[Jaywalk3r]

Well no, that's not "all they're saying."

Actually, it is.

But anyway, at least other people understand the point.

Unfortunately, you do not appear to be among them.

 

[bigshot]

Perhaps you don't know what I've investigated, and haven't investigated some of the things I have.

There are finite limits to human hearing... frequency ranges, levels of distortion, volume levels, noise levels, time error, etc. Beyond that point no human, even under the best circumstances, can possibly hear. There may be psychological or biological reasons why a person might hear less than that, but the upper limit is a ceiling no one with human ears can hear beyond. That is what I'm talking about and there's absolutely nothing vague or loosey goosey about it. If you've investigated those thresholds, you sure aren't talking like you have. I know by your own words that you haven't bothered to test your own perceptual thresholds with controlled listening tests. That was how I got a grasp on the thresholds, and that is what gave me the ability to look at an abstract number and know pretty well what it represents in real world sound.

 

[TheSonicTruth]

So what could be causing the "5-7kHz bump" someone mentioned they heard several posts back, when they were comparing DACs? That's a significant aberration!

 

[bigshot]

It was caused most likely by expectation bias. And a 1.5dB response deviation up in the upper treble isn't significant.

 

[Davesrose]

It was caused most likely by expectation bias. And a 1.5dB response deviation up in the upper treble isn't significant.

Upper treble is considered 6khz-20khz. Mile posts keep seem to be changing in this thread. One of the arguments was that we should just default to page one of this thread. Well if you do look at that post, most links are to the "standard" snake oil of cables and such. With the one DAC link, its conclusion was that there was still "hairs width between the two in terms of sound"...and that's with particular testing standards. Seems to me, time after time, we're not isolating to one particular component (and to try to isolate to DAC is hard). IE Bigshot maintaining any device that's "transparent" sounds the same: that's not the same as a DAC...where one of the arguments is that an iPhone is good for most headphones (in which it's not just DAC but its headphone amp). Also, when it comes to instruments, bigshot mentioned piano as being one of the hardest ones to reproduce: pipe organ is much harder: fundamental frequency ranges that go below and above regular measured auditory hearing.

 

[castleofargh]

hey guys, please try to ease up on the personal fighting. this is escalating for no good reason.

fundamentally both sides are following an evidence based rational, which is good and fairly rare in here. so maybe we could just appreciate that our thoughts are closer than the last pages might indicate. we're in agreement with almost everything that matters. the only difference is that when we lack conclusive data, we have different standby positions. as both sides agree that they are indeed just standby positions waiting for proper evidence, do we really have to fight to the death over how we should put our legs while waiting for the evidence bus? it's not that important IMO.




@Phronesis . we've been through this, but if it's not testable, what use is it to us? are we theoretical audiophiles now? you're stuck on subconscious being important in the listening process, and no doubt it is in general. but in this specific context, is it?
here are the conditions required to make 2 devices affect us differently at subconscious level but not at a conscious one:
1/ the variation between the gears must be too small to trigger a conscious reaction or be outside the range of hearing.
2/ the variation must be big enough to be properly and consistently perceived by the ear despite various external and internal noises. and the subconscious reaction must be triggered by hearing, not body shaking, vision, smell, preconception... otherwise it's not about hearing anymore.
3/ the subconscious reaction mustn't cause any change at a conscious level(else it would fall under what's audible in a listening test).
4/ it still has to matter for the brain. as the general consensus seems to be that we constantly discard a great deal of sensory data without really processing it much, if only for the sake of being able to function and have coherent thoughts.

how often would you expect those conditions to be met? given how those conditions have somewhat contradictory requirements, your only hope would be a range in the area strictly below the known hearing thresholds. so even with this rather rational but optimistic idea, what could have a subconscious impact while not generating a conscious one, is likely to be highly specific. at least in amplitude, but most likely in all aspects. making it something, even in this open minded hypothesis, that would require hyper specific variations to occur between 2 devices. making the all thing a pretty rare occurrence, if it can happen at all. so that's far from your quasi systematic need to bring it up as a variable we can't dismiss when audibility is questioned.
again, we don't know much about the brain, and the general concept of subconscious seems to agree with a lot of human behaviors. I'm not trying to reject that model. I'm just saying that it's unlikely to have a place in the hearing part of feeling differences between gears. now if we count the all experience sighted, then I'd be tempted to put subconscious stuff at the top of the list along with preconceptions. but it's a very different matter.

 

[bigshot]

Upper treble is considered 6khz-20khz.

