Resolution, speed, do these things really exist?

[192kbps]

Is It Possible for Headphones to Sound Fast? (Or Slow?)​

SoundStage! Solo | SoundStageSolo.com - Is It Possible for Headphones to Sound Fast? (Or Slow?)

Reviews of headphones, earphones, and headphone amplifiers

www.soundstagesolo.com

 

[kongwee]

Basically the driver able to move and stop at precise matter. Sometime speed can represent resolution in this aspect. Spectral Audio claim Mhz oscillation and Siltech claim Ghz transmission. Diamond Tweeter. Electrostatic driver, Ribbon driver. All about speed reveal details in one of the direction.

 

[audio2design]

Basically the driver able to move and stop at precise matter. Sometime speed can represent resolution in this aspect. Spectral Audio claim Mhz oscillation and Siltech claim Ghz transmission. Diamond Tweeter. Electrostatic driver, Ribbon driver. All about speed reveal details in one of the direction.

All those claims are nonsense. Frequency response and distortion at volume tell you speed and resolution. You need IMD to get a clearer picture of distortion. Waterfall closes out the measurement picture mainly.

 

[swamps]

When the typical audiophile says one headphone is "faster" than other they don't mean it literally, as said before if the driver can hit 20khz it is fast enough.
Now decay is both measurable and perceptible at extreme cases, if you take a headphone with a known resonance at say 10khz you can hear the extended ringing after the signal stops. How much it matters for actual music is another thing.
It really seems to boil down to frequency response, distortion, decay, on that order. Problem is there's no way to predict with high accuracy how each person's head will interact with the headphone and hence the weird peaks and mismatches that aren't in the frequency response plot and making my EQ process much more annoying.

 

[dc655321]

Many audiophiles and reviewers of higher-end audio products also seem to be able to separate a headphones FR performance from its technical performance are full of crap.

FIFY.

 

[tvrgeek]

As we may not have the final word on what parameters describe what we hear, we get terms like "fast" or "impact" which is totally meaningless but favorite terms of magazine writers or "soundstage" which has some FR component, some mechanical. ( Remember the Stax Nearphones?)

Decay, as measured in a CSD plot, can be informative if you understand what you are looking at. Plots can look terrible but actually show good behavior relative to possible. Darn physics. It takes force to stop things. It is not a favorite of reviewers as one can't put a number on it to say "bigger or smaller is better" It's a "it depends" measurement.

We can put a number on "area under the curve" for deviation from an assumed optimal response curve which may be a hint if a set of cans is useable without EQ. That would be most useful for portable use where no EQ is available. We still run the risk of not understand where deviation is more or less important. We can show bass roll off, but what tells us if the rest of the response has been modified away from "optimal" to make a balanced presentation? Maybe we want a little mid bass boost?

We can start with the help from the forum measurements to give a clue what we are hearing, but I suggest then using APO in 1/3 octave, Q 4.2 or something to tweak what sounds correct to each of us. Soprano voice for the upper mids. My experience is as simple a source as possible. Solo piano, solo bass, guitar, voice. We can get too confused by a very complex source. Once you have that dialed in, then bring in the fully orchestrated to see if your balance holds. Basically, if a female voice sounds natural, we can ignore a lot else.

 

[tvrgeek]

Basically the driver able to move and stop at precise matter. Sometime speed can represent resolution in this aspect. Spectral Audio claim Mhz oscillation and Siltech claim Ghz transmission. Diamond Tweeter. Electrostatic driver, Ribbon driver. All about speed reveal details in one of the direction.

Follow this advice and your legless reptiles will not make any annoying squeaky noises.

Seems there is this old "law of physics"
That which is motion shall remain in motion unless acted on by an outside force
That which is at rest shall remain at rest unless acted on by an outside force.

Well, we can apply whatever force necessary to get a driver moving. Unfortunately, there is a lag as our signal represents the desired motion and does not include the necessary force to overcome jerk. ( Jerk, acceleration, speed, Second derivatives) all transducers have mass. Damn physics again. We don't "stop" our transducer as much as need to control it's opposite direction with our signal. We would only need to stop a driver if we were producing a square wave. ( Impossible if there is mass involved. )

It is possible, with enough calculation and enough known about the transducer, to compensate for jerk by modifying the signal to produce the desired movement. No one has done that yet for a speaker I am aware of. If they ever do, we can get form the % distortion to where electronics are. That would be sweet. Maybe.

At least you mention two companies who are peddling such "speed" dribble, I know I don't need to look at their products. BTY, every actual objective test of ribbons I have seen shows them to have pretty bad distortion even if their CSD plot looks decent.

