[kiteki]

Hi,
 

 
 
 
This discussion started in this thread
 
 
The discussion revolved around acoustic measurement, digital measurement, the totality of data / total lack thereof, how data correlates to the listening experience, et cetera.
 
Using the Skullcandy MMM versus LCD-3 as a random example on acoustic data, I asked where the differences are.
 

 
 
You can easily equalize these deviations to elicit a 'golden FR' from the Skullcandy which turns it into the LCD-3, or is something else missing?
 
Apparently I was cherry picking data from flawed measurements, I was supposed to use innerfidelity or something.
 

 
 
 
So, now they look the same at 300Hz instead of 500Hz.
 
 
So can I easily 'golden FR' the Skullcandy into a $2000 headphone, or am I missing something here?  Apparently SWR isn't even important to measure.
 
Are the vital differences located in THD+N and IMD?  If that's true, then the Skullcandy is only a very slightly noiser, or slightly more distorted LCD-3?  I think this needs some explaining.
 
Novices tend to want to look at raw, unflinching data, rather than read glossy reviews subject to emotion.
 
 
What if the data is only 50% of the final sound spectra?  Then 70% human reviews are more accurate.
 
 
Is the lustrous musicality of the Denon Dx000 series in the FR?  Where?
 
Is the bass presence and extreme realism of the JH13 in the FR?  Is the featherlite bass, narrow soundspace, and white glistening highs of the RE272?
 
Is the harsh fire of the EX700 versus the liquid sky of the EX1000?
 
Sony MDR-V6 versus Tesla T1?

 

[kiteki]

Secondly, apparently it's not even necessary to measure SWR.
 

Originally Posted by ---- /img/forum/go_quote.gif  
You are going off topic...but..Square-wave responses are basically a fancy way of looking at the FR, check this out, I can derive any square-wave response based on the FR graph. Why would we want to measure that, FR are the result of what the driver is bringing in, in decibels. Those are all tuning aspects which will change FR response, two different things, looks like you're just trying to confuse the two together.

 
 
Needs clarification.  Especially for novices trying to make sense of all of these audio products online they can't listen to prior to purchase!

 

[Inks]

Novices tend to want to look at raw, unflinching data, rather than read glossy reviews subject to emotion.
So a budget headphone has similar peformance to a expensive one, no surprise there, don't think that discredits the graphs. Compare that Skullcandy to a HD800, no contest. SWR are a translation of the FR into a more concise matter, not changing anything. 
 
What if the data is only 50% of the final sound spectra?  Then 70% human reviews are more accurate.
What ifs, are just that. Unless you can provide and prove there's something significant that can't be shown, it's all speculation and wishful thinking. Our sound perceptions are a personal matter specially when there's no guidelines, translating that to a standard to which others should use for performance is very problematic. Graphing and using an average HRTF of the average human is the best approach, only a few graph this way though, Headroom and IF aren't it. 
 
Is the lustrous musicality of the Denon Dx000 series in the FR?  Where?
Define musicality first, that's a broad, personal useless term, that's the issue there. 
 
Is the bass presence and extreme realism of the JH13 in the FR?  Is the featherlite bass, narrow soundspace, and white glistening highs of the RE272?. JH13 do have boosted bass, realism? that's subjective unless you mean real to the recording, flat that can be measured. Don't know what you're getting at with all those comparisons, RE272 does graph compared to what it's said about it. 

 

[xnor]

Quote:

Originally Posted by kiteki /img/forum/go_quote.gif
 
You can easily equalize these deviations to elicit a 'golden FR' from the Skullcandy which turns it into the LCD-3, or is something else missing?

I wouldn't say that the LCD's have a 'golden FR'. Of course you could try to boost lower bass of the Skullcandy's by at least 10 dB, and flatten the upper mids, and use a couple of filters to copy the treble response (not that easy). Btw, those HR graphs are heavily smoothed. Take a look at the raw data to make the differences more obvious.
 
 
Quote:

Apparently I was cherry picking data from flawed measurements, I was supposed to use innerfidelity or something.
 

 
So, now they look the same at 300Hz instead of 500Hz.

How do you define 'the same'? Channel imbalance on the Skullcandy one, no?
 
 
Quote:

So can I easily 'golden FR' the Skullcandy into a $2000 headphone, or am I missing something here?  Apparently SWR isn't even important to measure.
 
