Could be, that I completely misunderstand, but me too (at least as a subset).
Two loudspeakers in front could only position an "image" if the receiver processes the sound field accordingly, therefore my example with a single microphone as receiver (at the usual listener position), which would produce exact the same output signal regardless if is (for a human listener) mid-center-right or mid-center-left; assuming for the sake of argument that the setup is perfectly symmetrical.
Nothing wrong with a bit of nitpicking every so often...
I tend to think that my "non sequitur line" is corrobated by your questions cited above.
In your post you described the listener as being able to suceed on average 18/20, which implies that chances of guessing are very low. If you want 'absolute truth' , I'm sorry, but statistical considerations can't provide that.
It's always the same problem, if you've found only one listener who (most likely) can 'hear it' , it doesn't mean that there aren't others as well.
As stated already in the past, large scale listening experiments are very rare, so conclusions to populations are usually not warranted. But, beside the experiments after wwII where sometimes tenthousands of humans participated in large scale tests, we do not have others and so we can not afford the luxury to dismiss the small sample size experimental results.
I know
Their discussions always go like this.........nowhere. Jakob is right of course but it's a purely semantic, pedantic contest between academics....
Nothing wrong with a bit of nitpicking every so often...
I tend to think that my "non sequitur line" is corrobated by your questions cited above.
You need to read some history of Markw4 claims. Then, you may want to take your "non sequitur line" and shove it.
Yes we can build on certain hypothesis, until proven wrong. I'd give up as soon as you'll specify what's the practical relevance of (for example) one or more persons confirmed to hear 0.0001% IMD.
And here we are again, running in circles. We have zero proof that a human can hear 0.0001% distortion. We also have zero proof that such a human that can hear 0.0001% doesn't exist. You say: "since we don't know if such a person exists or not, we cannot draw any general conclusion about the ability to hear 0.0001% distortions", I say "until a person confirmed to hear 0.0001% distortions shows up, we can happily and productively ignore this possibility and it's consequences".
That's in my view the difference between theory and practice. Last time I've checked, this forum is about practice and applied sciences, not about investigating theoretical possibilities (and nitpicking about, which I just did myself).
Last edited:
Have patience, N101N will appear soon for your full entertainment convenience. Don't forget the popcorn.
Hi,
So your theory is if a 0.9% second harmonic distortion audio clip is played on a speaker or IEM with 1% second harmonic distortion, no one can hear it? Are you sure? What if they merge & dissonate into more distortion outside of the 2nd harmonic? Or what if it's possible to hear distortion through distortion sometimes, for example all the people in diyaudio using speakers with let's say 2% THD & then they keep on raving about how good x amplifier sounds which is 0.001% THD versus their other amplifier which is 0.002% THD...... Again the op-amp & DAC developers I cited earlier, National Semiconductor in the U.S. & AKM in Japan, spent time in their listening rooms deciding which chip sounds best & I am pretty sure all their relevant op-amp & DAC chips are under 0.001% THD (especially on their equipment). I am not an expert on THD although it's either.. 1) you can hear distortion through distortion, 1.1) distorion plus distortion dissonates into other FR sectors or some similar effect, 2) They are hearing something else in the chips (like what? Phase variables, settling time, tiny frequency spikes(?) will an op-amp chip ever generate an audible frequency spike like at 7 kHz or 16 kHz ? That's why if I blind tested say LME49720 versus AD828 [or alternatively THS4062 if it's possibly a superior cousin to the 828, it's just I've never heard it in any circuit], then I didn't prove anything narrow, like it could be phase, settling time, a tiny error spike in the FR from 9 Hertz to 19 kHz; so if I purchased the necessary equipment & posted a screenshot that the THD is ultra-low, there are no tiny FR spikes, the settling time is identical, their phase is identical... then I am just proving I can hear something else in the chip which is not measured with the equipment..... however in general the human ear can't hear anything more than frequency response domain, time-domain & purity of the signal, so...... I don't understand why humans can identify for example PCM1704, versus Sabre ES9018 (when they are both ultra-low / low THD, flat FR, I assume equal specs in the time domain), so they have met the areas of the human ear (FR, time & purity of the signal), yet people can identify them easily every day over & over...... (how...) so either humans are super sensitive to the time domain & THD domain -- [with the right equipment, for example the first day I used the Etymotic er-4b everything sounded pretty similar...... (so again I cite this picture File sharing and storage made simple, that there is the information domain (like noise, at first this picture is just noise) then there is the perception domain once we get used to the noise, then we have permanent pre-processing which takes slices of information out of visual noise & slices of information out of audio noise (so like on the first day the Ety ER-4b is 'noisy' then later the pre-processing takes away the noise & just slices out the efficient or useful parts of the audio signal, just like with the noisy picture of that dalmatian)], -- or there is something else for instance the op-amp chip developers at National Semiconductor were hearing which is not time-domain, FR domain, THD domain. Nevertheless after they compared all their 0.00004% THD chip designs back and forth in the listening room then finally released the LME49720 as so on, yeah! Nice style!
