• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Audeze LCD-X Over Ear Open Back Headphone Review

OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
Note the LCD-X have a distortion minimum right around the frequency tested here, 40 Hz, so this result may be under-representative of their average distortion across the rest of the frequency range. See Innerfidelity's distortion measurements of them here:
I would caution against putting any trust in distortion measurements from Tyll. This wasn't his forte and there is no evidence that his measurements are correct. Here are the THD levels I have at the same level that the frequency response was measured for LCD-X (different than the dashboard):

1602831900743.png


As you see, there is no dip at all at 40 Hz. It wouldn't make sense to have a dip in distortion anyway. But rather, you are falling victim to misleading way THD percentage works in systems with varying frequency response like speakers and headphones. There, the amplitude changes and that causes THD% to vary even though the distortion mechanism itself does not. Here is Tyll's frequency response test that shows this:

1602832420168.png


Notice as I circled, his measurements show sudden increase in response around 40 Hz. So for the same distortion amount, the percentage will shrink.

In my testing I am using a single tone @40 Hz and assuring that the identical reproduction level is used in both headphones. As a result, frequency response variations are dialed out and THD ratios can be compared.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
May you help me, and us, understand what this mean?
It means that if you look what part of the spectrum of music on average is loudest, it is 40 Hz. So if something is going to induce distortion due to loudness, it would be the notes around 40 Hz.

Now, about the 650, do we have an understanding of what's happening there?
I do not as that goes beyond the scope of my testing. It is possible that the driver in 650 is much smaller and goes through much more excursion causing distortion to sharply increase. But I am guessing as the technologies here are very different. As we test more headphones this way, we get more insight.
 

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
3,303
Likes
3,838
It means that if you look what part of the spectrum of music on average is loudest, it is 40 Hz. So if something is going to induce distortion due to loudness, it would be the notes around 40 Hz.


I do not as that goes beyond the scope of my testing. It is possible that the driver in 650 is much smaller and goes through much more excursion causing distortion to sharply increase. But I am guessing as the technologies here are very different. As we test more headphones this way, we get more insight.
thanks
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
I don't think a single 40hz tone is enough to measure distortion in headphones... CSD would also be nice.
CSDs of all kinds are super complex to understand and produce. Be very careful in looking at them. People fall in love with the prettiness of the 3-D graph. The value is not there.

Still would love to see an IMD sweep or 32-tone test done for headphones, since multi-tone measurements aren't often done except just looking at square waves.
This is easy to run but getting meaningful results is fleeting. Reason is that the frequency response of the headphones varies a ton and that makes the 32-tone amplitudes all over the place. As such, you can't compare one headphone to another. We don't have that problem with electronics because frequency response is ruler flat.

I have spent days running different distortion tests. Acoustic products are very different animals and hard to characterize this way.
 
OP
amirm

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,381
Location
Seattle Area
While I applaud Amir considering adopting a headphone measurement rig to round out what ASR is capable of, given the number of other headphone measurement sources available (mentioned in the thread) I would prefer products measured where others are lacking. ie Speakers and amplifiers. I would prefer to somehow see a consistent DSP/room correction assessment method deployed, if practicable.
So far headphone testing is pretty quick. So I should be able to squeeze them in without a lot of impact on other tests.
 

crinacle

Member
Reviewer
Joined
Jul 31, 2019
Messages
60
Likes
788
Location
SG
CSDs of all kinds are super complex to understand and produce. Be very careful in looking at them. People fall in love with the prettiness of the 3-D graph. The value is not there.


This is easy to run but getting meaningful results is fleeting. Reason is that the frequency response of the headphones varies a ton and that makes the 32-tone amplitudes all over the place. As such, you can't compare one headphone to another. We don't have that problem with electronics because frequency response is ruler flat.

I have spent days running different distortion tests. Acoustic products are very different animals and hard to characterize this way.