 

[ADU]

FIFY.

Merci (I think).

So basically not buyin any of it?

 

[ADU]

All those claims are nonsense. Frequency response and distortion at volume tell you speed and resolution. You need IMD to get a clearer picture of distortion. Waterfall closes out the measurement picture mainly.

So speed exists then, but is a function of FR and distortion?

 

[dc655321]

Merci (I think).

So basically not buyin any of it?

Not really, no
The bandwidth of a system (frequency response) determines its ability to reproduce input signals within that bandwidth.
If a speaker system has a flat, anechoic bandwidth from 20Hz to 20kHz, then input signals in that range can be reproduced too.

I will grant that headphones are a bit different animal.
Same principles apply though.

 

[audio2design]

Any speaker with flat output from 50-250Hz approx and low distortion will be perceived as "fast".

 

[ahofer]

It would be fun to make a custom DSP with knobs to turn up and down the attributes that engineers know to be correlated with some of these words. Have a knob that turns up “resolution” or “slam” (Well, really only down, assuming we start with linear and low distortion).

 

[ADU]

It would be fun to make a custom DSP with knobs to turn up and down the attributes that engineers know to be correlated with some of these words. Have a knob that turns up “resolution” or “slam” (Well, really only down, assuming we start with linear and low distortion).

Fwiw, I believe the author of this Stereophile article on "euphonic distortion" created a tool for simulating some different kinds of distortion, which I think can still be found here...

Freeware

www.audiosignal.co.uk

I'll have to re-read it again, but I believe the article was mainly about amplifier distortion. And I think the utility was called "AddDistortion". Though I never actually tried it.

After experimenting with it and different kinds of distortion, I think the author, Keith Howard, basically concluded that no distortion was probably best. But also left the door well open for more and better research on the subject. He also did a follow-up on the subject later that same year with some more details and info.

Whether there has been any similar research by Stereophile since then, I don't know.

 

[ADU]

One of the main guys looking into the subject of distortion and headphones is Steve Temme, I believe. And this is a live presentation on the subject that he gave with Sean (now Doctor) Olive back in 2015 for their paper on The Correlation between Distortion Audibility and Listener Preference in Headphones...

 

[ADU]

I also found this video (suggested by Resolve) to be a pretty good primer on masking and different kinds of distortion, for those folks like me who may still be tryin to get up to speed on this subject...

 

[kongwee]

All those claims are nonsense. Frequency response and distortion at volume tell you speed and resolution. You need IMD to get a clearer picture of distortion. Waterfall closes out the measurement picture mainly.

Even looking at graph, you are visually looking at the compress value in dB. One dB increase is exponential of 10. 85 dB you can calculated many time the absolute value from 1db. How many interval can plot in linear graph of audio. Looking at the graph is not a linear relationship. 85 to 86 dB difference is not the same as 64db to 65dB. 0.001% to 0.0001%,not a much difference? Depend which dB region you are looking at. In the real world, I will not discount those diamond tweeter, electrostatic, ribbon and 500KHz > amplifier to give realistic presentation. New Class D oscillating above 500kHz just like Class AB. Of course, this is only one aspect of resolving detail and resolution. Some other electrical and physical property to factor in. Waterfall graph, decay graph, impedance graph, distortion graph......etc. Other aspect can mask the effect of speed. I am not designer, beyond my understand to engineer a certain aspect of audio.

Square wave graph is one way to look at speed. "zero" hz at flat responsive, "infinite" hz on rising or falling curve

 

[AyeYoYoYO]

A current/momentary lack of easily explained, consensus quantifiability does not denote inexistence.

 

[cheapmessiah]

As we may not have the final word on what parameters describe what we hear, we get terms like "fast" or "impact" which is totally meaningless but favorite terms of magazine writers or "soundstage" which has some FR component, some mechanical. ( Remember the Stax Nearphones?)

Decay, as measured in a CSD plot, can be informative if you understand what you are looking at. Plots can look terrible but actually show good behavior relative to possible. Darn physics. It takes force to stop things. It is not a favorite of reviewers as one can't put a number on it to say "bigger or smaller is better" It's a "it depends" measurement.

We can put a number on "area under the curve" for deviation from an assumed optimal response curve which may be a hint if a set of cans is useable without EQ. That would be most useful for portable use where no EQ is available. We still run the risk of not understand where deviation is more or less important. We can show bass roll off, but what tells us if the rest of the response has been modified away from "optimal" to make a balanced presentation? Maybe we want a little mid bass boost?

