements, found nothing of note, so we learned either my method isn't sensitive enough to pick up differences, or there aren't any differences. Edited October 27, 2020 by Ittaku 1 Link to comment Share on other sites More sharing options...
Eggmeister Posted October 27, 2020 Share Posted October 27, 2020 1 minute ago, Ittaku said: I did some measurements, found nothing of note. What did you measure ? Link to comment Share on other sites More sharing options...
Ittaku Posted October 27, 2020 Share Posted October 27, 2020 (edited) 7 minutes ago, Eggmeister said: What did you measure ? Standardised one minute of music -> Dac -> 24 bit 192kHz ADC measurements, FFT sweep. The variation between playback on the same DAC was +/- 0.02dB across all frequencies. Changing the power cord from a cheap kettle cable to a shielded "premium" audio quality cable on the DAC made no difference to the variation. Edited October 27, 2020 by Ittaku 5 Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 35 minutes ago, Telecine said: Has anyone actually measured anything yet or at least devised a measurement method? Test method proposed, awaiting some test equipment. Not using spectral methods - don't think they'll reveal much. Link to comment Share on other sites More sharing options...
aussievintage Posted October 27, 2020 Share Posted October 27, 2020 1 hour ago, rmpfyf said: Test method proposed, awaiting some test equipment. Is there a summary to catch us all up? Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 Just now, aussievintage said: Is there a summary to catch us all up? The important bits: FFTs at conventional lengths won't show it unless something is grossly out Time-dependent deltas probably the best way Audibility limits are not defined by SPL sensitivity If there is a difference we may not capture it Would be good to have some equipment to characterise some key externalities Anticipated that different equipment will react differently Would love to measure cables directly but expensive to do Link to comment Share on other sites More sharing options...
aussievintage Posted October 27, 2020 Share Posted October 27, 2020 Looking good I must say. Just some thoughts follow... 1 hour ago, rmpfyf said: FFTs at conventional lengths won't show it unless something is grossly out Still worth doing, as some claims indicate big differences are expected. Also, just worth doing once you have the data. 1 hour ago, rmpfyf said: Time-dependent deltas probably the best way What do you intend to do with the resulting data beyond just noting it exists, or not. Spectral analysis? What level is significant? 1 hour ago, rmpfyf said: Audibility limits are not defined by SPL sensitivity Not sure what this means in terms of the actual testing. Just trying to make it speaker sensitivity neutral? 1 hour ago, rmpfyf said: If there is a difference we may not capture it That's a bit self-defeatist. Isn't a valid result = no measured difference ? The above seems to invalidate it. 1 hour ago, rmpfyf said: Would be good to have some equipment to characterise some key externalities Better to control them, but if not, yes. 1 hour ago, rmpfyf said: Anticipated that different equipment will react differently A given by most I would think. 1 hour ago, rmpfyf said: Would love to measure cables directly but expensive to do To me that's a separate exercise. Keep it that way. Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 4 minutes ago, aussievintage said: Still worth doing, as some claims indicate big differences are expected. Also, just worth doing once you have the data. I think we'd need freq binning at a freq bin order of magnitude below sampling at least to start revealing decent details, this needs a ton of averaging to make good, so we're talking very long samples with potentially wandering externalities (power). Not impossible just difficult to realise practically if external variances start defining the effective noise floor. 7 minutes ago, aussievintage said: What do you intend to do with the resulting data beyond just noting it exists, or not. Spectral analysis? What level is significant? Time align it, compute amplitude deltas. That's it. No spectra - see comments above. The study referenced in the first post has amplitude differences cable-to-cable of 7LSB which I'll call 'significant' on account of never having seen a scientifically rigorous account of an audiophile cable. If there's a reliable 7 LSB difference between cables then everyone ever doubting the efficacy of these things can eat crow. If. 10 minutes ago, aussievintage said: Not sure what this means in terms of the actual testing. Just trying to make it speaker sensitivity neutral? Just that the test methods have practical limitations. 11 minutes ago, aussievintage said: That's a bit self-defeatist. Isn't a valid result = no measured difference ? The above seems to invalidate it. There's broad agreement between the two anoraks that Won't Give Up on this thread (@davewantsmoore and I) that no measured difference would only prove that there was no difference as evaluated by the limitations of the method employed. The resounding silence is testament to either (a) everyone else agreeing or (b, more likely) everyone else has lives and tuned out some time ago. 13 minutes ago, aussievintage said: Better to control them, but if not, yes. Indeed but practically hard to do. Hoping to get a hold of a power meter in the coming month to at least reject tests with silly DC, crazy AC voltages, mains transients, etc. 14 minutes ago, aussievintage said: A given by most I would think. Si. 14 minutes ago, aussievintage said: To me that's a separate exercise. Keep it that way. Also a separate budget Link to comment Share on other sites More sharing options...
