The highlight moments in life, when your in a state of blissful euphoria, "experiencing" something subjectively better, and are not "corrupted" with factual measurement data...
Thoughts on a bunch of DACs (and why delta-sigma kinda sucks, just to get you to think about stuff)
- Thread starter purrin
- Start date
johnjen
Headphoneus Supremus
You are certainly entitled to your opinion, as am I.
And if these terms cause you "psychological discomfort" then don't use them by all means.
As for placing me "in a disadvantaged position from the start, forcing to accept wrong assumptions as a truth", well, I simply don't agree.
Why?
Because it is my choice pure and simple to mean what I say, as it is your choice to let such meanings conform to those of others, if you so choose.
And really, do you want to be the grammar police?
And as was stated previously, to which I agree, I see this as a HOBBY, a FUN hobby and the picking of nits usually isn't much fun.
JJ
johnjen
Headphoneus Supremus
And speaking of opinions…
SOME measurements can "deliver objective, valid and reliable data" but not all.
Making measurements is a tricky business and as has been pointed out by others, can all to often result in impressive looking data that is utterly misleading and does not accurately, nor with precision, nor repeatability, reflect the DUT (Device Under Test).
Those 3 criteria (accuracy, precision, repeatability) comprise the bedrock for 'real' results using the scientific method and are all to often 'overlooked' (to put it mildly).
And as we all know, knowledge is based upon factual data that reflects, with accuracy, precision, and repeatability, the reality it is trying to measure.
So when I see a CFL desk lamp (or is it a led lamp?) in close proximity to a high precision data gathering device, for me it calls into question any and all results obtained by the aforementioned high precision data gathering device setup.
It makes me wonder what other 'quirks' are also in play, like say s/w setup/configuration and other environmental conditions.
AND
Can you tell me of ANY measurements that are designed to analyze and provide meaningfully relevant results,
for music, as it is being played,
you know dynamically?
Because THAT is what all of this audio equipment is meant and designed to do, play music, not test tones or sweeps or square waves etc.
Indeed as has also been pointed out recently, all of our 'standard tests' are woefully deficient at describing HOW a device will sound, which when it comes down to it is, the #1 with a bullet, criteria for WHY and HOW we actually use this gear.
Which points right back to what I wrote earlier "And since both approaches (objective and subjective) have big holes and problematic conclusions in their 'data gathering' and analytic abilities"…
Yes, SOME measurements can "deliver objective, valid and reliable data" but not all.
And even the most sophisticated tests available, all of which are mainly used to determine if any specific device is operating according to it's own specifications, these tests which comprise these very same specifications are not meant to relate to it's sound quality.
Why?
Because they weren't deigned for that purpose,
AND,
sound quality is a purely subjective evaluation.
IOW, how does a group of static measurements directly relate to the sound and how are they any different than other similar gear,
and here is the kicker,
which is 'better'?
Because in the end we ALL will choose what is 'better', it's just that some make choices that comprise criteria that is FAR different than others.
And this entire 2-sided 'debate' is an example of how we learn what IS and is NOT the best choice for one's personal criteria of what constitutes 'Better'..
Just some thoughts to ponder.
JJ
Sonic Defender
Headphoneus Supremus
Hello JJ,
That is quite a bit of information, thank you for the thoughtful reply. I understand about the strengths and weaknesses of measurement. There is always error involved, and error can come from more than one contributing source at the same time. To be fair, in the context of the discussion at hand we are talking pretty specifically about measuring audio devices using well validated methods and tools. The measurements will never be without error, but the accuracy can still be high enough to reliably draw conclusions from. Absolute facts are very hard to come by, and always subject to new investigation. Saying that, the set of objective measurements that are typically deployed by objectivists over at say ASR are pretty robust and have been well vetted for validity and reliability, same with the measurement tools. Are they absolutely perfect? Not to darn likely, but they are clearly more than adequate.
The claim of some special components of sound that nobody can describe, identify or measure is usually what is proposed by subjective proponents in objection to the clinical notion of measuring sound. I certainly cannot refute the existence of such unknown components of sound, or phenomena within the hearing brain, it may even be reasonable enough to speculate of their existence, but nonetheless it is nothing but speculation and as far from fact or true knowledge as can be possible at this point. Your points are well taken though, and I will; however, leave you with this thought. You mentioned the possible variability in measurement results as a reason that you don't like to be too bound by them. Fair enough. I think it may be worth considering how subjective observations are orders of magnitude more variable (look at the difficulty people have describing what they hear) and without any precision at all. Considering those confounds, and they aren't the only confounds found within subjective evaluation, how much confidence should be held in them?
