New Audeze LCD3
Jun 18, 2013 at 12:47 PM Post #7,531 of 11,521
^ Have you had a chance to read this article:
 
http://www.innerfidelity.com/content/expert-tests-innerfidelitys-headphone-measurement-repeatability-and-reproducibility
 
This is using the Neumann head (Tyll's setup), but I'm pretty sure that Audeze uses a similar head. While their setups are not identical (I would venture to say that Tyll's is likely more accurate/precise), the results show that beyond 4kHz there can be some dramatic swings in frequency response with the EXACT same pair of headphones due to very slight positional changes on the head. Tyll goes a few steps further by measuring 5 positions on the head and then averages them to help "smooth" out the data. Audeze DOES NOT do this. So their swings would be far greater (also documented in the above article).
 
So my guess (and money) are the differences you're seeing is very minute changes on positioning of the LCD-3s on the dummy head as they are all above the 4kHz range. 
size]

 
Jun 18, 2013 at 2:17 PM Post #7,533 of 11,521
Quote:
To add more data points to the discussion....


Plot from my originally purchased LCD-3:


Plot from my just returned LCD-3:

 
Did you read my post above? 
size]

 
So pretty much identical up to 4kHz and then some small variations MOST likely a result of positioning the headphones on the dummy head. As Audeze DOES NOT take an average of standard positions, they're graphs would vary far more than those on innerfidelity.com.
 
Honestly, I don't give much credence to the Audeze graphs (due to a single run on the Neumann head and high amounts of smoothing applied to the data) and think it might be easier to not include them. Have a look at how Sennheiser documents frequency responses for their HD800s? They are more well aware of measurement variations and are far more careful to give more "generic" graphs.
 
Jun 18, 2013 at 2:19 PM Post #7,534 of 11,521
I think the graph that Audeze provides is for one purpose only...to prove they tested it. They probably know what they are looking for in terms of defect/variance. They surely know we are going to compare online, so its not in their interest to be shipping highly variable product. I agree it is more than likely related to the positioning equipment/positioning etc.
 
The reason it became a discussion point recently though is because LCD-2 graphs seem to vary less. However LCD-2 headphones have much firmer pads. Perhaps that means the variance in dummy listening position is much less.
 
Jun 18, 2013 at 2:55 PM Post #7,535 of 11,521
Quote:
LCD-3 most definitely has recessed treble.  Doesn't change that the PS1000 has a nasty treble and midbass hump and no extension at either end.  Engaging short term, grating once it wears off, coming from a previous Grado fan.

 
I know there are plenty of brighter headphones than the LCD-3, but coming from the much beloved (by a lot of head-fiers) HD650, the LCD-3 sounds brighter and airy in comparison.  I wouldn't like it to be any brighter, but it's all about personal preferences of course.
 
So, quoting you when saying "LCD-3 most definitely has recessed treble", I wouldn't want it to sound like a downside of the Audeze, just some kind of sound signature.
 
Jun 18, 2013 at 3:41 PM Post #7,538 of 11,521
Perfect, we all seem to agree.  Would I venture to say the LCD-3 are one of the "consensus" planars, as the HD600/HD650 are one of the "consensus" dynamic drivers ?
 
Jun 18, 2013 at 3:58 PM Post #7,539 of 11,521
Sounds reasonable to me :wink:

That's a great article of MH's over on Inner Fidelity. And yep I agree - purpose of graph is mostly to show that it's been tested.

And add a little value to ownership, and provide we headfiers one more thing to debate :p
 
Jun 18, 2013 at 4:21 PM Post #7,540 of 11,521
Did you read my post above? 
size]



So pretty much identical up to 4kHz and then some small variations MOST likely a result of positioning the headphones on the dummy head. As Audeze DOES NOT take an average of standard positions, they're graphs would vary far more than those on innerfidelity.com.

