[ev13wt]

Venturing out of the Science Forum for a bit here, to let you all see the latest article. Of course this article does not talk about the finer things like ergonomics, cup resonance or riging, but how important is stuff like that anyhow?

http://www.audioholics.com/loudspeaker-design/price-sound-quality-headphones

If you are a normal person, and not one of the few audiophiles that can actually hear -30dB 10KHz ringing through the signal, this means that a good software EQ can basically make your Senny 800 sound like a Beyer with peak - if you want. Or if you remove the 8600 peak from a 770s, you get a pretty darn "flat" response.

What do you think? Does a 2000 vs. 200 USD headphone really bring that "huge" improvement to the table?

What is a "huge" improvement? It seems its not the frequency response.

 

[emrelights1973]

Then we are all screw**!

 

[youngarthur]

I first used headphones, when I was 12 years old. I am now 77. I have HEK/HD800/iSINE 20, and many others. Recently, I have been buying Chinese IEM, and am very, very surprised at the quality.They range from $11, up to the heady heights of $39!!.I had 846/W60 earphones, but sold both on hearing these Chinese earphones. I am sure this seems Stupid/foolhardy/insane, but FOR ME makes sense. I also have Beats X, and Solo 3. I don't like/dislike beats. I have a completely open mind on the subject, and buy earphones/headphones that I enjoy.I certainly will think very carefully, before spending over $100 on equipment. I can appreciate the differences between iSINE 20, and $40 Chinese earphones, but is the difference in sound worth the extra?. I won't comment on the bass/mid/treble/soundstage etc, because of my old ears.Your money, your choice.

 

[ev13wt]

youngauthor makes a great point: We all tend to fret about the tiniest differences when we should be more into listening to the music.

 

[Malfunkt]

I’ve seen this study crop up before, and without getting into too much detail I feel there may be flaws in interpreting the results of the study. Perhaps I’m an outlier, but in my experience, an HD650 (a top of its class dynamic driver reference) cannot be made to sound like say an LCD2 merely through earth. Simply, the hd650s bass extension cannot reach the same depth as the LCD2 and additionally the clarity is higher from the planar driver.

Another example, eq alone can’t make the hd600 sound like the hd700 or hd800 as there are unique properties from each headphones enclosure that allow them to present as more spacious.

The inverse is also true for the above examples. An LCD2 cannot be made to sound like and hd650. Hd700 can’t be made to sound like a 650. Close in frequency, yes, but there are other characteristics that will change the spectral delay of certain frequencies and cause resonances.


I guess it is how one determines the data and I would probably agree with the study for the most part that in general, equalization of the headphone is a major determining factor. But there may be flaws even in how this was determined. I mean if you listen to electronic music you may want a more bass specific signature versus a headphone for classical that is airy and detailed. These preferences can’t be averaged neatly.

I do feel that equalization can do quite a bit, but for most headphone enthusiasts we are also listening for those other technicalities and unique sonic characters.

Just like each human voice has its own formant signature, so do headphones. We can talk about averages and ideals, but we also enjoy diversity. I do feel the next big steps in headphones will involve dsp that is tailored to our hearing to help reproduce lifelike sound and mimic certain listening environments.

 

[ev13wt]

Good post, thanks. Of course it is not possible to boost drivers that simply can't do "those" frequencies. You are boosting into "nothing" using a ton of power. Not good. Never boost, only cut. On the other hand, you can make an LCD-2 sound like a 650!

The thing is - lets say we record my moms voice. Its pretty much irrelevant how we play the voice back, I will recognize "mom". If I can recognize my mom through a narrow frequency band, with a ton if noise, with a lossy compression played back on a 5mm driver. That said, the best recording and playback system in the world can only "represent an image" of my moms voice. If the above "phone" example represents 10% of real, a "cheaper" headphone gets to 80% of real, and a 4K headphone gets me to 85% of "real" - would I still buy the Fostex? Well... probably yes.

In the end though, I am only buying a prettier version of "suspending disbelief". I will never get to "the real thing".


*values pulled out of my a$$"

 

[Malfunkt]

Can you make your lcd/ sound like an hd650, not entirely. You could match frequencies but there are still differences in the physical characteristics that can lead to changes in the sound spectra. I talked about spectral delay, and that is a noticeable one in which essentially, different frequency ranges will have varying amounts of decay in the time domain. These can be seen on spectral plots.

Once that decay is there it can’t be reduced. Similar to adding a reverb to a sound sample, it can’t be easily removed using current technology - especially real-time.

It’s a major aspect of sound and i don’t know if this study took this into consideration.

 

[ev13wt]

Can you make your lcd/ sound like an hd650, not entirely. You could match frequencies but there are still differences in the physical characteristics that can lead to changes in the sound spectra. I talked about spectral delay, and that is a noticeable one in which essentially, different frequency ranges will have varying amounts of decay in the time domain. These can be seen on spectral plots.

Once that decay is there it can’t be reduced. Similar to adding a reverb to a sound sample, it can’t be easily removed using current technology - especially real-time.

It’s a major aspect of sound and i don’t know if this study took this into consideration.

