I've wondered myself, when you consider the lack of harmony within the equal temperament scale, how "euphonious" 2nd harmonic distortion is in equipment. It's probably ok with simple small ensembles, but when it comes to full orchestra I have my doubts.
Assuming that we are talking about multidimensional evaluations using music as stimuli. Is it true even for trained listeners? It looks like a reasonable hypothesis (the "..... their amount and effect is much less" part ) but is there experimental evidence to back it?
Given the example about comparisons of test protocols I've cited, it seems to be fairly safe to conclude that quite innocous looking details can have a quite severe impact.
Another example would be the case of the very high false response rate in "blind" tests when two identical stimuli were consecutively presented and the participants were asked to express their preference. Up to 80% false responses are normal rates; It is obviously not an easy task to examine this hypothesis, as one has to know about the correct response that listener have to give, and has to "blind" the sighted tests. One approach would be different cases for the DUTs but different or same electronics inside, the other usage of qualitative methods to explore the sighted results. Which means the participants are not blinded to the devices but to the perceivable differences.
+ 1000
It's unfortunately an anecdotal description (the company involved did not publish much about these experiment, did it?) and should be taken with a grain of salt. Just as an example, the same member posted with dead certainty that ABX tests were not used in codec research/evaluation, but someday jj posted "sure we used it all the time" and the usual "game of silence" followed; if ego issues seem to get involved it is often impossible for third parties to find out which informations are correct.
For your convience I quote your post:
Please post the link to the message(s) in which you posted the title of the thread and some of my replies on Hydrogenaudio. 🙂
That is the information that I was asking for and your answer - quoted at the beginning - somehow did not contain this information, surely just an accidential slip......
Jacob2 Of course, a test has to be tailored to suit its purpose, and of course, badly designed tests exist.......tests that make differences disappear. Did you teach me that?
Depends on the heatsink/amp.....some are essentially stone dead, typical consumer gear when the fins are tapped/struck ring in midrange frequencies ring quite loudly and for long time.
You need to get out more PMA, out into the real world.
Typically heatsink should be grounded to quench parasitics. The amp I built had collectors commoned to output line so I lapped flatpack transistors to the heatsink and omitted insulating washers. Crown MA series amps also have transistors directly mounted to heatsinks insulated from chassis.
Dan.
Says the man who lives about as far away from civilisation as it's possible to get 🙄
johnego,
As I understand it, ‘even temperament’ is the dividing an octave into a scale of twelve notes of equal length, which is the basis of western orchestral music where all the instruments are tuned to A3 at 440hz.
This means that the Gamelan, which is microtonal tuned to any one of a number of complex scales of notes within an octave, will to the western ear sound completely out of tune. It is not by the way, and neither is Oriental or Middle Eastern music.
When I heard the Gamelan orchestra that I spoke about, at first it sounded way out of tune. By the end of the evening, having gotten used to the sound, it all sounded beautifully in tune! Needless to say, it completely blew my mind. 🙂
I hope this helps.
ToS
Me? Show me a measurement of microphony in a non-idiotic amplifier design, that might be audible. The difference between us is that you permanently create stupid pseudotheories, without bringing any proof, any measurement, though I am supporting my occasional claims with measurements. To me, you are completely untrustworthy.
Never to me, ever, my introduction was viewing Lord Jim decades ago immediately thought it was the most heavenly sound. The differences in speakers East coast/West coast USA, UK, Japan, planar are so dramatic in of themselves that I don't think this argument has anywhere to go.
Max of course knows everything is in tune to me especially dropped A. 😀
Last edited:
Thanks; maybe I do now understand, you didn't want to imply that it could never happen (i.e. that differences disappear) and meant that the other believe the difference would _always_ (automagically) disappear, no matter if the test is well designed and executed. Right? 🙂
Hey man, don’t get pissy with me; save it for our euro brothers or at the very least elac!
It’s the scale they decided to use to control output levels....I don’t know the conversion as it seems rather complicated, but I can tell you it definately affects output levels......not quite 1:1 , 4 lufs is not 4db. It’s less, more like 1db but don’t know for sure maybe pavel does?
Scott,
Perhaps I did not express myself as clearly as I could of, and so I will try again.
