I won't speak for Pavel, but my point of describing relative to dBFS was to accommodate instantaneous vs average program level. Let us take the example of a DAC: if one decided to have a terrible gain structure that has him/her digitally attenuating as to push the signal into the noise floor of the DAC, then, yes, there's plenty of room to say that the DAC might have an audible effect. Lower the gain of subsequent stage(s) to something more modest and the DAC audibility may disappear.
Yes, we don't know if the constant noise heard in quiet passages were actually 'masked' by the music - it may well be that the noise was now a appearing as a dynamic noise that was still there but intermingled with the music signal & difficult to tease out for measurement purposes.
Yet, it could be that our perceptions of the music signal is being changed by this dynamic noise & only when it's removed at source that we identify the perceptual difference?
There seems to be a confusion between what is consciously audible & can be pointed to Vs what is actually effecting our auditory perception & not necessarily identified as an individual aspect of the sound. Again, auditory perception s what it's about.
Last edited:
I mean below instantaneous level and I think, according to my own experiments, that -60 dB is with great reserve and that even -40 dB would be very very difficult to tell, speaking about music. On pure sine wave of certain frequency and level, distortion of order 0.1% (H2) starts to be audible. With music, the level is much higher.
To put some literature as well -
Measurement of Harmonic Distortion Audibility Using a Simplified Psychoacoustic Model
AES E-Library >> Measurement of Harmonic Distortion Audibility Using a Simplified Psychoacoustic Model
Measurement of Harmonic Distortion Audibility Using A Simplified Psychoacoustic Model – Updated – Listen, Inc.
Just audible thresholds for harmonic distortion. | Researcher Information | J-GLOBAL
The Ear’s Own Nonlinear Distortion
The Ear’s Own Nonlinear Distortion | SpringerLink
Earl Geddes said we understand pretty well what about 95% of the population can hear from prior research. He said the other 5% has never been studied, and to do so would be costly. He said a preliminary study might cost a few tens of thousands of dollars. So far, nobody is willing to pay for it.
Seeing as you guys are convinced by Foobar ABX test results, here are a number of posted ABX results (by Ultmusicsnob) showing the audibility of <60dB in music
These are pre-prepared audio files with known levels of jitter & measurements.
Jitter Correlation to Audibility | Page 12 | Headphone Reviews and Discussion - Head-Fi.org
He states what he perceives in each of his ABX listening - in the case above "Getting into a pretty subtle area here. The earlier files seemed to be close together in terms of the audible effect of jitter, but path30jr is more challenging. Still using the same cue, quality of the piano attack in soft passages, but there's less of it here."
He has more ABX results posted in that thread prior to this one but this set of results I believe show he audibly perceives far lower dB levels of distortion than your 60dB - the subsequent comments & analysis are worth reading
@Markw4 - he may be on of the 5% group?
But as we all know, these "hundreds of tests" were not done on the same effect nor by the same people under the same conditions. Which means that you are in fact summing up all these different experimental results to conclude that indistinguishability was given for every single effect under test.
That approach surely doesn´t reflect the spirit of statistics and further it defies logic.
Invoking the spirit of statistics and taken the "hundreds of tests" at face value, then we could - under the assumption that the null hypothesis is true - ask for the expected number of positive test results given by the random guessing.
Using the usual SL = 0.05 and considering that, due to the discrete nature of small sample sizes in each test, the factual SL is probably more like 0.038 (16 trials with 12 hits needed for success) then the expected frequency for positive test results is 3-4 per hundred of tests.
So ask yourself how it could be that "hundreds of tests" all came out (allegedly) with negative results, you even could calculate the propability for such an outcome (hint it is "not very" likely) .....
Do I really have to do the search? My petard is killing me, atm.
I'll solder on & work through the pain, anyway.
Forgive if I'm wrong in my impression about this but I seem to remember many on a different thread seemed to defend Foobar ABX results but from this thread alone - a post from 2 day ago - could be still in the memory of people reading this thread, no? See you made me do this despite the pain of my hoisted petard when there was no need - I think you have a sadistic streak
John Curl's Blowtorch preamplifier part III
Last edited:
Come on billshurv, be fair. It might not be outspoken literally but in fact it surely must be the underlying premise, or maybe not the premise that it was "scientific" but that nevertheless the negative results allow the conclusion to be the "TRUTH" , namely that no perceptable difference exists.
Otherwise someone like DF96 would (implicitely) stating " i know it wasn´t up on par with scientific standards, i know it was only fun, i know that it could quite likely have been all wrong and i know that no further conclusions are justified but i declare the indistinguishability of all questionable effects" .
I hope it helps to realize at which point it gets difficult with the "it´s all fun" approach......
And this from you yesterday (how's your memory, billshurv?) - maybe you really are a forgetful sadist?
John Curl's Blowtorch preamplifier part III
I believe billshurv & many others are from the House of Slithering (Harry Potter reference) - those who try to say (but use politician's believable deniability) that Foobar ABX is a valid 'test' & they trust the results & when challenged with this begin the deconstruction of what they said ("you put words in my mouth", blah, blah)
Slithering is a description of the movement of a snake - you can never tie down a snake - they never take a final position - always slithering between positions
I thought you had more sense than to defend a troll who has a history of twisting people's words
I didn't read into Jakob2's words that they were intended in any way to defend a troll.
I thought he was on your ignore list
Jakob2 was never on my ignore list. For people who are on the list, I think you know their posts can still be read if one chooses to.
- Status
- Not open for further replies.