Ittaku Posted October 27, 2020 Share Posted October 27, 2020 1 minute ago, rmpfyf said: There's broad agreement between the two anoraks that Won't Give Up on this thread (@davewantsmoore and I) that no measured difference would only prove that there was no difference as evaluated by the limitations of the method employed. The resounding silence is testament to either (a) everyone else agreeing or (b, more likely) everyone else has lives and tuned out some time ago. Nah, most of us don't want to get into a bunfight again. The power cable lovers won't change their mind even if every measurement under the sun revealed no changes, and vice versa, so there's no one to convince. 4 Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 Just now, Ittaku said: Nah, most of us don't want to get into a bunfight again. The power cable lovers won't change their mind even if every measurement under the sun revealed no changes, and vice versa, so there's no one to convince. And there you go needing to Get Serious all over again Con. It's enough that cable people don't start with physics obviously on their side. Killjoy. Link to comment Share on other sites More sharing options...
Ittaku Posted October 27, 2020 Share Posted October 27, 2020 Just now, rmpfyf said: And there you go needing to Get Serious all over again Con. It's enough that cable people don't start with physics obviously on their side. I was just mansplaining... 1 Link to comment Share on other sites More sharing options...
rantan Posted October 27, 2020 Share Posted October 27, 2020 7 minutes ago, rmpfyf said: or (b, more likely) everyone else has lives and tuned out some time ago. This. 2 Link to comment Share on other sites More sharing options...
aussievintage Posted October 27, 2020 Share Posted October 27, 2020 1 hour ago, Ittaku said: Nah, most of us don't want to get into a bunfight again. The power cable lovers won't change their mind even if every measurement under the sun revealed no changes, and vice versa, so there's no one to convince. Oh so true Link to comment Share on other sites More sharing options...
Ittaku Posted October 27, 2020 Share Posted October 27, 2020 By the way, it did occur to me that you have a master plan - to try and find measurements that show a change to prove to the non-believers. There can be no outcome from this testing that affects the believers. Link to comment Share on other sites More sharing options...
Guest rmpfyf Posted October 27, 2020 Share Posted October 27, 2020 4 minutes ago, Ittaku said: By the way, it did occur to me that you have a master plan - to try and find measurements that show a change to prove to the non-believers. There can be no outcome from this testing that affects the believers. Not necessarily. I'm kind of hoping to find what I heard, which was not good. I'm not sure there's a scope for 'better'. I'm open-minded though. I'm not out to talk people out of brand or component religions. Also, I need a decent power instrument at home for some other work we're playing with and this is a nice way of finding one and giving it a bit of a run. Link to comment Share on other sites More sharing options...
MLXXX Posted October 27, 2020 Share Posted October 27, 2020 6 minutes ago, rmpfyf said: Not necessarily. I'm kind of hoping to find what I heard, which was not good. For heaven's sake, rmpfyf, simply record the output of your killer DAC with whatever ADC you have on hand, if that is truly your goal. For such an obvious audible difference in sound, you should have no trouble finding a difference in a recording even using a 16 bit 48kHz ADC. And record just a single channel if you don't have a stereo ADC handy. Then upload file A [using original power cord] and file B [other power cord] for us all to hear! This should not involve rocket science to achieve, if the difference is as severe as you have reported. In fact I am a little surprised you have apparently not already made such recordings! Link to comment Share on other sites More sharing options...
Recommended Posts