Hi Pavel, the idea is rather difficult for me to express clearly, but I believe that it still is a simple polynomial non-linearity. In case that I fail to get the point through, kindly allow me to explain one last time by the long way.
Please consider a case where source material avocado already has some H2 distortion, let me use the value of 0.3% and negative phase H2 just to explain my line of thought.
Focusing only on H2 distortion :
At the output of the 0.5% negative phase H2 mu follower tube stage, the resulting signal has ~ 0.2% and negative phase H2 due to distortion cancellation of an inverting stage. After phase inversion of the recording (made by by swapping +IN and -IN of the ADC), apricot finally has ~ 0.2% and positive phase H2. I arrived at this idea because similar to Nelson Pass observation about positive phase H2, apricot seems to have a little bit of enhanced detail (sparkle). Logically, it then becomes difficult to hear a difference between a 0.3% THD avocado and a 0.2% THD apricot.
When avocado is inverted before the tube stage, resulting apricot2 will have ~ 0.8% and negative phase H2. Perhaps easier to differentiate due to 0.3% and 0.8% wider THD spread with easier to perceive depth of soundstage difference.
It is just a possible scenario that I am by no means certain of, could be wildly mistaken or irrelevant. But hopefully I am now able to present the idea clear enough.
One thing, indra1, ABX blind testing for soundstage depth/width takes a lot of training & perseverance. See the descriptions here of ABX testing using just those cues
Gearslutz - View Single Post - Foobar 2000 ABX Test - Redbook vs 192/24
That whole thread is interesting
Gearslutz - View Single Post - Foobar 2000 ABX Test - Redbook vs 192/24
That whole thread is interesting
Indrawan, I am not sure about audibility of phase change in H2. I would like to see some evidence, supported by a well described DBT experiment, not only by quoting a German forum where some sighted listening test was performed. I know tons of "evidence" like this that support audibility of myrtle blocks under speaker cables etc. I do not see much reasons why a change in distortion component phase should be really audible.
BTW, it is no problem to me to prepare a test with H2 phase changed of 180°. Before doing that, I would need to know there are enough members willing to participate in such test.
BTW, it is no problem to me to prepare a test with H2 phase changed of 180°. Before doing that, I would need to know there are enough members willing to participate in such test.
Last edited:
Usually the evaluation of validity depends on the abitliy of a test to test what it is intended to do; normal routine is to test the reaction to the presentation of the independent variable, if noone knows what the independent variable really is, it ´s "hard" to judge the validity.
If you define validity in a way that as long as somebody somehow pushes a button so that the software is able to produce a result, then definitely every test is valid, but that isn´t the meaning of "test validity" .
That it wasn´t intended to be "scientific" doesn´t imo help, as the scientific method is just about using the tools in the right way to get correct, meaningful results. So using "it´s not scientific" is a bit like "i don´t really care about it" .
Looking at some of the posts in this thread (and others) it is imo quite obvious that a lot of people don´t realize the restrictions caused by the method choosen.
Last edited:
Pedant no. 2 stated that we were running in circles and i think he´s right in that.
I´d say his version of "whataboutism" (you criticize the "ABX" but what about "sighted listening" ? ) is one of the reasons (to be fair it´s used by others too) .
And i simply don´t understand why he maintains something like " blind is reality" although the contradicting evidence was brought to his attention several times before (in fact just recently) .
Being able to walking back a statement that isn´t correct, should be the basis of a sensible discussion. Falling back to produce a classic strawman instead will surely do nothing against "running in circles" .
I´d say his version of "whataboutism" (you criticize the "ABX" but what about "sighted listening" ? ) is one of the reasons (to be fair it´s used by others too) .
And i simply don´t understand why he maintains something like " blind is reality" although the contradicting evidence was brought to his attention several times before (in fact just recently) .
Being able to walking back a statement that isn´t correct, should be the basis of a sensible discussion. Falling back to produce a classic strawman instead will surely do nothing against "running in circles" .
Hi Pavel, perhaps there is a miss somewhere. No, I never posted any quote from any German forum in this thread. The quote I posted on #584 was from Nelson Pass article on H2 Generator available free on his Firstwatt site.
I never mean for you to prepare another test, I was only trying to convey my thoughts about the fruit test and point out a possible experience that you may find interesting by a simple phase inversion on Foobar during your listening sessions with the hybrid setup. I already have that particular experience and felt no interest for a test. No need for you to do more work.
Last edited:
@indra1, you might be interested in this thesis which examines changes in phase on the perception of timbre in complex sounds?
"The effects of background noise and test subject on the perceived amount of bass in phase-modified harmonic complex tones"
"The effects of background noise and test subject on the perceived amount of bass in phase-modified harmonic complex tones"
Thank you, it is not just equipment design aspect, the understanding of psychoacoustic and audibility of phase related issues is also important. Kindly refer to some that you have found interesting.
As you pointed out, one solution to phase coherence may lie within the digital domain. Perhaps I have an outdated approach, but I am still somehow intrigued to explore available solutions in the analog domain.
Last edited:
People might try this one first. It's a 70Hz sine with 100% H2. Actually, it's a slightly stretched harmonic that is at 140.5Hz, not 140.0Hz. The 0.5Hz offset gives a 2 seconds repetition rate of the phase of the harmonic slowly and continuosly rotating through 360°. So, regardless of the speaker's phase response there will always a section of that loop where the harmonic is in-phase and another where it is exactly out-of-phase.
If we were insensitive to phase shift (actually, to the waveform changes introduced by that) the whole thing would sound excatly the same all the time no matter what the phase relation is, but.... well, listen for yourself...
I haven't listened yet but will be interested in others listening
One psychoacoustic issue that this highlights is that we are more sensitive to the relative, dynamic shifts in the soundfield we are listening to rather than static distortions.
There has been a realization in psychoacoustic research that test signals need to be more realistic & complex rather than the simplistic tones used in the past. There's a good reason for this.
It's also of significance when we think about playback noise - static noise is easy to ignore (hear through), modulating noise seems to have a perceptual effect - a secondary rather than primary effect.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.