How to do a proper headphone review on Youtube (Oluv's Gadgets)

[Thomas_A]

Its pretty pointless anyway. The point we were going on then to dicuss is whether the changes in cans during the demos was audible - more than did the demos accurately reflect the sound of the cans . Once chap swears he cant tell a difference when Oluv changes the headphones during the demos. Everyone else says they can. Is this proof? Not really, but its too hot to argue.

During listening and comparing the original file with the cans, there is quite a resemblance of the original and for some headphones. Those ones also tend to measure quite neutral, like HD-600 and BD DT-250. The same goes if you have measured excessive highs; it is not difficult to hear this.

 

[Jimbob54]

I don't care about about even a billion people claiming to hear a sound difference, when there's no factual evidence to back it up.

I used a spectrum analyser myself, and it shows that there's a 4khz peak in the Verum. This confirms the tracks are not one and the same, but still who is to say he didn't mess with the EQ settings on the original track to give a perception that the headphones produce different sounds on the demo?

I mean this is a freaking science site, y'all should know your ears are useless and measured data is the only way to describe music with a certainty.

Absolutely nobody. Its not a good way to select headphones. Marginally better than no audition at all.

But then there is no point in reading any reviews , measurements or not, because you dont trust the reviewers not to mess.

Its just a possibility, but maybe the world isnt out to trick you, over and above standard marketing shenanigans and audiophoolery.

 

[amirm]

I don't know about the original track, but I seem to remember that he processes the audio coming from the headphones to make it more like the headphones actually sound to him.

I believe it was in this video where he was critiquing Zeos's headphone demo recording setup

It is worth watching the video if you want to understand his process. Of course it's about a year old, so it could have changed some in the time since.

Very well done video. But I don't get how Sonarworks correction gets into the picture. People listening online don't have that correction so they would not hear it the same way.

 

[richard12511]

I don't care about about even a billion people claiming to hear a sound difference, when there's no factual evidence to back it up.

I used a spectrum analyser myself, and it shows that there's a 4khz peak in the Verum. This confirms the tracks are not one and the same, but still who is to say he didn't mess with the EQ settings on the original track to give a perception that the headphones produce different sounds on the demo?

I mean this is a freaking science site, y'all should know your ears are useless and measured data is the only way to describe music with a certainty.

I don't think anyone was saying that those demos are a good representation of how those cans will sound. He very well may have altered the sound. Who knows.

We were just wondering why you kept claiming you couldn't hear any difference at all between the demo tracks when differences were obvious. We had to assume you were either somewhat deaf, or you were being dishonest. If it's the latter, the question is why? What do you have to gain by pretending to not be able to hear obvious differences?

 

[Jimbob54]

I mean this is a freaking science site, y'all should know your ears are useless and measured data is the only way to describe music with a certainty.

Not sure about that. The best way to determine if audio electronic components stay faithful to the source recording is measurements with an analyser.

Not sure measurements of transducers are the best way to describe "music"- pretty sure thats what hi fi systems do, pretty well. They represent the music the artist made and the producer/ engineer/ masterer recorded onto disc etc etc.

 

[raistlin65]

And are you listening to super flat dacs amps and headphones to make sure what he recorded is what your hear ?

Why, as an objectivist would you make the assumption that I use online headphone demos? Is that what you used to do before you learned better? I just offered that video in case you wanted to learn more about his process.

I've been a harsh critic of headphone demos ever since 2015 when Zeos posted the demo of the Brainwavz HM5 using this setup. Looks like a homemade foam box with a digital recorder stuck in it. lmao

 

[100rounddrum]

I don't think anyone was saying that those demos are a good representation of how those cans will sound. He very well may have altered the sound. Who knows.

We were just wondering why you kept claiming you couldn't hear any difference at all between the demo tracks when differences were obvious. We had to assume you were either somewhat deaf, or you were being dishonest. If it's the latter, the question is why? What do you have to gain by pretending to not be able to hear obvious differences?

