Audeze LCD-XC 21 & effects of small venting

[Robbo99999]

Thanx, I shall do it in proper time.
So far, i haven't found oratory's fr curve for LCD-XC 2021 model.
The newer model is different to older versions.
Can you send me a link, if there is one?

Ah, you've got the pre 2021 model then! (I thought you had the post 2021 model). Reddit is down at the moment for some reason but I was able to find some direct links from a post of Oratory's in the LCD XC 21 thread here on ASR:

Audeze LCD-XC Review (Closed-back Headphone)

@04gto Wow, I'm happy to hear someone say that the XC is their favorite headphone. So often it is overshadowed by the X. I have less than a month with my new XC and I love it of course! I know it's not fair to compare open backed vs. closed back but I can't help compare it to the X. They...

www.audiosciencereview.com

He posted those links on ASR in June 2022, so perhaps he's updated his pdf's since then, so might be better to look on Oratory's reddit when reddit is back working properly again, but those links can keep you going for now.

You don't have to go to all the trouble I outlined in REW to present your "miniDSP EARs mods" as theoretical GRAS measurements if you don't want to but it's the only real valid way of presenting your mods in GRAS measurement form which can be useful for people to put it into perspective, but it's not totally necessary as your miniDSP EARS measurements in your following graph are fine in terms of showing the difference of your mods:

Just you can't directly superimpose that over a GRAS measurement like you did later.

 

[Ken Tajalli]

Ah, you've got the pre 2021 model then! (I thought you had the post 2021 model). Reddit is down at the moment for some reason but I was able to find some direct links from a post of Oratory's in the LCD XC 21 thread here on ASR:

Audeze LCD-XC Review (Closed-back Headphone)

@04gto Wow, I'm happy to hear someone say that the XC is their favorite headphone. So often it is overshadowed by the X. I have less than a month with my new XC and I love it of course! I know it's not fair to compare open backed vs. closed back but I can't help compare it to the X. They...

www.audiosciencereview.com

He posted those links on ASR in June 2022, so perhaps he's updated his pdf's since then, so might be better to look on Oratory's reddit when reddit is back working properly again, but those links can keep you going for now.

You don't have to go to all the trouble I outlined in REW to present your "miniDSP EARs mods" as theoretical GRAS measurements if you don't want to but it's the only real valid way of presenting your mods in GRAS measurement form which can be useful for people to put it into perspective, but it's not totally necessary as your miniDSP EARS measurements in your following graph are fine in terms of showing the difference of your mods:

Just you can't directly superimpose that over a GRAS measurement like you did later.

No, I have the newer model.

 

[kemmler3D]

Hey, just wanted to say, big respect for actually doing the measurements (those EARS things aren't free). I was among the naysayers in the other thread, and I say with no irony, "well, you showed us".

I have said the same thing about EARS jigs recently... they're fine for checking relative changes, within limits, if you are careful.

This is a cool result and the effect is larger than I would have expected for such small "vents". I think this goes to show why fit and seal are so important for headphone performance. A tiny 1/16" (or so) hole causes an audible increase in bass... who would have guessed, other than headphone designers?

I also think this is a very safe / sane mod, as mods go. As long as you don't lose the screws, anyway...

 

[Ken Tajalli]

I have said the same thing about EARS jigs recently... they're fine for checking relative changes, within limits, if you are careful.

Absolutely, not just EARS, all of them.

This is a cool result and the effect is larger than I would have expected for such small "vents". I think this goes to show why fit and seal are so important for headphone performance. A tiny 1/16" (or so) hole causes an audible increase in bass... who would have guessed, other than headphone designers?

I did too!
But seriously, this is not new. Some thirty years ago or so, Dynaudio used to sell an adjustable speaker Vent, for DIY speakers. Its a trick! to give a smaller speaker box a virtual bigger characteristic. To my understanding.

I also think this is a very safe / sane mod, as mods go. As long as you don't lose the screws, anyway...

Original screws are common black steel US 4-40.
They don't even look good (to me).
Toughest job is to drill 1.5mm holes into stainless steel bolts that small.
8 of them! Broke 3 tips.

Glad you enjoyed this.

 

[kemmler3D]

I think the surprise is how small of a hole acts as an effective vent. From what I understand about speaker building, an aperiodic vent or port would be usually be larger relative to the enclosure. This is why headphone design seems like a dark art to me. Speaker performance is largely predictable from T/S parameters and the dimensions of the enclosure. I am not sure if the same is true about headphones, it doesn't seem so.

Audio wavelengths are large compared to the headphone enclosure up to 10khz or so... so I'm not sure what that implies for designing the enclosure. I guess Audeze and Dan Clark et al know...

 

[Ken Tajalli]

Just had a look at Oratory's measurement and EQ settings for this headphone.