No it isn't. The frequency scale is logarithmic. It's divided approximately into the nine audible octaves. The top octave is sometimes referred to as "air" because it generally isn't a part of the output of musical instruments, other than upper harmonics. That octave is the least important one in the audible range.

By the way, super sharp transients are harder for microphones and transducers to reproduce than super low and ultra high frequencies. Nyquist nails 20Hz to 20kHz perfectly and professional microphones generally do too with room to spare. Sub woofers go down to 12-15Hz or so and super tweeters can go beyond 20kHz easily. Sharp transients are a lot harder for microphones, speakers and headphones to handle. The piano is a percussion instrument and has some of the sharpest transients of any musical instrument. Recording it well is a challenge.

I'd suggest you go read that first post through all the way, including the links, and go through the links in my sig thoroughly too. Lots to learn in there.

 

[castleofargh]

I learned on this (the original interactive page is down )

 

[bigshot]

That one works too! That divides it into two octave chunks and extends the range beyond the range of human hearing to make up the extra octave. Not much above 10kHz!

 

[Davesrose]

That one works too! That divides it into two octave chunks and extends the range beyond the range of human hearing to make up the extra octave. Not much above 10kHz!

Yes, cherry picking sources does seem to work. Meanwhile, other sources have the organ going above 16khz (and entire range well exceeding piano). And CD music is recorded down to 20hz, while pipe organ can go below that.

https://www.hifisentralen.no/forume...ass-og-musikk-080830-frequency-ranges.xls.pdf

 

[Joe Bloggs]

Yes, cherry picking sources does seem to work. Meanwhile, other sources have the organ going above 16khz (and entire range well exceeding piano). And CD music is recorded down to 20hz, while pipe organ can go below that.

https://www.hifisentralen.no/forume...ass-og-musikk-080830-frequency-ranges.xls.pdf

"Going above" does not mean that that part is responsible for any appreciable portion of its perceived signature sound. (certainly not if I'm categorically deaf to the whole octave at 38 years old)

And there's no LOWER limit to what any PCM format can record. It will happily record and play back 0.0000001Hz if you have the recording and playback equipment for it. (movable concerete blocks maybe?)

 

[Phronesis]

@Phronesis . we've been through this, but if it's not testable, what use is it to us? are we theoretical audiophiles now? you're stuck on subconscious being important in the listening process, and no doubt it is in general. but in this specific context, is it?
here are the conditions required to make 2 devices affect us differently at subconscious level but not at a conscious one:
1/ the variation between the gears must be too small to trigger a conscious reaction or be outside the range of hearing.
2/ the variation must be big enough to be properly and consistently perceived by the ear despite various external and internal noises. and the subconscious reaction must be triggered by hearing, not body shaking, vision, smell, preconception... otherwise it's not about hearing anymore.
3/ the subconscious reaction mustn't cause any change at a conscious level(else it would fall under what's audible in a listening test).
4/ it still has to matter for the brain. as the general consensus seems to be that we constantly discard a great deal of sensory data without really processing it much, if only for the sake of being able to function and have coherent thoughts.

how often would you expect those conditions to be met? given how those conditions have somewhat contradictory requirements, your only hope would be a range in the area strictly below the known hearing thresholds. so even with this rather rational but optimistic idea, what could have a subconscious impact while not generating a conscious one, is likely to be highly specific. at least in amplitude, but most likely in all aspects. making it something, even in this open minded hypothesis, that would require hyper specific variations to occur between 2 devices. making the all thing a pretty rare occurrence, if it can happen at all. so that's far from your quasi systematic need to bring it up as a variable we can't dismiss when audibility is questioned.
again, we don't know much about the brain, and the general concept of subconscious seems to agree with a lot of human behaviors. I'm not trying to reject that model. I'm just saying that it's unlikely to have a place in the hearing part of feeling differences between gears. now if we count the all experience sighted, then I'd be tempted to put subconscious stuff at the top of the list along with preconceptions. but it's a very different matter.

I'm thinking not so much about thresholds of perception, but rather the effects of time and memory on the operation and interactions of conscious and subconscious perception (and the two sides of perception can be at odds with each other, as we clearly see with visual illusions). This paper I've cited before illustrates what I'm talking about as far as the time aspect: https://www.nature.com/articles/s41598-018-24528-3.

I believe this has a lot of implications for testing. For example, if some subconscious effects take several seconds to kick in, a test which uses short music segments may miss those effects. But if you use much longer music segments to try to capture those effects, memory operating in different timeframes comes into play, and memory of what was heard at the beginning of a segment will be especially unreliable. Making matters worse, we don't generally have a good understanding of what's going on in the brain/mind with auditory perception because the science isn't good enough yet, so we can only speculate about the kinds of potential time and memory effects I'm talking about.

