[gregorio]

Again, what is the only logical conclusion?

That you are indeed a troll and your last post appears to confirm that conclusion.

The issue of an audio reproduction system is to reproduce audio. The issue of your perception is an issue for those who created that audio signal and your personal perception, it is NOT an issue of the audio reproduction system whose ONLY job is to reproduce that audio signal! If you're going to keep repeating a fallacy, then I'm going to keep repeating the facts!

G

 

[Phronesis]

That you are indeed a troll and your last post appears to confirm that conclusion.

The issue of an audio reproduction system is to reproduce audio. The issue of your perception is an issue for those who created that audio signal and your personal perception, it is NOT an issue of the audio reproduction system whose ONLY job is to reproduce that audio signal! If you're going to keep repeating a fallacy, then I'm going to keep repeating the facts!

G

I won't respond in kind this time. Crying troll because you can't deal with someone disagreeing with you is childish.

 

[gregorio]

Crying troll because you can't deal with someone disagreeing with you is childish.

It's got nothing to do with you disagreeing with me, it's got everything to do with you repeatedly posting a fallacy which is contrary to the facts. I made that perfectly clear and your response is to misrepresent that fact and add an insult, which just further confirms that the only logical conclusion for your responses is trolling!

G

 

[KeithEmo]

ABOUT PERCEPTION.....

As often seems to be the case, there are two DIFFERENT arguments going on here - masquerading as one.

1) PERCEPTION of reproduced music. The goal of audio reproduction is to reproduce the audio signal... and how that signal is perceived isn't really part of that discussion. If I play the exact same signal through two copies of the exact same amplifier, and people all seem to like the red one better, then that is simply an observation on human perception. You can debate about how "the perception of the total experience" is different between them, but that really is a discussion about human perception, and not about the equipment involved. (Of course, it may still suggest that painting your amplifier red is an excellent idea.)

2) PERCEPTION as a factor when performing tests. However, when you are performing tests, the human listener becomes part of the test equipment, so there is no way to separate THEIR PERCEPTION from the results. There is some test hardware whose accuracy or sensitivity actually changes after it warms up... and it is quite common for electronic gear to be adjusted after a half hour warm-up. And, if you use that equipment before it is fully warmed up, then your results will be incorrect or not fully accurate. The problem with humans as test equipment is that we are subject to a whole slew of complicating factors, not all of which are well known, and many of which vary from person to person. This now becomes an issue of TEST EQUIPMENT CALIBRATION AND PERFORMANCE.

For example, human perception apparently varies depending on our state of alertness and awakeness. This means that, if you get up at 8 AM, the threshold at which you can notice THD may actually be measurably different at 8:30 AM than at 5 PM (because you are more alert at 8:30 AM). Likewise, our brains adjust to stimulus around us, and become more of less sensitive to certain things.

Start with the parameter that some folks seem to keep insisting is "dead simple" - the range of frequencies over which we can hear. Let's test the range of frequencies you can hear, with a continuous sine wave, in one ear at a time, at 85 dB SPL. HOWEVER, let's test it once 30 minutes after you wake up on Sunday morning, on a nice quiet day. NOW, go out and mow the lawn and trim the sidewalk with that noisy gas-powered weed whacker, and take the test again. My guess is the results will be different. Of course, most of us realize that our ears adjust to loud noises, and so actually mechanically perform differently for several hours after being subjected to high sound levels.

But, if you actually read any work on "brain science", you'll also know that what we notice and perceive can also be affected by the state of our brain. Therefore, we cannot rule out the possibility that, for example, our hearing thresholds actually are different when we're "focused" than when we're "distracted".... or when we're "tired" than when we're "alert". So, yes, it is actually possible that the measurable threshold of some seemingly simple metric like "how much THD is audible" may be different at different times of day, or even depending on what's going on outside the window at the test facility. Perhaps we are more relaxed, and so more perceptive, and more able to detect lower levels of THD, when there are tress outside the window instead of city traffic. And perhaps it's the exact opposite. Even worse, it may affect different people differently, or affect the same person differently at different times.