Frankly, we do need both subjective and objective data to better understand and communicate our experiences. I have only pointed out that some claims are extremely speculative and yet are presented as fact, that is a semantic issue, I do not per say have any issue with someone trying to relay their love of the hobby through discussing their subjective experience. It is what drew me to this community many years ago and keeps me part of it still. There is value in subjectivity and I have never felt or claimed otherwise. I feel that people may be taking my words and thoughts and interpreting them a little too broadly.
That is quite a bit of information, thank you for the thoughtful reply. I understand about the strengths and weaknesses of measurement. There is always error involved, and error can come from more than one contributing source at the same time. To be fair, in the context of the discussion at hand we are talking pretty specifically about measuring audio devices using well validated methods and tools. The measurements will never be without error, but the accuracy can still be high enough to reliably draw conclusions from. Absolute facts are very hard to come by, and always subject to new investigation. Saying that, the set of objective measurements that are typically deployed by objectivists over at say ASR are pretty robust and have been well vetted for validity and reliability, same with the measurement tools. Are they absolutely perfect? Not to darn likely, but they are clearly more than adequate.
The claim of some special components of sound that nobody can describe, identify or measure is usually what is proposed by subjective proponents in objection to the clinical notion of measuring sound. I certainly cannot refute the existence of such unknown components of sound, or phenomena within the hearing brain, it may even be reasonable enough to speculate of their existence, but nonetheless it is nothing but speculation and as far from fact or true knowledge as can be possible at this point. Your points are well taken though, and I will; however, leave you with this thought. You mentioned the possible variability in measurement results as a reason that you don't like to be too bound by them. Fair enough. I think it may be worth considering how subjective observations are orders of magnitude more variable (look at the difficulty people have describing what they hear) and without any precision at all. Considering those confounds, and they aren't the only confounds found within subjective evaluation, how much confidence should be held in them?
Frankly, we do need both subjective and objective data to better understand and communicate our experiences. I have only pointed out that some claims are extremely speculative and yet are presented as fact, that is a semantic issue, I do not per say have any issue with someone trying to relay their love of the hobby through discussing their subjective experience. It is what drew me to this community many years ago and keeps me part of it still. There is value in subjectivity and I have never felt or claimed otherwise. I feel that people may be taking my words and thoughts and interpreting them a little too broadly.
Last edited:
GearMe
Headphoneus Supremus
- Joined
- Feb 17, 2013
- Posts
- 4,370
- Likes
- 7,152
Interesting discussions on a variety of fronts...
As somebody that leans objectivist for electronics and subjectivist for transducers, my thinking is that a level-matched abx test would resolve the listening differences (if any) for electronics with the additional caveat that, if amps are being compared, they are both are 'up to the task' for the transducers they're being paired with.
In the end, if the test is well-designed and someone can consistently differentiate two pieces of gear that measure the same from an audibility perspective, then they can hear the difference even though it didn't show up in the measurements.
As somebody that leans objectivist for electronics and subjectivist for transducers, my thinking is that a level-matched abx test would resolve the listening differences (if any) for electronics with the additional caveat that, if amps are being compared, they are both are 'up to the task' for the transducers they're being paired with.
In the end, if the test is well-designed and someone can consistently differentiate two pieces of gear that measure the same from an audibility perspective, then they can hear the difference even though it didn't show up in the measurements.
johnjen
Headphoneus Supremus
Sonic Defender, wrote…
"The claim of some special components of sound that nobody can describe, identify or measure is usually what is proposed by subjective proponents in objection to the clinical notion of measuring sound. I certainly cannot refute the existence of such unknown components of sound, or phenomena within the hearing brain, it may even be reasonable enough to speculate of their existence, but nonetheless it is nothing but speculation and as far from fact or true knowledge as can be possible at this point."
Oh but SOME people CAN describe and identify "special components of sound" and quite well at that.
"I certainly cannot refute the existence of such unknown components of sound"
But for SOME they are NOT "unknown components of sound" at all.
Rather, some simply can't hear them and for a variety of reasons, and no it isn't "nothing but speculation and as far from fact or true knowledge as can be possible at this point" either.
Again SOME people simply can't hear these levels of SQ, while others CAN, and can tell when they are missing.
And to limit the scope of "true knowledge" to just gathering "clinical" data while in such an unnatural setting and while listening to music also hampers proper data gathering due to the way the brain actually operates while listening to music.