Well because I couldn't hear any difference, before I used the spectrum analyser.

I mean, that's why all these measuring applications and devices exist right? Because of foolery of one's ears.

Yeah 5 guys are able to hear a difference, but the next guy may hear absolutely no difference. This makes the claim that there is a difference between the different headphones demo'd factually useless.

Ofcours, after I've used the spectrum analyser, I've started to hear sound differences relative to what I saw on the analyser.

And I've been to the audiologist a week ago, he tested my hearing as perfectly fine and able to hear up to 16khz.

 

[Jimbob54]

Well because I couldn't hear any difference, before I used the spectrum analyser.

I mean, that's why all these measuring applications and devices exist right? Because of foolery of one's ears.

Yeah 5 guys are able to hear a difference, but the next guy may hear absolutely no difference. This makes the claim that there is a difference between the different headphones demo'd factually useless.

Ofcours, after I've used the spectrum analyser, I've started to hear sound differences relative to what I saw on the analyser.

And I've been to the audiologist a week ago, he tested my hearing as perfectly fine and able to hear up to 16khz.

Healthy scepticism is good , and you're right that 5 people saying white is black doesnt make the guy saying its white wrong. But dont be the guy with his eyes shut saying the sun hasn't risen today. I've done those popular online tests of different quality copies of the same track, trying to distinguish low bitrate from lossless and things inbetween.

I do badly when others claim high "success" rates. They may be fibbing, but I dont spend my time worrying that the test is fake.

 

[bobbooo]

I don't know about the original track, but I seem to remember that he processes the audio coming from the headphones to make it more like the headphones actually sound to him.

I believe it was in this video where he was critiquing Zeos's headphone demo recording setup

It is worth watching the video if you want to understand his process. Of course it's about a year old, so it could have changed some in the time since.

From that video it looks like he's using the same compensation curve for every headphone. This won't work for the miniDSP EARS, because it does not have the same acoustic impedance as the human ear, which results in the correct compensation curve needing to be different for every headphone. Professional acoustic engineer Oratory, who's specific job it is to measure headphones, explains all this with measurements in this excellent post. This all means Oluv's recordings aren't really of much use.

 

[Erik]

here is the red target curve that Harman found beeing equal to flat loudspeakers, which is NOT the Harman curve.
it was created in the process of creating the first Harman curve (which is not the red one).
in the graph one can see the big diference to the 2017 Harman curve versions

here is the target btw for using in REW for example: https://github.com/jaakkopasanen/AutoEq/blob/master/compensation/loudspeaker_in-room_flat_2013.csv

The red curve is a loudspeaker equalized to flat in-room frequency response.
This is not how anechoically flat loudspeakers measure when placed in a room.

 

[raistlin65]

From that video it looks like he's using the same compensation curve for every headphone. This won't work for the miniDSP EARS, because it does not have the same acoustic impedance as the human ear, which results in the correct compensation curve needing to be different for every headphone. Professional acoustic engineer Oratory, who's specific job it is to measure headphones, explains all this with measurements in this excellent post. This all means Oluv's recordings aren't really of much use.

I don't know. It's been a long time since I've watched that video.

But I do seem to remember that he doesn't claim that what he's doing is accurate. Rather his claim seems to be that the process he uses results in recordings which are closer to how the headphones sound to him than what the other YouTubers provide whom he critiques in the video.

 

[Objectivist01]

During listening and comparing the original file with the cans, there is quite a resemblance of the original and for some headphones. Those ones also tend to measure quite neutral, like HD-600 and BD DT-250. The same goes if you have measured excessive highs; it is not difficult to hear this.

Dude, you are playing it through a headphone which has its own coloration. So what you hear as neutral there is already colored.

 

[Thomas_A]

Dude, you are playing it through a headphone which has its own coloration. So what you hear as neutral there is already colored.

Sure but I am comparing the original file against the file measured through another headphone. Any coloration from my own headphone is added to both, and I am hearing the difference.