And, to make matters even more complicated, we humans vary physically over time. Caffeine is a drug which stimulates our central nervous system... and produces all sorts of easily measurable changes in our performance. Do you really believe it's reasonable to assume that the range of frequencies you can hear at 85 dB SPL is exactly the same after one cup of coffee and after three cups? After that third cup there will be easily measurable differences in things like your reaction time, and your ability to complete simple dexterity tests; so why would you assume there is no effect on your hearing acuity? Likewise, there are easily measurable differences in our performance when we are tired and when we are alert, and even when we are happy or depressed, or when we are focused or distracted. Since the human being is the test equipment here, it seems foolish to ignore factors that affect the accuracy and sensitivity of that test equipment. (And, yes, since the criteria is "what's audible to a human being", by definition the human being is the test equipment.)

If you want to determine some sort of "minimum threshold", with any degree of absolute accuracy, then at the very least you need to perform the test under optimum conditions.
And, yes, that means you'd also better document things like time of day, room temperature, lighting conditions, and other environmental factors in your test results.
And, yes, I guess that does mean that a lot of seemingly "simple threshold measurements" may not be so simple after all.
And, while you may choose to exclude "human perception" as a factor, the best you can possibly do is to control as well as you can against variations.

@gregorio Why do you keep repeating the same fallacy over and over again and your response is to ignore the question and just repeat the same fallacy yet again! How can that be interpreted as anything other than trolling?

The issue of audio reproduction systems involves both perception and reproducing an audio signal!! Why is this not self-evident to you? Why, even after it's been explained numerous times, is it still not self-evident to you? Again, what is the only logical conclusion?

 

[Phronesis]

I don't know why it's hard for some people to understand. Whether a difference is audible and whether an audible difference matters depends on the listener. Listeners aren't all the same, and each listener isn't the same at one time versus another. And audibility of differences isn't simply a matter thresholds, since we're talking about music, not just pure tones (reference to simple thresholds won't be helpful in the Yanny/Laurel example).

 

[gregorio]

As often seems to be the case, there are two DIFFERENT arguments going on here - masquerading as one.
1) PERCEPTION of reproduced music. The goal of audio reproduction is to reproduce the audio signal... and how that signal is perceived isn't really part of that discussion. If I play the exact same signal through two copies of the exact same amplifier, and people all seem to like the red one better, then that is simply an observation on human perception. You can debate about how "the perception of the total experience" is different between them, but that really is a discussion about human perception, and not about the equipment involved. (Of course, it may still suggest that painting your amplifier red is an excellent idea.)
2) PERCEPTION as a factor when performing tests. However, when you are performing tests, the human listener becomes part of the test equipment, so there is no way to separate THEIR PERCEPTION from the results.

1) Yes, I agree entirely.

2) This point I don't really agree with. What we actually hear and what we perceive are two different things. Perception is the identification, organisation and then interpretation by the brain of what we hear but as we know (and hopefully agree), this "interpretation" is a combination of all sorts of factors (sight, knowledge, past experience, biases, etc) and may have a lot, some, little or nothing at all in common with what our ears are actually hearing, which explains the difference between perception and hearing and therefore why they are two different things. When we perform hearing threshold tests, the goal is to discover the hearing threshold, NOT the perception threshold. This is why we need "controlled" tests, because "controlled" effectively means; to eliminate perception, thereby leaving us with only "hearing". In addition to the biases, sight and other factors which differentiates perception from hearing in the first place, there is another class of other factors which can affect our hearing thresholds/ability itself; fatigue (or it's opposite), temporary threshold shifts and conditions such as having a cold or hearing damage/deterioration for example. Regardless of this other class of factors, the goal is for perception NOT to be a factor when performing hearing threshold tests.

As @Phronesis was talking about audio reproduction systems then we are both agreed, perception is NOT part of the equation. But also, even when discussing hearing threshold tests, I believe that perception is still not part of the equation. In fact, the only time when perception should be part of the equation is when actually testing perception (or some aspect of it).

G

 

[bigshot]

He isn't a troll. It's a combination of being in love with the sound of his own words, the inability to really respond to anyone else's challenges to his arguments, and a desperate attempt to find a way to validate his audiophile solipsism. I think he actually believes what he says. He's just built a wall of rhetoric and logical fallacies around his misconceptions to defend them. There's really no reason to argue with him. He'll just go right back to his hippy dippy "we don't actually hear sounds- we perceive ideas about sound" argument again and it will all go in circles. I chalk him up as someone who is well spoken, but devoid of analytical thinking processes. We've seen the type here before. They are usually the loudest and most oblivious ones.