Which is why I find THD measurements for headphones to be pretty misleading, especially in the lower frequencies given that most measurements seem to be dominated by lower order harmonic distortion and so most susceptible to the effects of auditory masking.
 

mt196

Active Member
Joined
Apr 7, 2019
Messages
229
Likes
156
Location
Rome, Italy
Buy why should we take Harman as the main sounds signature? I don't like it and the are plenty of people not liking it. This is not something objective as a dac or amplifier sinad that is objective that should be the lowest as possible.

Also, in a forum that is all about the science, how can we overlook the fact that, is listening is also important for headphones as you said in the review, you can have biases due to having seen the measurements and having made up your mind about how they sound before listening to them. In addition, the same bias applies when you use eq that can sound better to you just because it is correcting what you didn't like visually on a graph
 

JohnYang1997

Master Contributor
Technical Expert
Audio Company
Joined
Dec 28, 2018
Messages
7,175
Likes
18,292
Location
China
Buy why should we take Harman as the main sounds signature? I don't like it and the are plenty of people not liking it. This is not something objective as a dac or amplifier sinad that is objective that should be the lowest as possible.

Also, in a forum that is all about the science, how can we overlook the fact that, is listening is also important for headphones as you said in the review, you can have biases due to having seen the measurements and having made up your mind about how they sound before listening to them. In addition, the same bias applies when you use eq that can sound better to you just because it is correcting what you didn't like visually on a graph
Then just read the raw frequency response.
 

mt196

Active Member
Joined
Apr 7, 2019
Messages
229
Likes
156
Location
Rome, Italy
Then just read the raw frequency response.
Yeah that is what I actually did and will do if headphones' reviews are done in this way. There is no point for me in judging a headphone based on the Harman curve or on thd that is truly misleading. I just hope people read through reviews completely and don't get fooled just by a headless panther...
 

magicscreen

Senior Member
Joined
May 21, 2019
Messages
300
Likes
176
"Headphone Bass Response
Here the response is "flat." Harman research backed by my own listening tests show that we crave much more bass than this. "

I do not understand this subjective point of view.
Then we can say that the 2nd harmonic distortion is more pleasant, we crave much more for it. And any tube amplifier has superior sound quality.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,687
Likes
4,068
Thanks @amirm , great review!

Why is the 40 Hz peak broader on the Audeze dashboard compared to the HD650 dashboard?
 

Chocomel

Active Member
Joined
Feb 18, 2019
Messages
106
Likes
322
Another question, what is the point of measuring imd at 132db? This is far from a normal case scenario

Amir said his rig isn't calibrated yet so the absolute level is probably off. But the idea is to see how loud the headphone can go without essentially "clipping". It's not something that's much of an issue in practice but in some cases people do encounter issues with it with headphones. For example with the focal open back headphones, although i do worry about either the focal QC or people's ears. Nevertheless it gives more of an idea of the performance of the headphone and in edge cases might explain what people are experiencing.
 

Saeniv

Member
Joined
Apr 30, 2020
Messages
12
Likes
3
Sorry for the opinion.
But Headphones should be about personal preference you have to hear theme to make your judgement.
I do understand that electronics should be judged by numbers and personal preference. Even that seems wrong from time to time. If you mostly use Flac's or any other format from Cd's, you never need 120db, or more.

But judging headphones on a "reference curve" seems completly wrong. Every earcanal is diffrent, every hearing abilities are diffrent, every preference is diffrent, and so on. Even the wearing comfort has impact on the subtle sound quality.

Where is the love for music and enjoyment.