Are the vital differences located in THD+N and IMD?  If that's true, then the Skullcandy is only a very slightly noiser, or slightly more distorted LCD-3?  I think this needs some explaining.
 
Novices tend to want to look at raw, unflinching data, rather than read glossy reviews subject to emotion.

Didn't Tyll review the Mix Master as a very good sounding headphone? So I don't see why it shouldn't compete with more expensive, also very good sounding headphones.
The square wave response is indeed redundant, but still a nice to have. (Those measurements cost those who make them a lot of time and money and we get them for free so I'm happy to see any kind of them.)
Sadly, we won't have IMD data available (I've suggested to plot the spectrum of a 100, 1k and 10k Hz sine wave playing simultaneously but nobody seems interested).
 
 
Quote:

Is the lustrous musicality of the Denon Dx000 series in the FR?  Where?

I don't even know what that means. Show it to me in the data ... oh wait ...
 
 
Quote:

Is the bass presence and extreme realism of the JH13 in the FR?  Is the featherlite bass, narrow soundspace, and white glistening highs of the RE272?

The JH13 has really boosted lower bass and doesn't seem to be diffuse field equalized. Maybe that makes it more realistic as an IEM?
 
 
Quote:

Is the harsh fire of the EX700 versus the liquid sky of the EX1000?

Seriously, how do you come up with these audiophile expressions? I'm not an in-ear guy but the EX1000 seems to have a problem with resonances. Maybe that smears the treble details into what someone would call 'liquid sky'?
 
 
Quote:

Sony MDR-V6 versus Tesla T1?

What about them? They sound different, measure different, look different.
 
 
edit: On a sidenote, what I don't get is how even tiny (< 0.5 dB) deviations from flat FR in amp and DAC measurements make people nearly freak out while > 5 dB differences in headphones are constantly labelled 'the same' or with 'no difference'.

 

[anetode]

Numbers may be a dangerous thing if you don't know how to use them.

A Mixmaster is a closed-back moving coil circumaural headphone with decent isolation and leather-covered foam earpads, whereas an LCD-3 is an open-backed isodynamic planar circumaural with tilted leather-covered earpads providing for a non-parallel angle of the transducer to the ear as well as a larger cup volume and different clamping pressure. Unless you have this information about the measured headphones, know how and why each of these factors affects the sound and you are familiar with the methodology of the tester you will be limited in how much you can glean from comparative measurements.

 

[kiteki]

Numbers may be a dangerous thing if you don't know how to use them.

A Mixmaster is a closed-back moving coil circumaural headphone with decent isolation and leather-covered foam earpads, whereas an LCD-3 is an open-backed isodynamic planar circumaural with tilted leather-covered earpads providing for a non-parallel angle of the transducer to the ear as well as a larger cup volume and different clamping pressure. Unless you have this information about the measured headphones, know how and why each of these factors affects the sound and you are familiar with the methodology of the tester you will be limited in how much you can glean from comparative measurements.

 
Exactly 100%
 
This information isn't present in the measurements, this is not a "what if" situation like Inks said, it's reality, this is the primary point I'm addressing.  The numbers are often represented in a sense of totality, which I find cynical or self-interested.
 
Unless there is a code in the numbers which always reveals if it's an isodynamic planar, or a moving coil, then the data is lacking, since humans can identify those two differences very quickly via listening.
 
 
As for the very different bass presence in the RE272 versus JH13 I can't find it in the FR.
 
http://www.innerfidelity.com/images/HiFiMANRE272.pdf
 

 
 
Then there is always this ...
 

 

[kiteki]

Quote:

Quote:

Originally Posted by kiteki /img/forum/go_quote.gif
 
You can easily equalize these deviations to elicit a 'golden FR' from the Skullcandy which turns it into the LCD-3, or is something else missing?

I wouldn't say that the LCD's have a 'golden FR'. Of course you could try to boost lower bass of the Skullcandy's by at least 10 dB, and flatten the upper mids, and use a couple of filters to copy the treble response (not that easy).

 
Sounds like an interesting experiment!  If I had some kind of frequency response or convolver to use in Foobar2000 on the Skullcandy MMM I'd probably buy it and then demo the LCD-3 with my laptop and MMM.
 