Honestly, I don't give much credence to the Audeze graphs (due to a single run on the Neumann head and high amounts of smoothing applied to the data) and think it might be easier to not include them. Have a look at how Sennheiser documents frequency responses for their HD800s? They are more well aware of measurement variations and are far more careful to give more "generic" graphs.


Of course MH, I read all of your posts. :)

What I did not mention, is the drivers in my LCD-3 had been replaced. So what the graphs represent are differences between those drivers and not test precision.

I do not put a lot of stock in those graphs. For one, they are an averaged composite of R & L channels and do not show the inbalance I was experiencing.

I just threw them for more data points for all to consider. Or not.
 
Jun 18, 2013 at 4:27 PM Post #7,541 of 11,521
Quote:
I think the graph that Audeze provides is for one purpose only...to prove they tested it. They probably know what they are looking for in terms of defect/variance. They surely know we are going to compare online, so its not in their interest to be shipping highly variable product. I agree it is more than likely related to the positioning equipment/positioning etc.
 
The reason it became a discussion point recently though is because LCD-2 graphs seem to vary less. However LCD-2 headphones have much firmer pads. Perhaps that means the variance in dummy listening position is much less.

 
Could very well be the softer pads and the levels of compression would vary more than the stiffer LCD-2 pads. Very good point. Tyll mentioned when we worked on the article that some headphones were more "position" dependent than others (with the HD800s being one of the least affected. 
 
Quote:
Of course MH, I read all of your posts. :)

What I did not mention, is the drivers in my LCD-3 had been replaced. So what the graphs represent are differences between those drivers and not test precision.

I do not put a lot of stock in those graphs. For one, they are an averaged composite of R & L channels and do not show the inbalance I was experiencing.

I just threw them for more data points for all to consider. Or not.

LoL, sometimes I wonder if anyone reads them. 
size]
 In the article, measuring the exact same pair of HD800s showed similar variation (precision) in the treble region. It (the variability) was reduced when taking 5 averages of 5 different standard positions. However, Audeze does not do this, so I would definitely expect their data to vary quite a bit more. So in the end, we can't statistically say that the differences between both drivers were from the drivers or from measurement variability. And I would venture to say (based on the data that I've seen to date) that the majority of it would be from the latter.
 
Cheers!
 
Jun 18, 2013 at 6:07 PM Post #7,543 of 11,521
Could very well be the softer pads and the levels of compression would vary more than the stiffer LCD-2 pads. Very good point. Tyll mentioned when we worked on the article that some headphones were more "position" dependent than others (with the HD800s being one of the least affected. 

LoL, sometimes I wonder if anyone reads them. 
size]

 In the article, measuring the exact same pair of HD800s showed similar variation (precision) in the treble region. It (the variability) was reduced when taking 5 averages of 5 different standard positions. However, Audeze does not do this, so I would definitely expect their data to vary quite a bit more. So in the end, we can't statistically say that the differences between both drivers were from the drivers or from measurement variability. And I would venture to say (based on the data that I've seen to date) that the majority of it would be from the latter.

Cheers!


Comparing the variability of multiple measurements from a single test setup of single pair HD800s with one off measurements from multiple test setups of multiple pairs LCD-3s is apples and oranges. All in my opinion, of course. :wink:
 
Jun 18, 2013 at 6:10 PM Post #7,544 of 11,521
Quote:
Comparing the variability of multiple measurements from a single test setup of single pair HD800s with one off measurements from multiple test setups of multiple pairs LCD-3s is apples and oranges. All in my opinion, of course. :wink:

Hard to say really. But Tyll did mention that the HD800s were amongst the "least" placement sensitive headphones he's tested (and he's tested a lot). So that's another piece of data to consider too. But they both share similarities in the Neumann head which can lead to these treble variations. So yes, we are extrapolating here, but in a logical way I think. 
size]

 
Jun 18, 2013 at 6:53 PM Post #7,545 of 11,521

Users who are viewing this thread

Back
Top