And lastly from those professional sound experts and others whose hearing I know DOES hear these finer details of SQ, they have raised some serious questions about ASR and its means and methods, not to mention it's reputation, which has come under scrutiny multiple times and in multiple ways.
Like I stated its not easy to make accurate, precise, and repeatable measurements in the first place, even from the professionals, and from the examples I have seen, ASR comes up rather short in this regard.
And drawing questionable conclusions all the while not actually listening to or for SQ, tells me his conclusions are fraught with bias which is the death knell for any form of useful information, at least for me who has decades of experience on both sides of this 'debate'.
And finally, we all make personal choices which by definition ARE subjective at there very core.
There is no way around this, we are after all subjective at OUR very core.
JJ
"The claim of some special components of sound that nobody can describe, identify or measure is usually what is proposed by subjective proponents in objection to the clinical notion of measuring sound. I certainly cannot refute the existence of such unknown components of sound, or phenomena within the hearing brain, it may even be reasonable enough to speculate of their existence, but nonetheless it is nothing but speculation and as far from fact or true knowledge as can be possible at this point."
Oh but SOME people CAN describe and identify "special components of sound" and quite well at that.
"I certainly cannot refute the existence of such unknown components of sound"
But for SOME they are NOT "unknown components of sound" at all.
Rather, some simply can't hear them and for a variety of reasons, and no it isn't "nothing but speculation and as far from fact or true knowledge as can be possible at this point" either.
Again SOME people simply can't hear these levels of SQ, while others CAN, and can tell when they are missing.
And to limit the scope of "true knowledge" to just gathering "clinical" data while in such an unnatural setting and while listening to music also hampers proper data gathering due to the way the brain actually operates while listening to music.
And lastly from those professional sound experts and others whose hearing I know DOES hear these finer details of SQ, they have raised some serious questions about ASR and its means and methods, not to mention it's reputation, which has come under scrutiny multiple times and in multiple ways.
Like I stated its not easy to make accurate, precise, and repeatable measurements in the first place, even from the professionals, and from the examples I have seen, ASR comes up rather short in this regard.
And drawing questionable conclusions all the while not actually listening to or for SQ, tells me his conclusions are fraught with bias which is the death knell for any form of useful information, at least for me who has decades of experience on both sides of this 'debate'.
And finally, we all make personal choices which by definition ARE subjective at there very core.
There is no way around this, we are after all subjective at OUR very core.
JJ
Last edited:
sajunky
Headphoneus Supremus
It is a subconscious discomfort, as it affects the process of collecting facts and interpreting results. We are accepting logical consequences which are behind a meaning of the term, not realising that it is linked to the false objectives.
For an example: A hard core objectivist claims that a crusty old objectivist is the one who believes audible transparency has already been reached (implicated subconsciously by 'objective' measurements), on the other side would be a subjectivist who goes wild over the newest OpAmp they swapped in (even though the previous one was audibly transparent). A subjectivist is the one who is supposed to argue who has the most pleasure from swapping opamps: objectivist or subjectivist. By accepting a subconcious message behind meaning of the term, a person would like to become an objectivist, as nobody wants to be incredible stupid, isn't? A defense method would lead to the elaborated dispute that measurements do not show what is heard by a human (which absolutely true and a good defense, in fact different opamps sound different). A problem is that a subjectivist forgets that there is no opamps in his/her system anymore, a system has achieved level of transparency that opamps are not able to deliver and that an 'objectivist' use a supermarket type of the equipment, brands like SMSL/Fiio.
Am main goal has been achieved, as a discussion is already diverted from the initial 'objectivists' claim that
"any reasonable DAC has a clock that performs well beyond the capability of any human ear. The whole magic clock tweaking stuff people do is absolutely measureable, but not resulting in audible difference."
Last edited:
This is a good point. BOTH perspectives are totally valid for me.
I have no longer any mental conflict between the topics discussed here...
Why?
Because I have been able to rationalize all data/measurements as "static" points in time. Very useful & necessary for "data-points" and expected performance of a device measured,
But I also differentiate it from "live". The continuously working/streaming "flow" of music "over time"(moving).
This is where I believe much of the misunderstandings exist between the "subjective" & "objective" sides.
An analogy would be like the physical measurements of an athelete...You can know their "specs" (height, weight, power, health, ratings/averages, like baseball stats...)
these "measurements" describe their "potential" and expected performance.