 

 

Is the bass presence and extreme realism of the JH13 in the FR? Is the featherlite bass, narrow soundspace, and white glistening highs of the RE272?

The JH13 has really boosted lower bass and doesn't seem to be diffuse field equalized. Maybe that makes it more realistic as an IEM?

 
Please explain.  Should I listen to live recordings, or synthetic studio music on the JH13 demo unit?
 
 
 

 

Is the lustrous musicality of the Denon Dx000 series in the FR? Where?

I don't even know what that means. Show it to me in the data ... oh wait ...

 
Lol.
 
 

Originally Posted by xnor /img/forum/go_quote.gif
 

Is the harsh fire of the EX700 versus the liquid sky of the EX1000?

 
Seriously, how do you come up with these audiophile expressions? I'm not an in-ear guy but the EX1000 seems to have a problem with resonances. Maybe that smears the treble details into what someone would call 'liquid sky'?

 
The EX700 is a bit deafness inducing with extensive listening, like when you've been to a loud concert and your ears are protecting themselves after.  On a 10 hour plane-trip listening to the EX700 my ears were not very happy.  This doesn't seem to happen with the EX600 which is more neutral and refreshing.  The EX1000 has some kind of effect which turns glass into liquid, sorry, I just use the pictures which come to my mind. =p
 
On a sidenote, I bought a second brand new EX700 and it sounds the same as my previous one (I was the third owner and applied burn-in), so, as far as I'm concerned burn-in doesn't exist, at least in this case. 
 
 

Originally Posted by xnor /img/forum/go_quote.gif
 

Sony MDR-V6 versus Tesla T1?

What about them? They sound different, measure different, look different.
 
edit: On a sidenote, what I don't get is how even tiny (< 0.5 dB) deviations from flat FR in amp and DAC measurements make people nearly freak out while > 5 dB differences in headphones are constantly labelled 'the same' or with 'no difference'.

 
Lol x2.
 
The whole extreme flat FR and 1 ohm output impedance is a purity / reference thinking.  What I don't get is... nevermind. =p

 

[xnor]

Quote:

Exactly 100%
 
This information isn't present in the measurements, this is not a "what if" situation like Inks said, it's reality, this is the primary point I'm addressing.  The numbers are often represented in a sense of totality, which I find cynical or self-interested.
 
Unless there is a code in the numbers which always reveals if it's an isodynamic planar, or a moving coil, then the data is lacking, since humans can identify those two differences very quickly via listening.

So you expect to see exactly what driver was used, what materials were used, what the shape and dimensions of the ear cup and pads is etc. from some measurements of the final result (sound pressure)? That's absurd.
Similarly, you don't know what engine or suspension is in a car by looking at the BHP or top speed.
 
But so what, that doesn't invalidate these measurements nor does it make them incomplete for what they are. The measurements are what they are and they tell what they're supposed to tell, not something else.
 
Now if you don't want to look at the data because it doesn't show you what you're looking for, that's fine. For example, we don't have measurements for comfort (ok, there is clamping force, but this is seldomly provided and there's still the question how soft the earpads are ..) so reviews can be quite helpful to fill in (subjective) data that we don't have measurements for (yet).
 
Another thing that I question is the last part of the quote above.

 

[kiteki]

Originally Posted by xnor /img/forum/go_quote.gif
 
So you expect to see exactly what driver was used, what materials were used, what the shape and dimensions of the ear cup and pads is etc. from some measurements of the final result (sound pressure)? That's absurd.
 
Similarly, you don't know what engine or suspension is in a car by looking at the BHP or top speed.
 
But so what, that doesn't invalidate these measurements nor does it make them incomplete for what they are. The measurements are what they are and they tell what they're supposed to tell, not something else.
 
Now if you don't want to look at the data because it doesn't show you what you're looking for, that's fine. For example, we don't have measurements for comfort (ok, there is clamping force, but this is seldomly provided and there's still the question how soft the earpads are ..) so reviews can be quite helpful to fill in (subjective) data that we don't have measurements for (yet).
 
Another thing that I question is the last part of the quote above.

 
No, I agree with everything in your latest post here!
 
The measurements tell what they tell, they are free, and useful if applied to exactly what they tell, so I'm not invalidating them, I'm only invalidating the perception that they tell everything, since they're very often referenced in this context!
 