Also alot of headphone measurments:
https://diyaudioheaven.wordpress.com/
 
Last edited:

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,462
Location
Land O’ Lakes, FL
Yep, this has now been automated and made freely available by Listen Inc. and courtesy of Harman:


The Excel (and other) files for both over and in-ear Harman preference rating calculations can be found in the link below, under the 'Get Sequence' section (a short registration form just needs to be filled out to download it):

https://www.listeninc.com/products/...-of-in-ear-headphones-using-the-harman-model/

The formula parameters (the standard deviation of the headphone's error curve and the slope of the logarithmic regression fit of the error) could even be presented to give a breakdown of the score, the former showing how much the frequency response deviates from neutral (the Harman target), and the latter particularly useful in determining whether the overall spectral balance of the headphone is warmer (slope < 0), brighter (slope > 0), or generally neutrally balanced across the entire audible range (slope ~ 0).

Oratory calculated a preference rating of 49/100 from his measurements of the Audeze LCD-X by the way, and 82/100 after applying his EQ settings (all displayed in his pdf for it, as he does for all the headphones he's measured).
Not sure what the stated accuracy is, but I would assume the preference ratings for headphones is not as accurate as with speakers. And, unless Amir does a whole suite of measurements (phase matching, group delay, etc.), we currently can’t do a deep dive to see how well the headphones image, how wide their soundstage is, how well the emulate the sound as being in front of us and not in our heads, etc.


And of course you have to actually wear the things so a good chunk of people will choose worse sounding headphones if they are noticeably more comfortable.

Unless I’m mistaken, here is the formula for AE/OE:
114.49-(12.62*SD)-(15.52*AS)
Where:
SD is standard deviation from 20Hz-10kHz
AD is absolute value of slope 20Hz-10kHz

There isn’t even a bass frequency portion like there is with the speaker preference rating.

Here is the accuracy of their model:
DCC365A5-0A8F-4D72-B38F-D0E0304FFEAE.jpeg



Compared to the preference rating for speakers:
18B33BFC-DC50-453A-A9AF-60A1419E4552.jpeg


For headphones, there were only 3 models that got scored >6/10 by humans, compared to ~15 models for speakers
 
Last edited:

bidn

Active Member
Joined
Aug 16, 2019
Messages
195
Likes
821
Location
Kingdom of the Netherlands
Outstanding test, as usual, Amir ! Such an excellent refresher compared to all the crap of the fake, nonsensical "reviews" abounding on the Internet and in audiophile magazines!

I bought (and still own) these headphones quite a number of years ago. I fully agree with you and confirms that they are complete failures re. high- fidelity as their high-mid and treble response are totally wrong!
I found a positive side though to this total failure : when I have a headache, I can't listen to fatiguing music like metal and in such a case I have found some utility for these definitely wrong- sounding headphones (because they suppress most of the treble!).

As a relative consolation, these are not even the worse headphones produced by Audeze : when the much more expensive LCD-4 arrived, I was more experienced and had at last realized by that time that most audiophile reviews are fully misleading (be it due to incompetency or financial reasons or both) and it was a shock to me to hear (auditioning them twice due to all the reviewers' audiophile praise they were receiving) these super expensive LCD-4 as being even worse than the LCD-X!!!! For me that is even more scandalous! Having learned from my mistakes, I fortunately didn't buy this even more expensive piece of crap.

BTW the only Audeze headphones that I found to have an acceptable sound (among those I heard) are the LCD-XC (not speaking of other issues like weight...).
 
Last edited:

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
3,303
Likes
3,838
Buy why should we take Harman as the main sounds signature? I don't like it and the are plenty of people not liking it. This is not something objective as a dac or amplifier sinad that is objective that should be the lowest as possible.

