[cel4145]

Over in the Introductions, Help and Recommendations, newbies are frequently citing having listened to headphone (and speaker) demo reviews on YouTube or recommending that others listen to them when deciding on which headphones to purchase, such as these by Z Reviews:


[VIDEO]https://www.youtube.com/watch?v=WGfeNapZvPY[/VIDEO]

[VIDEO]https://www.youtube.com/watch?v=4YtWGc7DgUU[/VIDEO]


I thought it would be helpful to have a thread that lays out the fallacy of using recordings of headphones or speakers (the transducers) to evaluate them as a resource that newbies could be pointed to when they champion this practice. Here are my thoughts:

1) Unless the transducers used to listen to the recording have a perceived neutral response, they will color the sound of the transducer recording being evaluated. I suppose if one really understood the frequency response of the headphones or speakers one is using to listen to the recordings, one might be able to make some judgement about how the recorded equipment would sound. But this seems highly unlikely with the newbies using this method. At best, assuming one uses videos produced by the same person, one could perhaps discover relative differences in bass, mids, or treble quality between two sets of recorded transducers, assuming that transducers used for listening to the recording have the necessary extension for the bass and the treble.

2) The recording itself may not be accurate, depending on the equipment used to do the recording and the setup.

3) Soundstage and imaging would seem difficult to accurately record, and of course the soundstage and imaging of the transducers used for the evaluation would affect how it is reproduced as well.

4) Obviously, if the transducers used for listening have less detail resolution than the transducers being evaluated, then one cannot evaluate that factor at all. For instance, the transient response of the transducers used for the evaluation could easily affect how the transient response of the transducers being evaluated are perceived.

***

I'm sure people will think of other factors as well. I will say that using this method of headphone and speaker evaluation, IMO, would likely provide more erroneous information than accurate information, making one more likely to make the wrong decision than one would make listening to the headphones or speakers in person.

 

[Head Injury]

1) Unless the transducers used to listen to the recording have a perceived neutral response, they will color the sound of the transducer recording being evaluated. I suppose if one really understood the frequency response of the headphones or speakers one is using to listen to the recordings, one might be able to make some judgement about how the recorded equipment would sound. But this seems highly unlikely with the newbies using this method. At best, assuming one uses videos produced by the same person, one could perhaps discover relative differences in bass, mids, or treble quality between two sets of recorded transducers, assuming that transducers used for listening to the recording have the necessary extension for the bass and the treble.

2) The recording itself may not be accurate, depending on the equipment used to do the recording and the setup.

3) Soundstage and imaging would seem difficult to accurately record, and of course the soundstage and imaging of the transducers used for the evaluation would affect how it is reproduced as well.

4) Obviously, if the transducers used for listening have less detail resolution than the transducers being evaluated, then one cannot evaluate that factor at all. For instance, the transient response of the transducers used for the evaluation could easily affect how the transient response of the transducers being evaluated are perceived.

Didn't watch your example videos, sorry.
 
1. Some of the videos I've seen (SonicSense reviews mostly) play the song right from the digital files, then switch to a recording of the song played through the headphones. This seems like the way to do it, because the transducer on the user's end will color both the original and recorded song equally. The closer the recording gets to the original, the more neutral the headphones are, regardless of the transducer used to play the YouTube video. The only problem then is the recording setup itself and how it colors the sound.
 
2. In the same situation as #1, #2 is not a problem if we're talking strictly neutrality and accuracy of reproduction. It can be when evaluating taste, of course.
 
3. For the most part soundstage and imaging are just functions of frequency response, channel balance, and other measurable phenomena. You can get at least an idea of it from videos like the ones I described in #1. Take this one. What I hear in it is the HD8 DJ compressing the soundstage slightly vs. the song file, the K267 expanding slightly, and the Beats...um. Anyway, this suggests the K267 will have a larger soundstage than the HD8 DJ.
 
4. Once again, in the case of videos made like I mention, you can get some idea of the detail lost vs. the original track and make a comparison that way.
 
No, you won't be hearing exactly what the headphone sounds like, but you can make comparisons to other headphones to get some idea what you like. They should not be the only reason you buy a headphone, but if done correctly they can be a useful tool. We know now, for example, that if you like to listen to music underwater, the Beats Pro is for you.

 

[cel4145]

Didn't watch your example videos, sorry.

1. Some of the videos I've seen (SonicSense reviews mostly) play the song right from the digital files, then switch to a recording of the song played through the headphones. This seems like the way to do it, because the transducer on the user's end will color both the original and recorded song equally. The closer the recording gets to the original, the more neutral the headphones are, regardless of the transducer used to play the YouTube video. The only problem then is the recording setup itself and how it colors the sound.

Still difficult to judge how a headphone will sound when it's colored by one's own headphones unless one's own headphones are perfectly neutral, even if comparing to the original digital file. Best you can get is more general relative differences--more or less bass, mids, treble. But it will always be "EQ'd" by the listeners transducer setup, so the sense of how the overall frequency response works for the headphones being evaluated will be quite different from reality. For instance, you could get compounding problems. If the listener's headphones are a bit warm, and then the headphones being evaluated are a bit warm but with slightly different frequency response characteristics, the resulting combination of the two could be overly warm. Or consider if one is listening with headphones with a spike at 8K to a recording of headphones with a spike at 10K. The combination of the two might make the recording sound overly bright, whereas the listener might be fine with the headphones being evaluated without the additional spike added.

3. For the most part soundstage and imaging are just functions of frequency response, channel balance, and other measurable phenomena.

I don't see how that's true. Otherwise, we'd have tons of closed headphones with soundstages like open headphones.

And if you listen to recordings of closed headphones with open headphones, it will exaggerate the soundstage and vice versa.

4. Once again, in the case of videos made like I mention, you can get some idea of the detail lost vs. the original track and make a comparison that way.

That's assuming ones listening setup is very good. The listener's transducers could hide differences in detail resolution between the original and the recording. Also, the impulse response could be a bit different and what you would be listening to is the sum of the impulse response of the recording and the listener's headphones, which could sound worse than the individual impulse response of either the recorded headphones or the listener's headphones.

No, you won't be hearing exactly what the headphone sounds like, but you can make comparisons to other headphones to get some idea what you like. They should not be the only reason you buy a headphone, but if done correctly they can be a useful tool. We know now, for example, that if you like to listen to music underwater, the Beats Pro is for you.

I still think that the potential for misinformation is so high that many people are more likely to walk away with the wrong impressions than the right ones, particularly newbies who are encouraged to do this that have poor or mediocre audio quality setups. Try evaluating two 4K TVs (or even two good 1080p TVs) by watching videos of them on a poorly calibrated, fairly inexpensive 720p TV and tell me you can discern the differences between the two good ones. You won't even be able to tell that the two 4K tv's produce a better picture than your own.

 

[Music Alchemist]

I remember watching a Sonic Sense video comparing the Audio-Technica ATH-M40x to other headphones and the recording itself. It was the only headphone that was indistinguishable from the recording in that video. However, when I bought them, I found that I disliked the sound and did not consider it to be anywhere close to accurate.
 
So yeah...going from those videos alone is not the best way to make buying decisions.