On the other hand JRC in Japan only released that stupid Muses01 which sort of sounds on the level of an OPA2134 or LT1115, at least, if it has a special sound then I didn't hear it... I put it in my iBasso amplifiers & quickly became tired & changed back to the normal LME49990 setup, or sometimes AD797 & stuff. (iBasso included AD797 in the kit & I think the manual said to roll the op-amps & decide if you like the 797 or the 2134 or other included chips.. you could also roll the buffers & roll the virtual-ground)
------
Edit - when I say ' junk ', of course all audio chips sound quite ok, even a PCM63 DAC without oversampling, using a circuit with the opposite of which type of resistors are said to sound nice, & with the opposite of which type of capacitors are said to sound nice, etc......, into some average spec op-amp like say, LT1115? (I used it in the iBasso portable amplifiers & the datasheet doesn't say it's special aside from voltage noise), I assume still sounds 'ok'... it's just comparing ok sound to excellent sound.
------
Furthermore, there's some people in this thread such as Abraxalito which say the PCM63 is "superior sound", I don't think that's true like what does the PCM63 have to offer compared to the ES9018 & so on? Unless humans can hear the R2R stuff or the "current stuff" without voltage & how is that possible ? Is it possible in the time-domain? Again aside from the minor conversations, the primary reason I'm posting in this thread is how can people identify an ES9018 versus a PCM1704 ......
Yes, often the cover is the only thing that can be saved in a book, especially if the content has been copied and pasted.
Aside from the black & white dots picture which is evidence of permanent visual pre-processing ( i.e. 'we' are walking perceptive lab equipment ), of course there's examples in audio as well.
-------~/
So for example, the sound files at this link seem to provide evidence of long-term pre-processing in audio, the first time you hear these "6-band noise-vocoded speech" files, it's a lot of "incoherent nonsense", then when you click on the "clear speech" file, then click on the 6-band file again, the 6-band file sounds completely different and is coherent, so then that stays in the pre-processing so if the human listens to the noisy file again 24 hours (or 24 years?) later, it's still equally coherent immediately (without the second file).
In other words, it's evidence that visuals & sound permanently change depending on how we process them (obviously, efficiently, to narrow down what is important then this is simple evidence we have a built in filter which filters away noisy incoherence & narrows down perception & stores it permanently in the pre-processing) & we can never undo what our pre-processing has decided. (Like in that black & white dots picture once your pre-processing decided what it is, it stays there)
I.e.
https://www.mediafire.com/convkey/26c4/4erz9nnkdmvn829zg.jpg first time viewing, nonsense picture, lots of noisy black & white
https://www.mediafire.com/convkey/26c4/4erz9nnkdmvn829zg.jpg 10 minutes later, a clear picture with clear structures
https://www.mediafire.com/convkey/26c4/4erz9nnkdmvn829zg.jpg 10 years later, a clear picture structures (immediately in milliseconds, no thinking required)
---
Audio example
http://www.mrc-cbu.cam.ac.uk/personal/matt.davis/vocode/a1_6.wav first time... nonsense noisy sound, can't hear anything, just black & white noise
http://www.mrc-cbu.cam.ac.uk/personal/matt.davis/vocode/a1.wav similar file which gives sonic information
http://www.mrc-cbu.cam.ac.uk/personal/matt.davis/vocode/a1_6.wav second time... now the sound is different
---
Therefore the first day I used the Etymotic ER-4B then my ES9018 desktop DAC sounded sort of normal-ish, then 1 week later or 2 weeks later the pre-processing has eventually narrowed down the more important sonic characteristics so I can more efficiently hear the ES9018 DAC difference (including the line-out op-amp, etc) & I cite the black & white dots picture as evidence that the human mind sorts information under this basis.
(I.e. prior to the Ety er-4b then I had it usually hooked up to a class-A amp into bookshelf speakers, which emitted an obvious sound, then that obvious sound was very quiet / difficult to hear in the er-4b at first)
(plus other people said the es9018 desktop dac + line-out op-amp -> class-A speaker amp -> bookshelf speakers sounded really great (with anything, normal music, normal movie) so their outsider opinion deletes the placebo effect, theory)
Last edited:
I know, but I would be pleased to visit Canada.
Evgeny Kissin is scheduled at the Roy Thomson Hall.
Hi peufeu =)
I don't have any equipment to check transient response for ringing, frequency response up to several times GBW.
Nevertheless, 0.00005% THD op-amps with rare ringing or rare suspicious spikes, will still retain their transparent & fresh sound signature, don't you think so? Technically speaking
Hi Kastor L,
You do realize that NOS DACs absolutely require a vicious 7th order (or higher order) filter - right?
Why is it that idiots who build or buy NOS DACs do not include the necessary filters? Do you know why there is a nasty filter there???? Mostly to keep the 44.1KHz sampling frequency out of your system! It is a high amplitude signal that can possibly end your tweeter. Mind you, maybe it is useful to run the amp over the crossover zone (if that amplifier has problems there).
Anyway, running a non-oversampling DAC without the filter is a game only someone completely clueless to how this stuff works will do. It is a very stupid thing to do.
-Chris
You do realize that NOS DACs absolutely require a vicious 7th order (or higher order) filter - right?
Why is it that idiots who build or buy NOS DACs do not include the necessary filters? Do you know why there is a nasty filter there???? Mostly to keep the 44.1KHz sampling frequency out of your system! It is a high amplitude signal that can possibly end your tweeter. Mind you, maybe it is useful to run the amp over the crossover zone (if that amplifier has problems there).
Anyway, running a non-oversampling DAC without the filter is a game only someone completely clueless to how this stuff works will do. It is a very stupid thing to do.
-Chris
- Status
- Not open for further replies.