Then you observe them play, in actual "realtime".. continuous live streaming, within your variables & settings( your ears, headphones, gear, room, streaming, etc).
Why do we have "correction data" algorithms, even in micro SD cards, if the real world is that simple?
rude awakening,,, ANS=In real world, nothing is simple..
Last edited:
Another observation which is more on topic with the thread title, is that I am of the opinion that Delta Sigma has reached performance levels, to no longer be any compromise in comparison to an R2R.
Specifically. in the volatile and fast evolving DAP market, I have observed and tried various R2R & Delta Sigma implementations.
Although with desktop units, anything goes.
Specifically. in the volatile and fast evolving DAP market, I have observed and tried various R2R & Delta Sigma implementations.
Although with desktop units, anything goes.
johnjen
Headphoneus Supremus
One of my, admittedly subjective assessments, is centered around this idea of the "flow of music over time".
I call it having a sound field that plays the music in 4D.
That being, the sound field is stable for each and every 'Voice' ('Voice' means any instrument being played, including singers etc.) in 3D space but also this stability remains thru time (4D) and even if a particular 'Voice' is supposed to move, it does so with stability (ie no jumping nor uneven movements within the sound field, especially as the 'Voice' rises and falls in frequency) just like in 'real life'.
I would ask how ANYONE could measure that acoustic trait, yet it is clearly audible, especially when you know to listen for it.
JJ
Last edited:
sajunky
Headphoneus Supremus
I do never experience a voice positional stability problems, perhaps it is because I am sensitive to false harmonics, my reaction to this comes first. Simplified sound in result of DSP filtering cause bluring individual voices into one, which most of people hear as an increased clarity. However because individual voices carry different harmonics and reverbations, it leads to false harmonics. I perceive it similar to the unacceptable level of THD distortions. Phase linearity problems also cause 'jumping', but it is in result of chosing wrong filters, I don't think you meant that.
All laboratory equipment work on a principle of time averaging. The above mentioned "Voice raises and falls" cannot be detected as averaging time is longer than these events.
castleofargh
Sound Science Forum Moderator
- Joined
- Jul 2, 2011
- Posts
- 10,452
- Likes
- 6,077
I suspect that you're talking about something that stereo albums on headphones cannot do well.
Tell me if I understand you correctly. If we consider a real sound source in front of you, it will simply feel like it's over there, and if it was to move, it would do in a way that agrees with the actual displacement. No jumping around, coming closer or going up, just the audio impression in agreement with the actual displacement.
But with headphones, that might not be the case. Things might feel like they go up or move closer, or kind of teleport from a position to the next before resuming a more reasonable movement. To make things worst, if you happen to move your head, the already small stage could collapse inside your head(at least for a moment).
If it's something like that, the best answer would be speakers, or a fairly good HRTF simulation(as in one that is custom to you, or one that happens to work fine for you with some luck).
I use a Realiser A16. For those not familiar with that toy, you measure some speakers in a room while looking at a bunch of directions while having mics in your ears. You then also measure the FR of the headphone you'll use with the same mics in your ears. Then by the magical power convolution, the A16 tries to make your headphone sound like your speakers, and the signal changes as you turn your head to maintain a correction that comes closest to one you measured in that general direction. My entirely biased opinion is that the simulation works well.
If you can procure some binaural microphones(maybe start with super cheap little mics and some DIY to have them sit near your ear canal), you can use impulcifer which is free. Troubleshootings and discussions here https://www.head-fi.org/threads/rec...ses-for-speaker-virtualization.890719/page-40
And if I'm completely missing your point, well, ooops. ^_^
justsomesonyfan
Headphoneus Supremus
- Joined
- Sep 24, 2013
- Posts
- 1,580
- Likes
- 1,901
fellas.. gonna revive this thread, anyone tried ak4499ex vs the old r2r big hitters?
pcm63? pcm1704uk?
pcm63? pcm1704uk?
sajunky
Headphoneus Supremus
Many tried, no chance. You can try this one:
http://audio-gd.com/HE/HE7MK3/HE7mk3EN.htm
In a modern design, no more vintage sound from PCM1704-UK chip (six of them). Or eight.
chesebert
18 Years An Extra-Hardcore Head-Fi'er
- Joined
- May 17, 2004
- Posts
- 8,898
- Likes
- 4,055
I am burning in a new dac based on PCM1795 - it’s relatively old and cheap. It must have been over a decade since my last dac that used only IC without some FPGA upsampling/reconstruction.
Users who are viewing this thread
