This is a review and detailed measurements of the LCD-X planar magnetic headphone. It is on kind loan from a member. The LCD-X costs US $1,199 from Amazon including prime shipping by itself. The version I have came with travel case and such which costs US $1,700 I believe.
This is one sturdy, large and solid headphone:
It feels very heavy but when I wore it for a couple of hours, I did not feel the weight. But rather, felt the pressure around my cheeks where the cups are.
It comes with flexible and thin braided line as you see which was nice. The removable connectors have a high quality feel to them as well.
This headphone marks my first attempt at formal testing of headphones with a proper test suite using the loaned Gras 45CA from Audio Precision/GRAS:
Nice thing about this fixture is that we have a lot of research as to what response out of this fixture correlates with listener preference.
The large cups on the LCD-X and easy of use of 45CA made it quite easy to get good measurements out of this fixture. There are dual couplers (microphones) in the 45CA allowing me to adjust the headphone until I get both time (phase) and amplitudes close.
Being used, as you see in the picture, the pads deformed differently from each other but despite that, I got pretty close out of each side.
Headphone Distortion Measurement
I spent a good number of hours trying to come up with a useful dashboard as we have in our electronic tests to determine distortion. Initially I used dual-tone intermodulation test but that proved to be problematic. The frequency response of these headphones are all over the place and that messes with the amplitude of the primary tones. So consistency was not there. I went back to single tone then but instead of using 1 kHz as we do in electronic testing, I chose 40 Hz. Based on a survey of library, this is the frequency with the highest amplitude so makes for a good test of bass distortion. To have a reference, here is the dashboard for Sennheiser HD-650:
Focusing on the FFT top right, we see our primary peak at 40 Hz. There is clear second harmonic peak at 80 Hz and then third harmonic at 120 Hz. The rest of the harmonics are there as well going up to 240 Hz+. I truncated the spectrum after 400 Hz since it is just noise in my room (I plan to get a sound proof box for the future).
Our friend the SINAD is back
indicating a distortion-free range of just 26 dB. BTW, the SPL readings are approximate as I have not calibrated the 45CA level (will do in the future). But it is kept consistent between tests so we are dealing with same level of loudness.
Here is the LCD-X response:
Wow! There is essentially no harmonic distortion out of this headphone! We have a small peak at 80 Hz to the tune of -79 dB. SINAD should be that but due to the fact that it includes room noise, it degrades down to 61 dB or so. But even then, it is a whopping 35 dB better than the Sennheiser HD-650.
Looks like I lost the gain setting in the dashboard. Empirically, the LCD-X is far more efficient than the HD-650. It took much less drive to get to the same 100 dBSPL.
To measure the loudness limit, I did use IMD signal. Here is the comparison of the two:
We can clearly see how much lower the distortion is in the LCD-X (in red) and how much louder it can play.
Headphone Frequency Response Measurements
By far the most important characteristic of a headphone is frequency response. Here, we measure that and compare it to the preference test from Harman research using the same measurement fixture (very important);
The ideal headphone would track the dashed blue line. We obviously are not there. Follow how I read the measurements as it is not trivial to do so.
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 have learned to specially appreciate deep bass. To get there, you need to get close to the dashed blue line. Alas, many headphones are not even flat let alone track that blue line so this is not a defect in the headphone but a selection of a different "target curve."
Upper bass region in hundreds of hertz is a bit proud of our curve but that is fine.
Headphone Mid-range Response
As we cross 1 kHz, really bad things happen. Instead of peaking, the response stays flat and then actually droops down. We are talking worse case shortfall of 17 dB! There is no way, no how this is right. Big chunk of the spectrum is taken out of the music and thrown out.
Upper mid-range post 5 kHz actually recovers, nearly reaching our blue target.
Treble Response
Here we get into many complexities. For one, how accurate these fixtures are in this region is subject to debate and real issues. In addition, the response changes drastically with small physical movements of the headphone as you see in differing response between the two channels (red and green). There are also reflections between the driver and the fixture causing comb filtering (adding and subtracting). Finally, preference gets into the question of how much high frequency we want and desire.
Bottom line, I look at that region but mostly blur it out. In this case if you stand back far enough, the response is more or less what we have in blue.
Subjective Listening Tests
The state of headphone measurements is not where it is for speakers due to complexities of what I have explained above and then some. So it is important to also listen. For me, equalization is key to validating the measurements. If I make a correction of significant response errors and result is positive, then likely measurements told us the truth.
