Oh, my bad. It seems like he acknowledges dynamics/distortion are important and not entirely dependent on tonality?Read the whole of that thread on twitter.
Oh, my bad. It seems like he acknowledges dynamics/distortion are important and not entirely dependent on tonality?Read the whole of that thread on twitter.
I must confess I dont understand the "dynamics/ distortion" split used there- especially the dynamics part. And yes, if you know the tonality/ FR at SPL X dB but it gets increasingly noticeable distortion above that it will affect the tonality accordingly.Oh, my bad. It seems like he acknowledges dynamics/distortion are important and not entirely dependent on tonality?
Distortion and dynamics are related, but not the same thing. Distortion is a side-effect of imperfect dynamic reproduction.I must confess I dont understand the "dynamics/ distortion" split used there- especially the dynamics part. And yes, if you know the tonality/ FR at SPL X dB but it gets increasingly noticeable distortion above that it will affect the tonality accordingly.
This might be my last reply, I'll try to sum it up. All you perceive in terms of headphone sound or any other sound is the frequency response at your eardrum. This is affected by the measured frequency response of the headphone (on a dummy head, lets say GRAS), it's measured distortion/compression, and "headphone transfer function" (we'll call it) when you wear the headphone on your own head (which is how your own ear & head anatomy influence the headphones transmitted frequency response when measured at your own eardrum vs that of the dummy head). Your "micro & macro dynamics" that you mention, is literally just a semantic label you've put on an element of what you yourself experience when you listen to headphones, but whatever that is....it's explainable by the frequency response that is being received at your eardrum which is a result of all the stuff I underlined in this paragraph.Yes, this is exactly why it should be obvious that headphones sound different partly due to how they handle variations in loudness over time. No headphone is capable of reproducing audio perfectly. Headphones will reproduce audio with imperfect tonality (partly due to the changes at the eardrum you keep mentioning). It stands to reason that headphones will also reproduce dynamics (variations in loudness over time) imperfectly, in large part due to drivers having different physical mechanisms for producing sound which have levels of responsiveness that vary with frequency, as seen in those graphs for the RS1x.
I've also been slightly disappointed at the level of discussion. People seem to think either dynamics aren't real or don't make a difference. The entire point of mastering audio recordings is to refine dynamics. You can't deny the importance of dynamics without denying the importance of mastering.
I think I finally get what you are getting at. And I still have to answer the same. It would show up on the FR graph.Distortion and dynamics are related, but not the same thing. Distortion is a side-effect of imperfect dynamic reproduction.
If you used the reply function, there may not be such confusion.No, you seem to confuse me with someone else, please look what I am saying and what not, I was talking of dynamics and transients in the sense of CSD, which is a real thing. Please get this right.
I genuinely appreciate the reply, I'll try to keep this concise but I believe we misunderstand each other somewhat. I'm aware of how tonality can affect your perception of certain dynamic qualities in music. I'm also well aware that factors like unit-to-unit variation, pad and tip size/wear, and person-to-person differences in how headphones sound to our eardrums can affect our perception of music. I say this as someone who spends an embarrassing amount of time tip-rolling my IEMs and someone who spent an even more embarrassing amount of time manipulating and mixing sounds in various DAWs. Because of your comments, I'm also now aware that FR graphs on the internet are imperfect, especially when applied to our personal perceptions of these headphones.This might be my last reply, I'll try to sum it up. All you perceive in terms of headphone sound or any other sound is the frequency response at your eardrum.
I think you're getting close.I think I finally get what you are getting at. And I still have to answer the same. It would show up on the FR graph.
Say headphones #1 has the same FR as headphones #2 both @80dBSPL. Both have no compression issues.
Now lets assume #1 has more trouble with dynamics as you call it than #2 and shows compression when playing loud (lets say 100dBSPL). I think you described it as the peaks in the signal get topped off. (correct?)
When you measure them @100dBSPL they would not show the same FR anymore. #1 would show dips at the frequencies it has trouble with playing loud. And the THD on #1 would shout up at those frequencies. So these headphones would actually not measure the same.
Did I finally understand what you were asking about?
The same FR = the same sound is simply not true, as you neglect transient behavior, i. e. CSD and any time based errors in general as the FR plot is a purely static thing. And spatial differences as well. Being objectivist and scientific does not mean being simplistic.Your claim is that this is because of some imaginary traits called "micro dynamics" and "macro dynamics".