At the beginning of the performance, I thought that the orchestra was out of tune. At the end of the performance I realised that it was - me - not the orchestra, that who was out of tune before the performance had even begun. Ever since then, I have loved Gamelan, African Indian, Middle Eastern, and Oriental musics. Indeed all musics.
My thoughts about unequal but even temperament tuning in regards to loudspeaker design, are just that - thoughts. Even I know when I am not entirely sure what I am talking about, but you Scott, are one of the smartest people out there, and I always listen to you in the same way you listen to me. The difference being you are more often likely to make sense than I ever do.
The day is good 🙂 ToS
Last edited:
As I have already mentioned twice, post #21590, I asked you a question and provided a link. No answer to my question and no reading of the link. Sure looks like an intentional slip.....
Nothing negative intended, I wouldn't notice if something was in or out of tune I only listen to what someone wants to present.
"Everything is music only the listening stops" John Cage IIRC.
Yes, what I meant to say is that all sighted comparative tests are invalid. Especially audio developers should not rely on their own ears in sighted tests.
If an auditory difference exists, a blind test can be designed to bring it out. It is equally possible to design a blind test that will not bring an auditory difference out. One should aim for the former and avoid the latter.
I went in '83, it was £12 to get in, didn't pay though, just walked past the guy selling tickets 😀 Went Stonehenge Festival that year too and '84
As you were asking before about the "teaching", I've to ask now: Did I teach you that? (Just kidding)
In fact, one can't say that "all sighted comparative tests are invalid" ,because how could we know about it?
As I've said before (scottjoplin will love the repetition 🙂 ) the problem with "sighted listening tests" is, that you can't show the (internal) validity, so in every scientific approach you'll avoid it (by introducing the "blind" property).
But being unable to show (internal) validity is something different to state that a listening test was not valid (i.e. was/is invalid).
To state "invalidity" would require a quite elaborate experiment; what if one prefers always the same DUT in "sighted" and "blind" controlled listening tests?
Usually audio developers do rely on measurements, their own ears and ears from other people, but admitted that more extreme point of views are out there.
The problem is what I've tried to address before, we can't read listener's mind and up to now (that might be different in the near or not so near future) can't access the physiological response to any stimulus.
Doing subjective perceptual evaluation (especially in case of multidimensional evaluations) means to know how to listen and means being able to control bias impacts (surely not perfect but up to a certain degree).
If it could not be learned to control bias impacts (in the sense mentioned above) "blind" tests in most cases would not work either, as so many bias effects exist that are still at work under "blind" conditions.
As it is usually a matter of costs/efficiency "blind" controlled listening tests are just an additional part of the toolbox.
But I totally agree that one should be able to learn to get the same result under "blind" controlled listening test conditions as under "sighted" controlled listening test conditions. And be it only just as a sanity check. It will not help in forum discussions, though.
Imo, most people don't want to hear about positive test results contradicting their prior belief of inaudibility. 😉
Nah, you asserted that you've provided evidence for your claim from 9th of June, but I hope we can agree that there is no link in your post #21590 (not in #21519 either) .
In your post #21519 you posted these quotes from the hydrogenaud.io forum:
You failed to provide evidence for your claim from 9th of June.
Last edited:
Oh, open up just about any typical consumer Yamaha or Sony or Pioneer etc amp and you will find ringing heatsinks and microphonic gain stages. I have typically observed such heatsinks ringing because of direct mechanical stimulation. I have observed such heatsinks being excited by OPS currents also, just connect a dummy load and listen to the heatsinks ring in sympathy to the music, it sounds a bit like stylus direct acoustic sound but has ringing decay time.
Pavel, I haven't provided any 'pseudotheories' but I have stated interesting observations and I have provided loopback recording proofs of my observations but apparently that is not good enough. I do not dismiss your efforts in achieving low measured distortions but this is not the last word in subjectively good sound. Materials used in the system are important also, and are ultimately subjectively mission critical and still dominant despite how ever low the measurement numbers can go. I am saying that system sound can be controlled and corrected by application of chosen materials in the right places and that this technique is useful in any quality level of system including systems like your measured high performance system. Dan.
Last edited:
- Status
- Not open for further replies.