My first impression of LCD-X was that it sounded all wrong. I certainly did not sound like any speaker system I have listened to. So I pulled up the DSP/equalization settings in my Roon player and using the frequency graph dialed in two corrections:
The left side is a canned set of 3 filters meant to boost the low frequencies more and more with a bit of protection against clipping (the rolling down at lowest frequencies). I could have used a simpler one with a "shelf" filter but this worked good enough so I went with it.
Next and specific to LCD-X, I dialed in a massive 8 dB peak at 3000 Hz with a "Q" of 2 which meant that it was quite wide and filled most of that drop in mid-range. I was surprised how well this worked. I thought surely it would be too much but it was not.
With both filters in place, the sound transformed and become very nice. I set my Topping A90 headphone amplifier, set the gain to max and turned up the volume. Man, this headphone was producing exemplary bass performance and good tonality. I am talking about my usual ear lobes resonating. I then looked at the volume control and there was more room to go so I went for it! For just 1 second that I could tolerate it, it felt like I had a subwoofer in my head!!! If our hearing did not get damaged doing this, I would listen to these headphones this way all day long. It was uncanny how there was no distortion from either the amp or the headphone.
I then turned off the EQ and sound became totally uninteresting again.
Note that due to high boost of both bass and mid-range, I had to allow a lot of headroom to avoid digital clipping. Roon is nice in that it has an indicator when this happens. So pragmatically I dialed in about -3.5 dB before clipping went away across my sample music including those with heavy bass.
EDIT: forgot to include soundstage evaluation in the original review. So here it goes.
Soundstage was kind of interesting. Most of the instruments in the test track would dangle behind my head and a few inches to the left whereas mono content (vocals for example) would land in the middle of my head. Of note, equalization had a large effect on this causing more openness, showing that frequency response is what determines a lot of this, not any kind of inherent effect by itself.
Headphone EQ Verification
The testing was subjective and sighted so maybe I was all wrong. Fortunately we have a verification tool of some sorts. Audeze has worked with Roon to put in a set of canned EQs for their headphones. So I decided to measure the one for LCD-X and compare it to mine:
Very interesting! Once again we read the graph from left to right. They have put in a bass shelf recognizing lack of out put there (in red). Mine (in Green) is of course much more boosted but you know I like my clean bass per earlier remarks. Most importantly they also boost the region in 3 kHz region but again, not as much as I did as to say "our default is not that far off."
They also put in a third correction around 9 kHz which I did not bother with. That may or may not be needed. In general, I am not a fan of narrow high frequency corrections because our hearing discriminations (frequency resolution) is poor at high frequencies. So narrow correction like that will not sound like the way it looks (it will have a smaller effect).
All in all, we have agreement on what is deficient in the headphone. It is just a matter of degree. I think Audeze targets the out of box as being reference results but in my book, that is not reference. It is deficient.
Conclusions
I watched a bunch of video reviews of LCD-X online. At least a couple swore that this is the "end-game" type of headphone with sound that is better than anything else by far. I heard none of their accolades of "speed," "resolution," etc. How on earth do you get resolution when a good chunk of your mid-range is gone? As for speed, it seems people confuse lack of bass with speed. Yes, if you take away the bass notes, the sound is flatter and seems to linger less. But that is just a frequency response error, not any kind of speed thing. The base notes all have the same frequency and move very slow anyway. If you sped them up, they would change their sound! Anyway, we digress. Kudos to a few reviewers who noted the errors in mid-range.
Out of the box, the response error is so large that I would say LCD-X is a failed design. Whether that is intention or what, I don't know. Fortunately equalization is very effective and completely transforms this headphone. Combined with extreme power handling, efficiency and super low distortion, this becomes one nice headphone once signal processing is applied.
Alas, selling a headphone for $1,200+ that doesn't remotely sound right out of the box doesn't sit well with me. So I am going to give the Audeze LCD-X my first thumbs down of any headphone I have tested. If you own it, equalize it. Otherwise you and I are not going to be friends!
-----------
As always, questions, comments, recommendations, etc. are welcome.
Our septic is finally online and everything is back to the way it was. So decided to wash up some of the Apples and dehydrate them (they are starting to go bad at alarming rate);
They look beautiful on the outside but unfortunately, many have worm holes in them. It is quite time consuming to cut those sections out. Reminds of my audio testing. A lot of stuff looks great on the outside until you look on the inside (using measurements)!
Anyway, they dried overnight and tasted wonderful this morning. A bit tart and sweet at the same time.