We clearly say that it is FR. The same FR at the eardrum = sounds the same. That is also what Dr. Sean Olive says. He explains that what you measure as the same FR will most likely be different from the same FR on someone's head because of many different reasons. And the headphones will most likely not measure the same at the eardrum every time you wear them. All FR related.
If the driver can play the complete music spectrum (20Hz - 20kHz) it can follow all signals contained within the music signal. Or it would show in the distortion data and FR at the peak SPL value of the signal.I think you're getting close.
You're talking about distortion in headphones that results from clipping. You're correct that this would show up on an FR, but this is a very specific type of dynamic imperfection/compression issue that isn't directly indicative of overall dynamic ability.
Dynamic ability overall is more like "compression ratio" (or maybe "compression profile"?) during music playback. As long as the "compression ratio" doesn't result in hugely overemphasized peaks in loudness, there won't be any clipping distortion.
If the relative loudness of frequency A in a drum hit is recorded as
12-15-100-0, a headphone, being an imperfect playback device, might "compress" these to relative loudness levels to
15-20-95-5 during playback. This will not result in any clipping and therefore this will not change the FR, but the sound is being expressed imperfectly in a way that is both audible and measurable. If the headphone played the 100 in the recording as 102, that's when you get clipping and resulting changes to FR (assuming that the headphone is incapable of cleanly outputting 102).
The bit in your second paragraph, where you say "We perceive sounds as frequency responses at our ear drum that change over a period of time" and your second bit where you say "What I'm referring to as dynamics is Loudness x Time" - I think I know what you're getting at, but I think that's actually information that is already covered by the measured frequency sweep of the headphone, all of that information is in the frequency response of the headphone and it's associated distortion measurements. The same is true for the last bit where you say "During music playback, different headphones will handle spikes in loudness (the ones recorded into the song) differently. There are also differences in how accurately and distinctly headphones will handle small changes in loudness as well, resulting in audible differences in qualities like texture, soundstage depth, and impact". That is all literally in the frequency response and distortion measurements, if you get the frequency response right in a low distortion headphone then all of those elements will be optimised.......it's easy to see the influence on these qualities you mention when using EQ to manipulate headphones. It's obvious that by shifting around parts of the frequency response through EQ will influence all the points you've mentioned. And as to your perceptions of applying "a very rudimentary dynamic effect to a piece of music by manually adjusting volume level during playback. Just spike the volume knob for a split second and take note of how it affects your music." then of course that will affect how it sounds, that's a change in overall SPL which affects tonality due to Fletcher Munson effects, that's a seperate phenomenon unrelated to the headphone itself but rather a property of the human ear. But no, it's all in the frequency response received at your eardrum which is influenced by the measured frequency response of the headphone (assuming your unit is measured accurately), it's distortion, and your "headphone transfer function" of your own anatomy, the detail of which I mentioned in my prior post(s).I genuinely appreciate the reply, I'll try to keep this concise but I believe we misunderstand each other somewhat. I'm aware of how tonality can affect your perception of certain dynamic qualities in music. I'm also well aware that factors like unit-to-unit variation, pad and tip size/wear, and person-to-person differences in how headphones sound to our eardrums can affect our perception of music. I say this as someone who spends an embarrassing amount of time tip-rolling my IEMs and someone who spent an even more embarrassing amount of time manipulating and mixing sounds in various DAWs. Because of your comments, I'm also now aware that FR graphs on the internet are imperfect, especially when applied to our personal perceptions of these headphones.
But your second sentence is incorrect. We perceive sounds as frequency responses at our ear drum that change over a period of time. Sound is objectively a 3D experience:
1st dimension - Frequencies
2nd dimension - Loudness (When you plot Frequency and Loudness, you get FR)
3rd dimension - Time
This is why I say any sound can be reproduced by playing the right frequencies at the right loudness levels at the right times.
What I'm referring to as dynamics is Loudness x Time.
If you've spent your time as an audio enthusiast relying solely FR to conceptualize sound, I can understand how this seems like bullcrap, but bear with me: there is no way to EQ a sound to make it identical to a louder sound. There is no way to adjust the dynamics (loudness over a period of time) of a sound to make it identical to an EQ'd sound. Test this for yourself:
Apply a very rudimentary dynamic effect to a piece of music by manually adjusting volume level during playback. Just spike the volume knob for a split second and take note of how it affects your music.