Also, in a forum that is all about the science, how can we overlook the fact that, is listening is also important for headphones as you said in the review, you can have biases due to having seen the measurements and having made up your mind about how they sound before listening to them. In addition, the same bias applies when you use eq that can sound better to you just because it is correcting what you didn't like visually on a graph
Is it fully objective? it might not, but it's based on science, not on taste. Olive's team went to great lenght to characterize that curve, real physics, maths, acoustics. taking into account how frequencies hit your heardrums differently if the source is very near your hear compared to how it would be perceived in a more "natural" listening environment. Think about it, would a piano sound natural if you had your hear in the piano, 1 cm from the hammer hitting the string? I'm not saying that's what headphone do, but I am saying that we don't hear the same thing, from a same source, depending of the environment. the shape of the hears have a much larger effect on the actual perceived sound with headphones than with speakers. Binaural recordings would negate this but those are rare. Mastering and mixing engineers, still mix and master with speakers as the main reference, and Haman's studies are based on that. When they do use headphone, to asses if their mix translate, if the accepted reference is to have non flat response, that's what they'll use as well. Is it a perfect model? no, but it's a better model than flat. Flat wouldn't make sense. They happen to have the most torough analysis until now. It's not subjective to asses that it's better than no compensation at all, but yes it could also be incomplete and at some point some of theses findings could be contradicted, but it is quite objective that we do need some sort of transfer function.
 

Jimbob54

Grand Contributor
Forum Donor
Joined
Oct 25, 2019
Messages
11,064
Likes
14,694
So we're going all in on the Harman target curve?

Would appear so- I am not sure at all this is a good thing, but as long as readers understand that is one of the key deciders for the panther- so be it.

What I find interesting is that from a technical perspective (distortion etc) its great- but from a stock tuning angle- dud.

Which makes me wonder @amirm, should there be a two tier rating system? Most here know that EQ is pretty essential to get the best out of almost all transducers. A headless panther feels right for the stock performance, but not for what this (or perhaps any) HP can do.

Maybe headless should be reserved for a HP that is neither well made, bad FR as stock and also cant easily be EQ'd to something far closer to target without distorting?

Im overly complicating and as a subjective fan of the LCD-X should just take it on the chin . :rolleyes:
 

mt196

Active Member
Joined
Apr 7, 2019
Messages
229
Likes
156
Location
Rome, Italy
Is it fully objective? it might not, but it's based on science, not on taste. Olive's team went to great lenght to characterize that curve, real physics, maths, acoustics. taking into account how frequencies hit your heardrums differently if the source is very near your hear compared to how it would be perceived in a more "natural" listening environment. Think about it, would a piano sound natural if you had your hear in the piano, 1 cm from the hammer hitting the string? I'm not saying that's what headphone do, but I am saying that we don't hear the same thing, from a same source, depending of the environment. the shape of the hears have a much larger effect on the actual perceived sound with headphones than with speakers. Binaural recordings would negate this but those are rare. Mastering and mixing engineers, still mix and master with speakers as the main reference, and Haman's studies are based on that. When they do use headphone, to asses if their mix translate, if the accepted reference is to have non flat response, that's what they'll use as well. Is it a perfect model? no, but it's a better model than flat. Flat wouldn't make sense. They happen to have the most torough analysis until now. It's not subjective to asses that it's better than no compensation at all, but yes it could also be incomplete and at some point some of theses findings could be contradicted, but it is quite objective that we do need some sort of transfer function.
I agree I must say. In fact, my major criticisms were about: non calibrating the measuring device because ok it is consistent with its own measures, but what about consistency with headphones measurements over the web? (There are a lot, conversely to DAC and Amp measurements that cost a lot more to do). Another problem is measuring IMD nowhere near the standard use case, that makes only things perform worse, who cares if my headphones have a problem at 130db, I care if there are a 90... and finally, but probably more importantly, the fact that reviews include a subjective part in which there are a lot of biases going on (like having seen at graphs before listening and thus partially having made your mind about how headphones will sound, or even stating that EQ made them better without considering that what you hear as better may be partly biased by the fact that you are correcting the part of the graph that you didn't like, but that maybe you don't actually hear). Audio Science should be about science, measurements, and graphs, not subjectivism and biases hidden behind a measurement device. If you just look at graphs, ASR is a immensely valuable source, but if you start looking at the subjective part of the reviews and the panther rating system, it just becomes similar to a lot of here criticized review websites.
 
Top Bottom