As usual, I crave as much money as I can get out of you all so please donate what you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
This is one sturdy, large and solid headphone:
It feels very heavy but when I wore it for a couple of hours, I did not feel the weight. But rather, felt the pressure around my cheeks where the cups are.
It comes with flexible and thin braided line as you see which was nice. The removable connectors have a high quality feel to them as well.
This headphone marks my first attempt at formal testing of headphones with a proper test suite using the loaned Gras 45CA from Audio Precision/GRAS:
Nice thing about this fixture is that we have a lot of research as to what response out of this fixture correlates with listener preference.
The large cups on the LCD-X and easy of use of 45CA made it quite easy to get good measurements out of this fixture. There are dual couplers (microphones) in the 45CA allowing me to adjust the headphone until I get both time (phase) and amplitudes close.
Being used, as you see in the picture, the pads deformed differently from each other but despite that, I got pretty close out of each side.
Headphone Distortion Measurement
I spent a good number of hours trying to come up with a useful dashboard as we have in our electronic tests to determine distortion. Initially I used dual-tone intermodulation test but that proved to be problematic. The frequency response of these headphones are all over the place and that messes with the amplitude of the primary tones. So consistency was not there. I went back to single tone then but instead of using 1 kHz as we do in electronic testing, I chose 40 Hz. Based on a survey of library, this is the frequency with the highest amplitude so makes for a good test of bass distortion. To have a reference, here is the dashboard for Sennheiser HD-650:
Focusing on the FFT top right, we see our primary peak at 40 Hz. There is clear second harmonic peak at 80 Hz and then third harmonic at 120 Hz. The rest of the harmonics are there as well going up to 240 Hz+. I truncated the spectrum after 400 Hz since it is just noise in my room (I plan to get a sound proof box for the future).
Our friend the SINAD is back
indicating a distortion-free range of just 26 dB. BTW, the SPL readings are approximate as I have not calibrated the 45CA level (will do in the future). But it is kept consistent between tests so we are dealing with same level of loudness.Here is the LCD-X response:
Wow! There is essentially no harmonic distortion out of this headphone! We have a small peak at 80 Hz to the tune of -79 dB. SINAD should be that but due to the fact that it includes room noise, it degrades down to 61 dB or so. But even then, it is a whopping 35 dB better than the Sennheiser HD-650.
Looks like I lost the gain setting in the dashboard. Empirically, the LCD-X is far more efficient than the HD-650. It took much less drive to get to the same 100 dBSPL.
To measure the loudness limit, I did use IMD signal. Here is the comparison of the two:
We can clearly see how much lower the distortion is in the LCD-X (in red) and how much louder it can play.
Headphone Frequency Response Measurements
By far the most important characteristic of a headphone is frequency response. Here, we measure that and compare it to the preference test from Harman research using the same measurement fixture (very important);
The ideal headphone would track the dashed blue line. We obviously are not there. Follow how I read the measurements as it is not trivial to do so.
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 have learned to specially appreciate deep bass. To get there, you need to get close to the dashed blue line. Alas, many headphones are not even flat let alone track that blue line so this is not a defect in the headphone but a selection of a different "target curve."
Upper bass region in hundreds of hertz is a bit proud of our curve but that is fine.
Headphone Mid-range Response
As we cross 1 kHz, really bad things happen. Instead of peaking, the response stays flat and then actually droops down. We are talking worse case shortfall of 17 dB! There is no way, no how this is right. Big chunk of the spectrum is taken out of the music and thrown out.
Upper mid-range post 5 kHz actually recovers, nearly reaching our blue target.
Treble Response
Here we get into many complexities. For one, how accurate these fixtures are in this region is subject to debate and real issues. In addition, the response changes drastically with small physical movements of the headphone as you see in differing response between the two channels (red and green). There are also reflections between the driver and the fixture causing comb filtering (adding and subtracting). Finally, preference gets into the question of how much high frequency we want and desire.
Bottom line, I look at that region but mostly blur it out. In this case if you stand back far enough, the response is more or less what we have in blue.
Subjective Listening Tests
The state of headphone measurements is not where it is for speakers due to complexities of what I have explained above and then some. So it is important to also listen. For me, equalization is key to validating the measurements. If I make a correction of significant response errors and result is positive, then likely measurements told us the truth.