The mistake is his refusal to address audio fidelity. Is sound accurate and properly conveyed? Perception is great, but a DAC or amp or headphones can't change perception as much as a comfy chair and a glass of wine can. Taken to the next level, we wouldn't be discussing audio equipment at all any more, because "they just produce sound, they don't affect our perception". It's solipsism. And solipsism is an entertaining way to create circular arguments in Introduction to Philosophy 101. Mental monkey spanking is a time honored tradition in first year college students.

The thrust of his argument goes back to the Greek philosopher Gorgias who said...

1) Nothing exists.

2) Even if something exists, we can't know anything about it.

3) Even if we can know something about it, we are unable to communicate it with others.

Why do people talk about Socrates and Plato all the time, and you never hear about Gorgias? Well, that's simple. It's because his whole philosophy is based on an argumentative trick. The purpose of the idea is to shut down conversation. People use arguments like this when they don't have a real argument. They lob a solipsist bomb into the midst of discussion and let it explode and make everyone shut up, or at least divert them off into discussing purely solipsist matters.

 

[Phronesis]

If you're going to make a distinction between hearing and perception, you'll need to explain what you mean (please cite references). Obviously, both involve processing by the brain because hearing/perception need to be registered by a listener. When judging whether Systems A and B 'sound different' to listener when playing music (not simple test tones), I don't see how we aren't talking about differences in perception. System A could have measurably more distortion of some type than System B, but they may not sound different to a given listener at a given time for some particular music content, either consciously or subconsciously. Conversely, the two systems could sound different because there are real audible differences under the given conditions, and a particular set of measurements which are performed may or may not be indicative of why they sound different.

 

[Phronesis]

He isn't a troll. It's a combination of being in love with the sound of his own words, inability to really comprehend anyone else's challenges to his arguments, and a desperate attempt to find a way to validate his audiophile solipsism. I think he actually believes what he says. He's just built a wall of rhetoric and logical fallacies around his misconceptions to defend them. There's really no reason to argue with him. He'll just go right back to his hippy dippy "we don't actually hear sounds- we perceive ideas about sound" argument again and it will all go in circles. I chalk him up as someone who is well spoken, but devoid of analytical thinking processes. We've seen the type here before. They are usually the loudest and most oblivious ones.

The irony is that essentially the same (with a few modifications) could be said about you. The difference is that I'll be tired of going in circles long before I rack up more than 17,000 posts. Some people have posted things here lately which have made the time spent worthwhile, but eventually it'll be almost all repetition and for me it will be time to move on.

 

[gregorio]

[1] I don't know why it's hard for some people to understand.
[2] And audibility of differences isn't simply a matter thresholds, since we're talking about music, not just pure tones (reference to simple thresholds won't be helpful in the Yanny/Laurel example).

1. Is that rhetorical? If not, it absolutely should be!!

2. I'll ignore the obvious contradiction that the Yanni/Laurel example is not music but let's run with your example anyway. Should an audio reproduction system just reproduce the audio or should perception play a part in the reproduction system? If, as you assert, perception should play a part, then who's perception, yours or your wife's? Should it process the sound in some way so only Laurel is perceived or so that only Yanny can be perceived? How would it know which one, how would it achieve that result and how would this apply to music, maybe a DAC just for Mozart, a different DAC for the Prodigy, etc.? How does any of this make any sense to you? An audio reproduction system just reproduces audio, that's it, it can't and doesn't know what you perceive! Don't you think it's up to the creators of the audio file and you and your wife's perception, rather than the audio reproduction system?

The difference is that I'll be tired of going in circles long before I rack up more than 17,000 posts.

Promise? Is it too much to ask that you'll stop going in circles before you rack up 388?

G

 

[Phronesis]

2. I'll ignore the obvious contradiction that the Yanni/Laurel example is not music but let's run with your example anyway. Should an audio reproduction system just reproduce the audio or should perception play a part in the reproduction system? If, as you assert, perception should play a part, then who's perception, yours or your wife's? Should it process the sound in some way so only Laurel is perceived or so that only Yanny can be perceived? How would it know which one, how would it achieve that result and how would this apply to music, maybe a DAC just for Mozart, a different DAC for the Prodigy, etc.? How does any of this make any sense to you? An audio reproduction system just reproduces audio, that's it, it can't and doesn't know what you perceive! Don't you think it's up to the creators of the audio file and you and your wife's perception, rather than the audio reproduction system?