Try to EQ your music to produce the same effect. You can even adjust gain. You will find this is impossible.
This sounds really stupid and obvious, but this is dynamics at its simplest. During music playback, different headphones will handle spikes in loudness (the ones recorded into the song) differently. There are also differences in how accurately and distinctly headphones will handle small changes in loudness as well, resulting in audible differences in qualities like texture, soundstage depth, and impact.
Yes, all headphones can follow all signals very well. The type of distortion I believe you're thinking of only occurs if it follows a certain kind of signal (the loudest frequencies at the loudest moments) particularly badly.If the driver can play the complete music spectrum (20Hz - 20kHz) it can follow all signals contained within the music signal. Or it would show in the distortion data and FR at the peak SPL value of the signal.
I'll make one last attempt to explain what I'm talking about with a one-minute video and I'll concede if this doesn't work:The bit in your second paragraph, where you say "We perceive sounds as frequency responses at our ear drum that change over a period of time" and your second bit where you say "What I'm referring to as...
Hello....yep, I watched that video, that's "supporting" what I've been saying.....ie that changes to the frequency response will affect the frequency response of the headphone and therefore the frequency response received at your ear & therefore how it sounds, what that video is showing is totally expected.....and to be honest it's not showing much to be fair.Yes, all headphones can follow all signals very well. The type of distortion I believe you're thinking of only occurs if it follows a certain kind of signal (the loudest frequencies at the loudest moments) particularly badly.
I'll make one last attempt to explain what I'm talking about with a one-minute video and I'll concede if this doesn't work:
This will show you what music "looks" like in terms of FR. For the purposes of this discussion, let's pretend this is measuring a headphone's output. You can think of the adjustments to EQ as equivalent to adjustments in the headphone's tonality. Do you see how frequencies across the spectrum move up and down throughout playback? These aren't recording artifacts, this is showing that these frequencies vary in loudness over time. The accuracy and precision of these movements is what I'm referring to as dynamic ability.
You might want to say that inaccuracies in these movements would result in distortion, and therefore show up in FR. The movements of these frequencies should only result in distortion if the loudest moments are too loud for the headphone to handle, leading to clipping of the audio signal (which is distortion which is shown in FR). Most headphones are good at keeping the loudest moments from getting too loud. But what if a headphone makes the quietest moments less quiet than the recording intended? What if it plays slightly loud frequencies a little louder than intended? No distortion, but still inaccurate.
How can differences in the accuracy and precision of these movements be explained with a headphone's FR, which is basically just a very specific and imperfect EQ profile? A graph plotting Frequency against Volume does not contain any information about Time.
Appreciate your time. I agree that the phenomenon you're describing is almost completely negligible when it comes to evaluating headphones. I just assume tonality/FR for a headphone remains constant throughout music playback.Hello....yep, I watched that video, that's "supporting" what I've been saying.....ie that changes to the frequency response will affect the frequency response of the headphone and therefore the frequency response received at your ear & therefore how it sounds, what that video is showing is totally expected.....
This comment is curious to me. While I haven't heard any of the better headphones reviewed here, I was under the impression that IEM's gave better and easier sub bass because they seal the ear canal. My experience with a couple of them has been great.I have yet to hear an IEM with good sub-bass. Response of the Blessing 2 was better than average (with EQ) but still very far from any over the ear headphone.
A frequency being played slightly quieter than normal won't result in distortion, it's the opposite, if you put less energy into the driver at that frequency level then it won't be forced to physically move as much, so it will have lower distortion as it's further away from it's mechanical extremes & limits of movement. Thinking about it, and I don't know if it's true, an edge case might be when you're at the very lowest SPL limits where the driver is producing virtually no measurable sound then it could be that you'd get some kind of incomplete response that wouldn't reflect the input signal accurately that might be measured as distortion, but that's just an intuition, but that would be at such low levels that it would be of no listening level consequence anyway.Appreciate your time. I agree that the phenomenon you're describing is almost completely negligible when it comes to evaluating headphones. I just assume tonality/FR for a headphone remains constant throughout music playback.
I don't understand why a frequency being played slightly quieter than normal would result in distortion.