It's an exacting science to know - or pursue - what they don't tell, however this is usually whisked off.
 

---

 
It's valid to question if humans can tell apart isodynamic planar versus moving coil versus balanced armature et cetera, since some can't.
 
There's lots of cases where people hear things which don't exist, or make up complete nonsense, however that doesn't invalidate the real perceptivity of differences.
 
I usually just look at statistics, if 1000+ people can hear a specific difference the statistics of chance are pretty thin, et cetera. =)

 

[Head Injury]

Quote:

 
No, I agree with everything in your latest post here!
 
The measurements tell what they tell, they are free, and useful if applied to exactly what they tell, so I'm not invalidating them, I'm only invalidating the perception that they tell everything, since they're very often referenced in this context!
 
It's an exacting science to know - or pursue - what they don't tell, however this is usually whisked off.

I hope no one's ever claimed that THD and frequency response tell everything there is to know about a headphone. They would be very wrong.
 
I also hope you're not trying to use this argument to suggest that measurements don't tell everything about amps or DACs. That would be a red herring.

 

[kiteki]

Originally Posted by Head Injury /img/forum/go_quote.gif
 
I hope no one's ever claimed that THD and frequency response tell everything there is to know about a headphone. They would be very wrong.

 
They do, sometimes only FR, other times CSD, IR, IMD and THD+N.
 
 

Originally Posted by Head Injury /img/forum/go_quote.gif
 
I also hope you're not trying to use this argument to suggest that measurements don't tell everything about amps or DACs. That would be a red herring.

 
Nope, acoustic and digital are very different.  In theory if we had 100% perfect speakers in anechoic chambers, or 100% perfect IEM's, I think the imperfections of digital (ADC, DAC, amplifiers etc.) would be fairly minimal, I think they'd be pretty close to perfect sound. 

 

[xnor]

But sometimes FR is enough: imagine you want a neutral headphone and the FR shows a big bass boost - next one please, unless you intend to equalize it.

 

[kiteki]

Originally Posted by xnor /img/forum/go_quote.gif
 
But sometimes FR is enough: imagine you want a neutral headphone and the FR shows a big bass boost - next one please, unless you intend to equalize it.

 
Yes, sometimes.  As an example, I didn't like the bass boost in JVC FX500, FR confirmed it, and then I removed it with an acoustic equalizer. ;o)
 
http://www.head-fi.org/t/614513/jvc-fx500-yellow-filter-black-filter-and-cotton-removal-review#post_8459857

 

[xnor]

Quote:

 
Yes, sometimes.  As an example, I didn't like the bass boost in JVC FX500, FR confirmed it, and then I removed it with an acoustic equalizer. ;o)
 
http://www.head-fi.org/t/614513/jvc-fx500-yellow-filter-black-filter-and-cotton-removal-review#post_8459857

That's a very nice example. I really like mods, even if they just improve comfort (which often does also have an influence on sound).
I'd love to see measurements for all headphones pre and post "famous" mods. It wouldn't only be useful to confirm what people hear but also help create a better understanding of how mods change the sound, how we can use mechanical means to equalize headphones to one's liking within its limitations of course, reduce peaks in the FR due to resonances, how much damping material is too much and so on..

 

[Joe Bloggs]

Quote:

 
So can I easily 'golden FR' the Skullcandy into a $2000 headphone, or am I missing something here?

 
It's not 'easy', but equalization has turned my $10 Philips SHE3580 (by all accounts a pretty good pair of phones to start with) into THE reference headphones for me.  The $200 Etys don't sound worse, they just don't sound better, and they break on me all the time so I'm selling them.
 
http://www.head-fi.org/t/561951/philips-she3580-iem-review-10-giant-killers-er-4p-gathering-dust
 
These days whenever I run a new pair of phones through my full parametric equalization procedure and equalize them to my reference FR (which equalizes the phones at every point in the audible spectrum (rather than say a 10 band or even 30 band graphic equalizer), evening out especially all those resonance peaks in the treble that you can't measure with an acoustic coupler or even a dummy head, but only with your own ears), every pair of phones end up sounding 98% like my reference SHE3580.  It's like there's nowhere to go for me anymore from 'perfection'... and perfection can be had for $10 if you know what you're doing.
 
So my answer to you would be a definite 'yes', except for the 'easy' part.  It took me months of research to nail down my methodology.