The answers are apparent from the example. It's ideally supposed to sound like Laurel to all listeners, but some listeners are apparently hearing it as Yanny because of issues with distortion, noise, etc. There needs to be better recording and/or fidelity. Which one does it sound like to you? I can't even say which listeners have 'better' hearing in this case, all I can say that they hear the same objective sound differently. I think it's a fascinating case study and I look forward to reading a proper expert analysis of it.

 

[analogsurviver]

Hate to break it to you, but PCM is inherently free from this problem too! There is no phase misalignment between channels of PCM, zero, none at all. Historically there was one combination of specific equipment that led to this myth, recording on the EIAJ/Sony PCM-F1 family of semi-pro video-based PCM converters, then directly transferring the data to CD without a dub to PCM1620/30 or using one of the timing correction boxes of the time. The PCM-F1 family of ADCs sampled Left and Right alternately, and played back that way too, so within the system there was no interchannel phase or timing error. However, PCM1620/30, which was Sony's pro video-based PCM converter, and an integral part of CD authoring, sampled L and R simultaneously, as does the CD format. There was a potential for uncorrected F1 material to play from a CD with on channel delayed 1/2 sampling clock cycle. The error was so small it could be corrected with an analog all-pass filter (and was!), but that's THE ONLY combination that resulted in that error. DAT machines eliminated alternate channel sampling, and so did everything after that. You can prove it easily with basic testing, which I would suggest you do before posting nonsense like this.

The entire argument as to the inefficacy of PCM to handle binaural is myth based fabrication.

OK, the first I want to reply to the part that PCM is perfect regarding the phase delay between the two channels

It is NOT.

Many, many generations of processors after those you've mentioned, such errors STILL occur. And, to the contrary of your claim, not only with the one historical combination, but MANY combinations that are in the use today.

Want proof ? Record a square wave to any digital recording device in PCM - mono signal from a real hardware signal generator connected to both left and right channel input. You can check for the output using any oscilloscope fast enough not to introduce any error in phase coherence between the 2 channels by itself. Some slow digital (storage) oscilloscope might be not be good enough, but in general any oscilloscope should be suitable. Normally and usually, if the recording device is a soundcard and computer, the output of the DAC part of the soundcard will still be in sync at the output - IF you play back the recording ( or monitor the output during the recording ) using the very software as used for the recording in the first place.

Now store the PCM file just recorded to any storage you are normally using. Play back the PCM file loaded from this storage using the same soundcard with the same computer using ANOTHER playback software. You might still be lucky to still have the output in sync - but that is no longer guaranteed in all cases. It is perfectly possible for the error to occur as early as here.

Now send the same PCM file to a friend/acquaintance over the internet - which he/she will be playing on another computer with different settings, using yet another type of soundcard/DAC and yet another playback software. Of course, he/she still needs to check the output using an oscilloscope. He/she MIGHT get lucky to have the output of the both channels still in sync - but guaranteed is this definitely NOT.

Just because I did not post the evidence of such an error in this thread, that does not mean I did not do the basic testing.
For details, you can read my review of the iFi Micro iDSD Black Label DAC at

http://www.head-fi.org/users/335227/reviews .

After clicking that link, you are likely to get :
The requested page could not be found. We may be updating this page, so check back at a later time!

Now, I hope this does not mean I will have to upload all the photos of the square wave performance of the said DAC, both in PCM in DSD modes - AGAIN;
might/may be that all the pics are gone after the last major changes of head-fi.

As explained in my previous post, the delay of one channel versus the other is EXACTLY the rise time at whatever sampling frequency setting up to 192kHz. I do not have a PCM recorder with higher sampling frequency capability to test the said DAC, which is spec'd at 768kHz/32bit.

And no way this error can creep up just with this particular DAC - there must be other combinations of software and hardware that mess up the phase/sinhronization between the 2 channels in PCM.