My first impression of LCD-X was that it sounded all wrong. I certainly did not sound like any speaker system I have listened to. So I pulled up the DSP/equalization settings in my Roon player and using the frequency graph dialed in two corrections:
The left side is a canned set of 3 filters meant to boost the low frequencies more and more with a bit of protection against clipping (the rolling down at lowest frequencies). I could have used a simpler one with a "shelf" filter but this worked good enough so I went with it.
Next and specific to LCD-X, I dialed in a massive 8 dB peak at 3000 Hz with a "Q" of 2 which meant that it was quite wide and filled most of that drop in mid-range. I was surprised how well this worked. I thought surely it would be too much but it was not.
With both filters in place, the sound transformed and become very nice. I set my Topping A90 headphone amplifier, set the gain to max and turned up the volume. Man, this headphone was producing exemplary bass performance and good tonality. I am talking about my usual ear lobes resonating. I then looked at the volume control and there was more room to go so I went for it! For just 1 second that I could tolerate it, it felt like I had a subwoofer in my head!!!
If our hearing did not get damaged doing this, I would listen to these headphones this way all day long. It was uncanny how there was no distortion from either the amp or the headphone.I then turned off the EQ and sound became totally uninteresting again.
Note that due to high boost of both bass and mid-range, I had to allow a lot of headroom to avoid digital clipping. Roon is nice in that it has an indicator when this happens. So pragmatically I dialed in about -3.5 dB before clipping went away across my sample music including those with heavy bass.
EDIT: forgot to include soundstage evaluation in the original review. So here it goes.
Soundstage was kind of interesting. Most of the instruments in the test track would dangle behind my head and a few inches to the left whereas mono content (vocals for example) would land in the middle of my head. Of note, equalization had a large effect on this causing more openness, showing that frequency response is what determines a lot of this, not any kind of inherent effect by itself.
Headphone EQ Verification
The testing was subjective and sighted so maybe I was all wrong. Fortunately we have a verification tool of some sorts. Audeze has worked with Roon to put in a set of canned EQs for their headphones. So I decided to measure the one for LCD-X and compare it to mine:
Very interesting! Once again we read the graph from left to right. They have put in a bass shelf recognizing lack of out put there (in red). Mine (in Green) is of course much more boosted but you know I like my clean bass per earlier remarks.
Most importantly they also boost the region in 3 kHz region but again, not as much as I did as to say "our default is not that far off."They also put in a third correction around 9 kHz which I did not bother with. That may or may not be needed. In general, I am not a fan of narrow high frequency corrections because our hearing discriminations (frequency resolution) is poor at high frequencies. So narrow correction like that will not sound like the way it looks (it will have a smaller effect).
All in all, we have agreement on what is deficient in the headphone. It is just a matter of degree. I think Audeze targets the out of box as being reference results but in my book, that is not reference. It is deficient.
Conclusions
I watched a bunch of video reviews of LCD-X online. At least a couple swore that this is the "end-game" type of headphone with sound that is better than anything else by far. I heard none of their accolades of "speed," "resolution," etc. How on earth do you get resolution when a good chunk of your mid-range is gone? As for speed, it seems people confuse lack of bass with speed. Yes, if you take away the bass notes, the sound is flatter and seems to linger less. But that is just a frequency response error, not any kind of speed thing. The base notes all have the same frequency and move very slow anyway. If you sped them up, they would change their sound! Anyway, we digress. Kudos to a few reviewers who noted the errors in mid-range.
Out of the box, the response error is so large that I would say LCD-X is a failed design. Whether that is intention or what, I don't know. Fortunately equalization is very effective and completely transforms this headphone. Combined with extreme power handling, efficiency and super low distortion, this becomes one nice headphone once signal processing is applied.
Alas, selling a headphone for $1,200+ that doesn't remotely sound right out of the box doesn't sit well with me. So I am going to give the Audeze LCD-X my first thumbs down of any headphone I have tested. If you own it, equalize it. Otherwise you and I are not going to be friends!
-----------
As always, questions, comments, recommendations, etc. are welcome.
Our septic is finally online and everything is back to the way it was. So decided to wash up some of the Apples and dehydrate them (they are starting to go bad at alarming rate);
They look beautiful on the outside but unfortunately, many have worm holes in them. It is quite time consuming to cut those sections out. Reminds of my audio testing. A lot of stuff looks great on the outside until you look on the inside (using measurements)!
Anyway, they dried overnight and tasted wonderful this morning. A bit tart and sweet at the same time.
As usual, I crave as much money as I can get out of you all so please donate what you can using : https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/
Last edited:
