Dynamic range of music, in relation to headphones

Sep 15, 2013 at 5:08 PM Thread Starter Post #1 of 24

MatsGyver

Head-Fier
Joined
Jul 20, 2013
Posts
53
Likes
13
I probably wrote 700 words that explained this to a larger extent, but headfi forum would not let me upload pictures, and would not let me go back from the picture menu, so I lost it all. So I`ll make it as short, precise and to the point as I can.

A much longer article was originally written in my main language, and I wrote it as a personal piece mainly for myself, to try to make sense of the subjective vs objective view that we find on headfi and HiFi in general.

In my experience, when you have two strong conterparts like this that dont see eye to eye, both usually have something they are right about, but the message gets lost in communication, or lack of.

I mean its not like more precise headphones are the most musical if the recording is really bad and has a dynamic range compressed -6dB from clipping.

And maybe electrostatic headphones actually have the potential to be more precise, but what if they cant produce the dynamic range of the most dynamic classical recordings? Are they still better?

Or that much of todays music with a dynamic range of -10dB sound bad when played the way it is through precise Stax, Sennheisers or Audeze. Maybe Beats can be more "musical" because it gives music with low dynamic range more dynamics because it is underdamped and unprecise.

Dynamic means movement right? and more dynamic range means more movement? and more movement means you need harder damping to keep the same ammount of mass under control. Knowing every action has an equal and opposite reaction.

In relation to musicality, not precision Im starting to think modern music sound more musical and more dynamic, when played through underdamped headphones like Beats. Or in rock, played with Grados. Because this type of Music usually have low dynamic range and because the Beats and Grados are underdamped.

Audiophile recordings will sound better with music that has more dynamic range because quality headphones have lighter diaphragms and are harder damped and therefore have better control on the diaphragm making them more detailed.

With really dynamic recordings orthodynamics and electrostatics dont have the physical space to fully express the largest dynamic movement of the recording, probably because they are not dynamic enough. Dynamic drivers are not constricted inside a 2,5mm space between large magnets, and may have an advantage over electrostatics and orthodynamics in this case. It is a question of weighing the pros against the cons, and if the dynamic range is more important maybe the Sennheiser HD800 may be a better suited headphone for music with very high dynamic range, even though the Stax is more detailed.

To make a long story short. Dynamics of music matter. The most dynamic recordings may be better on dynamic headphones like Sennheiser HD800 because its diaphragm can work unrestricted. For less extreme dynamic recordnings the electrostatics may be better and more precise. But for the recordings with the lowest dynamic range and for the music that is the most artificial and compressed, the less precise dynamic headphones can again sound better because they are underdamped and less precise. They can even give music with little dynamic range more dynamics because of the unpresise loose damping. It may be less true to the actual recording, but it doesent matter if the actual song playing actually sounds better. Precise headphones reveal more of the distortion, compression and clipping that we find in bad recordings and recordings in general that are heavily influenced by the loudnesswar, and distortion is the least musical sound I know of. Less is better.

In my experience you cant get a headphone that is both precise and detailed for the best recordings, that at the same time smooths over rough recordings making them at sound fuller. If such a headphone exist, I would definitely want one.

I also made these graphs, sorting music genres after dynamic range (vertical) and tonality (horisontal). And plotted in the different headphones after tonality (horisontal) and damping (vertical), to see the headphones tonality and damping-factor in relation to music genres dynamic range and tonality.

The graphs display dynamic range of music genres in relation to headphone measurements like damping. Instead of in relation to nothing. Or headphone measurement in relation to something. The graph comes remarkably close to the subjective opinions on witch headphones that are best suited for each genre. And can maybe even give insight to why Beats have 70% of the headphone marked, and I dont think its all down to marketing.

In relation to nothing the best measuring headhone will be classified as the best, but Im not convinced that the best measuring allways sound best.
 
Sep 15, 2013 at 5:14 PM Post #2 of 24
Edit. Headfi still wont let me upload. So here are direct links instead:

Maybe someone can upload them for me?

Oh. And sorry for my mad Paint skills, I dont have photoshop installed on this computer and I am highly skilled with PS. ;)

http://s22.postimg.org/re00wd5nl/oversiktheadfi7.jpg

http://s23.postimg.org/j4d68yobv/oversiktheadfienglish.jpg
 
Sep 17, 2013 at 11:46 AM Post #6 of 24
Maybe, I dont know. I was hoping someone here could clarify this in relation to dynamics in music. There is a difference between Geen Day on -6dB and Tchaikovsky on -36dB.

2,5mm is the same no matter the driver size.
 
Sep 17, 2013 at 11:59 AM Post #7 of 24
2,5mm is the same no matter the driver size.

 
A larger driver needs less excursion to move the same amount of air. In practice, the LCD-2 seems to be able to play a 20 Hz tone at 100 dB SPL with less than 0.1% distortion, something that few dynamic headphones can match.
 
Sep 17, 2013 at 12:18 PM Post #8 of 24
Yes. But the frequency response sweeps dont take into account that the same orthodynamic and electrostatic driver has to produce tones at 30 to 10khz at the same time in real music. Probably adding a bit of distortion in the process.
 
Sep 17, 2013 at 12:37 PM Post #9 of 24
Yes. But the frequency response sweeps dont take into account that the same orthodynamic and electrostatic driver has to produce tones at 30 to 10khz at the same time in real music. Probably adding a bit of distortion in the process.

 
Low frequencies are the most difficult to reproduce at high SPL, because they require the most driver excursion. The InnerFidelity PDF files do not include distortion tests at higher than 100 dB SPL, though, so it is not known how loud the sound needs to be before the driver begins to distort significantly. But if the headphone can play a 100 dB sweep cleanly, it should also be able to play music with 100 dB peaks (which might not be loud enough for very "dynamic" music, however).
 
Sep 18, 2013 at 9:26 AM Post #10 of 24
Thats wierd. Distortion is less on SPL 100dB than on 90dB on both LCD-2 and HD800. With the exeption of 20-300hz range thats marginally worse on the HD800.

Do any of the measurements give indication of speed? If I interpret the square wave 30hz, and 300hz, and impulse response the HD800 covers a larger distance in a shorter ammount of time, compared to LCD-2 and Stax SR009.

If I interpret this right the HD800 driver is in fact more dynamic. It is faster, but have a larger overshoot on the rebound. Probably due to less mechanical damping. Stax and Audeze is faster to reset into a standstill position. Probably an indication of better damping ability.
 
Sep 18, 2013 at 9:36 AM Post #11 of 24
Thats wierd. Distortion is less on SPL 100dB than on 90dB on both LCD-2 and HD800. With the exeption of 20-300hz range thats marginally worse on the HD800.

 
That is because InnerFidelity graphs show THD+N, rather than THD. So, at the lower SPL, microphone and ambient noise will often show up as if it was distortion. For example, if there is 40 dB unweighted noise SPL (THD+N calculations are usually unweighted, and in that case 30-40 dB is quite possible), it will show up as 0.3% THD+N at 90 dB, even if there is actually no distortion at all. In reality, transducer distortion typically increases with excursion.
 
Quote:
Do any of the measurements give indication of speed?

 
For that, first we need to define accurately what "speed" is.
 
Sep 18, 2013 at 9:49 AM Post #12 of 24
That is because InnerFidelity graphs show THD+N, rather than THD. So, at the lower SPL, microphone and ambient noise will often show up as if it was distortion. For example, if there is 40 dB unweighted noise SPL (THD+N calculations are usually unweighted, and in that case 30-40 dB is quite possible), it will show up as 0.3% THD+N at 90 dB, even if there is actually no distortion at all. In reality, transducer distortion typically increases with excursion.

THD should increase in tact with higher SPL, so that it was opposite suprised me. But your explenation is logical.

For that, first we need to define accurately what "speed" is.


I would define it as the time it uses to reach peak point, and return back to starting position. Not to a standstill,

If it covers more ground, in a Shorter amount of time, I would Call it faster.

On a nother note. Why do headfi Automatic put large nouns infront of random Words. It is annoying as hell.
 
Sep 18, 2013 at 11:34 AM Post #13 of 24
If it covers more ground in a shorter amount of time, I would call it broken.

The time it takes to reach the peak point and return to the starting position is a function of frequency. If something did this "faster" or "slower," it would have to be producing a frequency other than the frequency it's being fed. In other words, broken.

Your notion of "speed" makes no sense.

se
 
Sep 18, 2013 at 12:13 PM Post #14 of 24
Take these two HD800 measurements.

A
http://www.innerfidelity.com/images/SennheiserHD800B.pdf

B
http://www.innerfidelity.com/images/SennheiserHD800.pdf

One has a faster IR rise and fall, but requires less voltage to do the work.

The other has a slower IR rise and fall, and does not travel as far as the first, and needs more voltage to do the work.

Whats the difference? Diaphragm weight, mechanical damping or something else? Witch is faster? I would say A is faster but less controlled. Witch is more precise? I would say B.
 

Users who are viewing this thread

Back
Top