It probably didn't, but the effects of this are so much tinier than, lets say a room and speakers, that I feel it is almost irrelevant in comparing headphones. Can one hear it? Yes. In headphones, what is relevant? There is basically no reverbation in relevant frequencies, certainly not higher than 5ms?

Different maybe, but "better"? Not sure about that one. Its not an untreated room. People say: "Use headphones to take the room out of sound reproduction!"

 

[Audioholic123]

What is a "huge" improvement? It seems its not the frequency response.

The guy that wrote that Audioholics article seems to be using the art of measuring Headphone Frequency Response as an argument against expensive headphones. HFR is not the be-all and end-all of sound quality.

If I were to write an article like that, i'd have focused on driver technology.

 

[Malfunkt]

It probably didn't, but the effects of this are so much tinier than, lets say a room and speakers, that I feel it is almost irrelevant in comparing headphones. Can one hear it? Yes. In headphones, what is relevant? There is basically no reverbation in relevant frequencies, certainly not higher than 5ms?

Different maybe, but "better"? Not sure about that one. Its not an untreated room. People say: "Use headphones to take the room out of sound reproduction!"

Going back to the title of the article "No correlation between price and sound quality". I think saying there is absolutely no correlation is too far, and this is just hedging solely on the aspect that cheap headphones can be EQ corrected. However, as I've given example, I don't even think this is possible for the most part. I've tried many cheap headphones, and they can't be eq'd to get the performance of say an LCD2. They can't increase their clarity, they can't extend as far into the low-end and the top-end. For the most part.

I think the study can be misleading to the layman, thinking they can just EQ a pair of Audio Technica M50x and come out with an HD800, it just isn't going to happen.

Also, I can pick an average of people, and they may not be able to discern all the differences or care to (the latter is probably more true).

Personally, I pretty much reached end game with my LCD2 (rev 1) and they respond well to EQ. But try as a I may, EQ to the LCD2 won't bring the holographic sound of the HD700, as so much of that effect is from the interaction of the HD700 driver within it seashell resonant-like enclosure. A DSP may be able to simulate this type of effect however, something like Atmos or perhaps the most mature version of Dolby.

@ev13wt have you used Sonarworks? It has EQ profiles. I usually don't bother with EQ as I find it easy to just accept whatever I'm listening to and fill in the blanks.

And honestly, that last bit of statement is an important realization. I was fortunate enough to get all my gear pretty cheap, and having heard 2-channel systems more expensive than my condo, I don't really feel I'm lacking anything (even though those B&W 802s were revelatory) and I'm sure there is someone out there just loving their HD558 or Beats - and for the most part they are close enough within the general ballpark of being able to enjoy music.

 

[ubs28]

Cool. Now I need to find a pair of $50 headphone and then they will sound like the Orpheus 2 with EQ applied.

 

[FastAndClean]

I find that with proper EQ you can improve greatly the sound of the headphones if they have good enough driver to keep the control without adding distortions, planars usually have lower distortions than dynamic drivers and they are my choice at the moment

 

[FastAndClean]

i had great results before with beyerdynamic dt150, i added velour pads that made the bass flater, the soundstage was bigger, and the overall sound was very clean and tight, very high level of performance for the price

 

[mtoc]

yeah, study is more tell-dy than using own ears

 

[Sefelt103]

An interesting article. However I don't really believe there is no correlation between sound quality and price in headphones. I would argue that there is an arbitrary correlation between the two. As to whether there is no correlation between frequency response and price I would tend to agree with this statement. The article contends that the frequency response is "the dominant factor that determines sound quality". I would disagree with this. The frequency response is but one of a number of factors relating to sound quality. The design/type of the headphone, physical dimensions of the headphone, driver type/location, acoustics of the enclosure, materials, earpads or reflections will all significantly affect sound quality. Comparing Sennheisers HD650 frequency response with Beyerdynamics DT880 at Headroom shows two very similar frequency response curves and yet these headphones have little in common. The graphs tell you nothing about the thicker bassy nature of the HD650 or of its much more colourised sound or the difference in soundstage. I find frequency response to be a poor indicator of the sound of a headphone, there are just too many other factors involved. The article also talks about the chimera of making a Sennheiser HD518 into sounding like a more expensive model. Using equalisation or DSP (digital sound processing) to turn one headphone into another has been a dream since low distortion equalisation became a possibility. There are still many difficulties to face in this regard. One is if the low born transducer can't match the speed and accuracy of the exotic transducer. Another major problem will also be distortion. The signal must be altered to create the bass response of a larger driver, altered to create the soundscape of an angled driver, altered with further equalisation to match the acoustics etc. All these alterations inevitably lead to distortion of which there is very little of in some high grade headphones. If the headphone to be simulated also has a colourised sound then that's even more distortion. To turn the sound waves of a HD518 into a HE90 would be quite an achievement. I don't think high end headphone manufacturers will be too worried about these studies especially when so many audiophiles are of the subjective type. There was another thread on here about a $2 software DAC being indistinguishable from a hardware one costing $2000. You do pay for sound quality but in some ways it's almost free.