We can start with the help from the forum measurements to give a clue what we are hearing, but I suggest then using APO in 1/3 octave, Q 4.2 or something to tweak what sounds correct to each of us. Soprano voice for the upper mids. My experience is as simple a source as possible. Solo piano, solo bass, guitar, voice. We can get too confused by a very complex source. Once you have that dialed in, then bring in the fully orchestrated to see if your balance holds. Basically, if a female voice sounds natural, we can ignore a lot else.

I always get amazed at how people and reviewrs talk about soundstage as something thats dependant on the gear to such extent that amplifiers and cables affect it "significantly" by their personal standards.

And I do understand, and believe, that driver positioning on headphones relative to the ears can more or less modify soundstage, after all soundstage is nothing more than time delay due to our ears position relative to the sound source and the time delay from the reflections relative to the that same sound source. But when that "soundstage" concept gets used to rate an amplifiers, or even worse to compare TOTL amplifiers between each other, or even TOTL DACs, I cant help but get both confused and laugh at the extent some people will get to try to give some esoterical "aura" to relatively simpe functioning and well understood devices.

And ofcourse, you cant have soundstage unless the recording was engineered to transmit such quality, but my concern was directed towards the reproduction side of things.

 

[AyeYoYoYO]

I always get amazed at how people and reviewrs talk about soundstage as something thats dependant on the gear to such extent that amplifiers and cables affect it "significantly" by their personal standards.

And I do understand, and believe, that driver positioning on headphones relative to the ears can more or less modify soundstage, after all soundstage is nothing more than time delay due to our ears position relative to the sound source and the time delay from the reflections relative to the that same sound source. But when that "soundstage" concept gets used to rate an amplifiers, or even worse to compare TOTL amplifiers between each other, or even TOTL DACs, I cant help but get both confused and laugh at the extent some people will get to try to give some esoterical "aura" to relatively simpe functioning and well understood devices.

And ofcourse, you cant have soundstage unless the recording was engineered to transmit such quality, but my concern was directed towards the reproduction side of things.

There are a small number of headphones, many from AKG, which can even stretch&pull some impression of soundstage out of recordings which simply didn’t have much there to begin with. For these reasons, I sometimes feel the K5/K6/K7/K8 lines should receive an extra adjective modifier to the phrase “professional reference monitor”, perhaps there are more comprehensive words which encompass all the words like “archival”, “remastering”, “restoration-grade”, etc LBS

 

[ADU]

I always get amazed at how people and reviewrs talk about soundstage as something thats dependant on the gear to such extent that amplifiers and cables affect it "significantly" by their personal standards.

And I do understand, and believe, that driver positioning on headphones relative to the ears can more or less modify soundstage, after all soundstage is nothing more than time delay due to our ears position relative to the sound source and the time delay from the reflections relative to the that same sound source. But when that "soundstage" concept gets used to rate an amplifiers, or even worse to compare TOTL amplifiers between each other, or even TOTL DACs, I cant help but get both confused and laugh at the extent some people will get to try to give some esoterical "aura" to relatively simpe functioning and well understood devices.

And ofcourse, you cant have soundstage unless the recording was engineered to transmit such quality, but my concern was directed towards the reproduction side of things.

This is goin slightly off the OP's original topic, but what I really don't understand is why more people/entities/magazines/reviewers/etc. aren't making the effort to do more in-ear measurements of actual speakers, so that they (and we) can see a little better what's actually going on at the DRP/eardrum in terms of all of this stuff.... FR, distortion, time, etc..

I think that kind of "in situ" information would just be invaluable to the headphone community. And I know that some audio and headphone manufacturers are doing this behind close doors. And probably learning alot from it that the rest of us in the public sphere don't know yet. And are likely applying it in the development of their own products. I'm not just talkin about some of the smaller guys, like the DCAs, HFMs, Audezes, Focals, ZMFs, and folks like that (some of whom may simply not have the necessary resources for this kind of testing). But also many of the bigger headphone manufacturers, including Sony, Apple, Sennheiser, AudioTechnica, Samsung/Harman, etc.. And probably also some of the other larger Chinese manufacturers that are trying to make some more in-roads into these markets.

I do get the sense that there is some resistance to this kind of thing though, both from some reviewers and also some audiophiles. And I guess I can understand some of that. Because it takes a little bit of the mystery and fun out of just trying out different devices. And having large collections of headphones, each with their own unique sound... It is a bit of a mystery to me why there isn't more research being done in this specific area though, which is out in the open for all to see and appreciate.

If I was a headphone reviewer, and had a little $$ to work with to set up a good measurement system, then I would try to get or put together one that included a well-designed anthropomorphic mannikin, so that I could do both some headphone and also in-ear speaker measurements with it. It just makes sense to me to have all of that information together in one place.