Bottom line for me right now is that I can believe that controlled blind testing with short music segments is *probably* sufficient to rule out the existence of large differences which people often claim to hear (perceive), and my experience shows that you don't always even need blinding if the other controls are good enough. But I'm not sure that such testing is sufficient to either rule in or rule out more subtle differences.

 

[Davesrose]

"Going above" does not mean that that part is responsible for any appreciable portion of its perceived signature sound. (certainly not if I'm categorically deaf to the whole octave at 38 years old)

And there's no LOWER limit to what any PCM format can record. It will happily record and play back 0.0000001Hz if you have the recording and playback equipment for it. (movable concerete blocks maybe?)

I think that then begs the question if audio reproduction should strive for recording aspects of the original recording venue or be within what’s considered average human hearing. I’ve seen similar debates about human perception vs digital reproduction in photography. White balance, for example is separate then human perception (which has rods and cones for different acuity at different wavelengths for daylight vs night).

Also when it comes to whether CDs do have content below 20hz: in CD mastering sounds are evidentially commonly filtered out below 20hz. Does seem to make it harder to send fundamental sub bass sounds to a subwoofer.

 

[Phronesis]

I think that then begs the question if audio reproduction should strive for recording aspects of the original recording venue or be within what’s considered average human hearing. I’ve seen similar debates about human perception vs digital reproduction in photography. White balance, for example is separate then human perception (which has rods and cones for different acuity at different wavelengths for daylight vs night).

This one has been discussed in multiple threads. Generally, recordings are engineered to create an experience which isn't necessarily meant to capture a live performance, and often there's no single live performance to capture anyway. So the best you can do is to make it sound like it did for the recording engineers, using their studio monitors in their studio. But if you're using headphones, matching the sound of studio monitors is already impossible because the acoustics are so different, so the best you can do is some sort of approximation which sounds good enough. For me, that's my ultimate criterion for evaluating gear: do I like how the music I typically listen to sounds on that gear? I don't get hung up over an ideal of "fidelity" which turns out to be ambiguous or impossible.

 

[KeithEmo]

Just to add a little actual current data to the "well known fact" that "human hearing is limited to frequencies between 20 Hz and 20 kHz"....

According to WikiPedia:
"The commonly stated range of human hearing is 20 Hz to 20 kHz. Under ideal laboratory conditions, humans can hear sound as low as 12 Hz and as high as 28 kHz, though the threshold increases sharply at 15 kHz in adults, corresponding to the last auditory channel of the cochlea."

Also, here's an interesting study that was done to determine whether the low frequency sounds produced by wind turbines could be heard by humans...
https://medicalxpress.com/news/2015-07-inaudible.html

Here was their conclusion:
"Test persons were asked about their subjective hearing experience, and these (also quantitative) statements were then compared by means of imaging procedures, namely by magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI). The results have shown that humans hear lower sounds - namely from 8 hertz on - which, after all, is a whole octave lower than had previously been assumed: an excitation of the primary auditory cortex could be detected down to this frequency. All persons concerned explicitly stated that they had heard something - whereby this perception had not always been tonal. In addition, the observations showed a reaction in certain parts of the brain which play a role in emotions.

Interestingly, along with effects on the auditory cortex, they noted effects on the areas affecting emotions.
(So, according to this study, you can not only hear as low as 8 Hz, but it can affect your emotions as well.)

I'm thinking not so much about thresholds of perception, but rather the effects of time and memory on the operation and interactions of conscious and subconscious perception (and the two sides of perception can be at odds with each other, as we clearly see with visual illusions). This paper I've cited before illustrates what I'm talking about as far as the time aspect: https://www.nature.com/articles/s41598-018-24528-3.

I believe this has a lot of implications for testing. For example, if some subconscious effects take several seconds to kick in, a test which uses short music segments may miss those effects. But if you use much longer music segments to try to capture those effects, memory operating in different timeframes comes into play, and memory of what was heard at the beginning of a segment will be especially unreliable. Making matters worse, we don't generally have a good understanding of what's going on in the brain/mind with auditory perception because the science isn't good enough yet, so we can only speculate about the kinds of potential time and memory effects I'm talking about.

Bottom line for me right now is that I can believe that controlled blind testing with short music segments is *probably* sufficient to rule out the existence of large differences which people often claim to hear (perceive), and my experience shows that you don't always even needing blinding if the other controls are good enough. But I'm not sure that such testing is sufficient to either rule in or rule out more subtle differences.