 

[Mad_Economist]

From that video it looks like he's using the same compensation curve for every headphone. This won't work for the miniDSP EARS, because it does not have the same acoustic impedance as the human ear, which results in the correct compensation curve needing to be different for every headphone. Professional acoustic engineer Oratory, who's specific job it is to measure headphones, explains all this with measurements in this excellent post. This all means Oluv's recordings aren't really of much use.

Oluv primarily uses a set of blocked canal binaural microphones in his own ears at the moment. When he used an EARS, he adjusted the recording equalization curve on a per headphone basis, but for obvious reasons that's not really the preferable solution.

 

[Objectivist01]

Sure but I am comparing the original file against the file measured through another headphone. Any coloration from my own headphone is added to both, and I am hearing the difference.

And yet you don't know how the orginal file is as your headphone is still colored.

Example: Consider your headphone has a dip of 6db at 3khz. Headphone A has a boost of 6db at 3khz, headphone B has -3db at 3khz. Now what how since your heaphone pushes everything at 3khz to -6db at 3khz how would you know the difference between these headphones ?

 

[Jimbob54]

And yet you don't know how the orginal file is as your headphone is still colored.

Example: Consider your headphone has a dip of 6db at 3khz. Headphone A has a boost of 6db at 3khz, headphone B has -3db at 3khz. Now what how since your heaphone pushes everything at 3khz to -6db at 3khz how would you know the difference between these headphones ?

What? In demos like the one in the op, if he instead played a 3khz 0db test tone, to your ears through your hp, A would be louder than B. That's not so hard is it? Regardless of your cans, assuming both tests recorded reasonably well, the brighter demo set will sound brighter. End of

Neither will sound as they would if you were listening to them on your ears. That's not the point. It shows (to a degree) their relative characteristics

EDIT- lets put it another way. Youre recording an electric guitarist. You ask him to play one note with amp turned to 10, your record it. You ask him to play same but with amp at 5 and record. Same mics, same levels through the recording function. You play both back through headphones that have a large dip or bump at the frequency as that note. The first note played by the guitarist will be louder than the second, all other things being equal. That would be the same whatever the frequency response of your transducers

 

[lashto]

Does he level-match?

As far as I can see and hear, yes he does level matching. And generally tries to do pretty serious calibration. See my previous post for an example.

From that video it looks like he's using the same compensation curve for every headphone. This won't work for the miniDSP EARS, because it does not have the same acoustic impedance as the human ear, which results in the correct compensation curve needing to be different for every headphone. Professional acoustic engineer Oratory, who's specific job it is to measure headphones, explains all this with measurements in this excellent post. This all means Oluv's recordings aren't really of much use.

don't have a link but there is a video where he also says that miniDSP Ears is quite ******/useless and that's why the newer videos use binaural, in-ear mics.

Not sure why some people are so up in arms about this. Of course the setup has various limitations but it's an excellent resource for A/B testing. I wish he did a lot more of those tests.

 

[JohnYang1997]

He uses ears to eq recordings for every headphones to sound that resembles the original on hd600. IMO vastly more representative than any other sound check sources.

 

[JohnYang1997]

Oluv primarily uses a set of blocked canal binaural microphones in his own ears at the moment. When he used an EARS, he adjusted the recording equalization curve on a per headphone basis, but for obvious reasons that's not really the preferable solution.

Very true. Yet much better than anything else in comparison. Binaural microphone + EQ still sounds much better than EARS+EQ.

 

[NDC]

I was quite interested in Oluv's Focal Elex vs HD800s comparison video. I have always thought of the HD800s as a reference pair of headphones but his perspective was that they are just "decent" but not worth the asking price. The Focal Elex's sounded a lot better compared to the HD800s via the binaural recording through my HD58x's. I just got the most recent Drop run of the HE4xx's and am pretty impressed by the sound of them as well... they sound better than the binaural recordings of the HD800s to my ears! I was totally expecting the HD800s to blow everything I had out of the water...