It seems there is a hump at 210Hz, though with my measurements, it is not there, but his rig is much better than mine, so assuming he is correct, it seems the 1-2dB dip I am getting at the same frequency, is pretty much needed!
Indeed, the mod. is bang on according to Oratory.
About 5dB general boost and 1dB reduction at 210Hz.
I would love to hear Audeze's opinion on this.

 

[Robbo99999]

Just had a look at Oratory's measurement and EQ settings for this headphone.
View attachment 272226
It seems there is a hump at 210Hz, though with my measurements, it is not there, but his rig is much better than mine, so assuming he is correct, it seems the 1-2dB dip I am getting at the same frequency, is pretty much needed!
Indeed, the mod. is bang on according to Oratory.
About 5dB general boost and 1dB reduction at 210Hz.
I would love to hear Audeze's opinion on this.

View attachment 272227

Here's Oratory's measurement, there is no hump at 200Hz:

He's including the -1dB Peak Filter at that point just to bring it down to Harman Curve at that point. How you decide to align the Target on the measurement before you start creating the EQ will also influence where the various EQ filters are created - for instance if he had aligned that measurement at 1kHz with the Harman Curve then 200Hz point would be bang on the Harman Curve so he wouldn't need to put a -1dB peak filter at that point. So it's actually best to look at the actual measurement (lol) rather than the EQ Curve (nor individual filter list) if you want to determine if there is a peak at any point in the measured frequency response, ha!

Your mod does look like it synergises well with Oratory's measurement though if you eyeball it. So for the top two lines in your graph (the mods that produce the top two most bass - I'll attach your graph below), you can assume that you're quite close to Harman bass from 300Hz downwards. In fact if you look at Oratory's measurement (the above graph), then just imagine it tracking Harman from 300Hz down to about 30Hz - and then of course you can see the rest of the frequency response above 300Hz in Oratory's graph, you can see that's not perfectly tracking Harman up there, but it's quite easy to imagine the effect of your mods on Oratory's above measurement.

 

[Ken Tajalli]

Here's Oratory's measurement, there is no hump at 200Hz:

He's including the -1dB Peak Filter at that point just to bring it down to Harman Curve at that point. How you decide to align the Target on the measurement before you start creating the EQ will also influence where the various EQ filters are created - for instance if he had aligned that measurement at 1kHz with the Harman Curve then 200Hz point would be bang on the Harman Curve so he wouldn't need to put a -1dB peak filter at that point. So it's actually best to look at the actual measurement (lol) rather than the EQ Curve (nor individual filter list) if you want to determine if there is a peak at any point in the measured frequency response, ha!

Your mod does look like it synergises well with Oratory's measurement though if you eyeball it. So for the top two lines in your graph (the mods that produce the top two most bass - I'll attach your graph below), you can assume that you're quite close to Harman bass from 300Hz downwards. In fact if you look at Oratory's measurement (the above graph), then just imagine it tracking Harman from 300Hz down to about 30Hz - and then of course you can see the rest of the frequency response above 300Hz in Oratory's graph, you can see that's not perfectly tracking Harman up there, but it's quite easy to imagine the effect of your mods on Oratory's above measurement.

This is what I meant, and yes you are correct, only when you apply the Harman curve, it would appear that way.
At any rate, I was just patting myself on the back!
It seems that the slight dip at 210Hz after the mod. was a fortunate side effect, if you eyeball the Oratory's curve.
But what's a dB among friends?
It also appears that I had managed to get a better seal than both Resolve and now Oratory, because my bass does not droop below 60Hz! Or is that a characteristic of the GRAS rig?
But the shape of EARS curve and GRAS curve at bass, is very similar - so I am happy about the EARS, it seems it is doing its job.
At any rate, I am done with this mod.
After enlarging the vent holes to 1.5mm (that precise drilling took some doing), at least to my ears, the LCD-XC 21 requires no EQ to the bass region. It still needs abit of low treble reduction ( the 6k peak. the 4k droop doesn't bother me), but it is now a far simpler EQ profile, and best of all, no preamp cut.
I suppose, if one was to use one of those DCA Aeon's paper filters on these, it might be possible to not EQ at all.

 

[Robbo99999]

View attachment 272354
This is what I meant, and yes you are correct, only when you apply the Harman curve, it would appear that way.
At any rate, I was just patting myself on the back!
It seems that the slight dip at 210Hz after the mod. was a fortunate side effect, if you eyeball the Oratory's curve.
But what's a dB among friends?
It also appears that I had managed to get a better seal than both Resolve and now Oratory, because my bass does not droop below 60Hz! Or is that a characteristic of the GRAS rig?
But the shape of EARS curve and GRAS curve at bass, is very similar - so I am happy about the EARS, it seems it is doing its job.
At any rate, I am done with this mod.
After enlarging the vent holes to 1.5mm (that precise drilling took some doing), at least to my ears, the LCD-XC 21 requires no EQ to the bass region. It still needs abit of low treble reduction ( the 6k peak. the 4k droop doesn't bother me), but it is now a far simpler EQ profile, and best of all, no preamp cut.
I suppose, if one was to use one of those DCA Aeon's paper filters on these, it might be possible to not EQ at all.