How many times did you see ANY digital product actually tested for the (lack of) channel sinhronization lately ? The honest answer can be no other than NONE - because, even if the DACs are being tested using various "soundcards" ( Audio Precision is also nothing but a soundcard, but with the lowest distortion and highest S/N, an interface speciffically made to interface audio gear for computer based measurements ) - and very rarely BOTH channels are being measured while operating SIMOULTANEOUSLY - as most measuring software can only measure one channel at a time.

I took the channel sinchronicity in PCM also for granted - all CD-R recorders I have been working with and all Korg DSD recorders I am still using have perfect channel sinchronization - even MR-1 in MP3 recording mode.

The "fun" - or better said "not-so-fun" - begins the moment the files recorded on these machines enter the computer - and continue its way trough various Scillas and Caribdas of PCM digital., until they finally reach the headphones or speakers. No control over all these potential but deffinitely possible problems once the perfectly sinchronized / zero delay / phase cooherent 2 channel PCM files are sent out in a big wide digital world...

DSD, on the contrast, IS inherently free of lack of channel sinchronization. No DSD capable device I have yet been able to test did not test any different but perfect in this regard - regardless of the playback software (differences) , as long as it has been NATIVE playback of DSD.

If the DSD playback involves PCM at any stage, such as DoP - who knows what might happen, given the number of hardware and software combinations being simply too large to test them all.

 

[bigshot]

The irony is that essentially the same (with a few modifications) could be said about you.

Oh that's a good one! Don't answer the points made, just turn them back around. Nice technique!

 

[KeithEmo]

I agree with you absolutely.

The only catch on #2 is that, since we cannot "tap directly into the wires", we are always going to be stuck with some degree of interaction with human perception.
The closest we can come to controlling the situation would be to use the MRI and not ask the human involved at all.
But then we run the risk of considering something "audible" if everyone's brain reacts to the stimulus - even if nobody actually clams to perceive it at all.
Also, according to the latest brain research, there are actually be PHYSICAL interactions between our perceptions and other stuff going on in our brain.
In other words, the PHYSICAL threshold of what we can hear may actually vary depending on what we're thinking at the time (our brain can functionally alter the sensitivity of the physical apparatus).

I don't know if this has been shown to occur with hearing, but it most certainly has been shown to occur with our other senses, including sight...
The eye actually PHYSICALLY sees or fails to see certain things depending on what the brain has "told it to look for".
The part of our brain that processes sight can actually be programmed to be more or less sensitive to specific patterns... at which point the threshold at which those patterns become "visible" changes.
Of course, this is only possible because every one of our senses involves processing in the brain... there is no such thing as "a sensory threshold independent of the brain".
Our ears DO NOT function "like a microphone connected to a meter"; it's more like "a microphone that is controlled by and sends its output to a really complicated computer".
So, if you really want to determine "an accurate minimum threshold of detection" for something, then you must determine the brain state that makes the individual most sensitive to that stimulus, and then ensure to test under that condition.

1) Yes, I agree entirely.

2) This point I don't really agree with. What we actually hear and what we perceive are two different things. Perception is the identification, organisation and then interpretation by the brain of what we hear but as we know (and hopefully agree), this "interpretation" is a combination of all sorts of factors (sight, knowledge, past experience, biases, etc) and may have a lot, some, little or nothing at all in common with what our ears are actually hearing, which explains the difference between perception and hearing and therefore why they are two different things. When we perform hearing threshold tests, the goal is to discover the hearing threshold, NOT the perception threshold. This is why we need "controlled" tests, because "controlled" effectively means; to eliminate perception, thereby leaving us with only "hearing". In addition to the biases, sight and other factors which differentiates perception from hearing in the first place, there is another class of other factors which can affect our hearing thresholds/ability itself; fatigue (or it's opposite), temporary threshold shifts and conditions such as having a cold or hearing damage/deterioration for example. Regardless of this other class of factors, the goal is for perception NOT to be a factor when performing hearing threshold tests.

As @Phronesis was talking about audio reproduction systems then we are both agreed, perception is NOT part of the equation. But also, even when discussing hearing threshold tests, I believe that perception is still not part of the equation. In fact, the only time when perception should be part of the equation is when actually testing perception (or some aspect of it).

G

 

[bigshot]

How do MRI scans relate to actual hearing in more than just a general way? Can we track the MRI pattern for an oboe, or a frequency response curve, or the sound of Mozart? We can do all of this by simply putting on a set of headphones.