That bass mod has worked out about as good as it could have for Harman Curve Bass fans, and the one & two screw mods with less bass could cater for people who prefer less than Harman bass.

In terms of miniDSP EARS bass measurements differing from GRAS - if you're using one of the flat cheek GRAS measuring devices which Amir uses and Oratory also then bass seal shouldn't really be a factor between them, but I'm unsure if bass measurements on miniDSP EARS are expected to be identical or very close for all types of headphone models.......for sure though miniDSP EARS most accurate part of the measurement will be in the bass area though, as miniDSP EARS frequency response goes nuts over 1kHz so you certainly can't expect anything over 1kHz on the miniDSP EARS to mirror GRAS measurements. My K702 unit that I sent to Oratory to measure, and which I then subsequently (a year or so later) measured on my then newly purchased miniDSP ears - the response 20Hz to 1000Hz did not vary greatly from the GRAS measurement. The best answer for you though is graph I've just remember I had from Oratory (he must have posted it somewhere):

Black line is showing the average difference between flat cheek GRAS and (of course flat cheek) miniDSP EARS when different headphone models were measured between the two measuring rigs. So that curve shows the miniDSP EARS overreads below 500Hz, and overreads the most at around 65Hz. Due to the slight u-shaped area of that overreading then that means that miniDSP EARS measurements should show a faster acceleration of bass drop off than a corresponding GRAS measurement (even though miniDSP EARS shows a bit more bass in general. So if you're well extended in your bass on the miniDSP EARS (as in no early roll-off) then this information is saying that your bass is likely to be well extended too. Your well extended bass looks pretty similar to Amir's in terms of 35Hz rolloff:

Note: Oratory also uses the BC version of the GRAS, which is the version with the more lifelike head (not flat cheek), but he also uses the flat cheek AG version too - he says he tries to display best case bass in terms of seal so a bit surprised to see Oratory's measurement starting to roll off at 60Hz (praps he mainly did that work on the BC version rather than AG). You should be pretty confident in the degree of bass extension your miniDSP EARS is showing though, judging from this information.

 

[Ken Tajalli]

FWIW
I superimposed the curve I got from my sample, in factory condition, on top of Oratory's curve, adjusting for scale in both horizontal and vertical. Take a look:

This damn EARS is not doing that bad!
I was expecting a lot more deviation.

 

[Robbo99999]

FWIW
I superimposed the curve I got from my sample, in factory condition, on top of Oratory's curve, adjusting for scale in both horizontal and vertical. Take a look:

View attachment 272969

This damn EARS is not doing that bad!
I was expecting a lot more deviation.

DOH! Be careful here, you can't just superimpose your miniDSP EARS measurement directly over the Harman Compensated GRAS measurement - doing so would imply that both miniDSP EARS rig and GRAS rig operate to the same standard, and the compensation files that miniDSP offer are of "unknown calculation origin" and are not equivalent to the GRAS standard. If there's someone here that knows how miniDSP EARS came up with their various compensation files then I wouldn't mind hearing about it, I've not really looked into it to be honest, but I've assumed that they have measured a number of different headphone models and then compared them to somekind of standard of their choosing through which they've developed their compensation curves - I immediately discounted their compensation curves when I received my miniDSP EARS due to the limitations of the miniDSP EARS never being able to faithfully replicate a GRAS measuring device in terms of creating a headphone target that would be applicable to all headphone models - my understanding is that this is because the impedance of miniDSP EARS is too far off the impedance of a human ear to be able to have a target curve that is valid between different headphone models. If you were to measure a specific unit of a headphone on the GRAS though, and then measure that very same unit on your miniDSP EARS, then you could create a compensation curve that could turn your miniDSP EARS measurement into a GRAS measurement, but that would only be valid for that particular headphone model, it wouldn't translate to other headphone models, it would be a different compensation curve for each different headphone model, as shown by the evidence of Oratory's graph which I showed in an earlier post: .

So you can't expect to use the compensation files that miniDSP EARS include as means of comparing your headphone measurements to GRAS measurements, which is what you've tried to do there in your post.

 

[Ken Tajalli]

DOH! Be careful here, you can't just superimpose your miniDSP EARS measurement directly over the Harman Compensated GRAS measurement

Its a free country, I can superimpose my own face on Trumps body, if I feel like it! (joking)

- doing so would imply (NOT by me) that both miniDSP EARS rig and GRAS rig operate to the same standard, and the compensation files that miniDSP offer are of "unknown calculation origin" and are not equivalent to the GRAS standard. If there's someone here that knows how miniDSP EARS came up with their various compensation files then I wouldn't mind hearing about it ( Me too, would help with optimization), I've not really looked into it to be honest, but I've assumed that they have measured a number of different headphone models and then compared them to somekind of standard of their choosing through which they've developed their compensation curves - I immediately discounted their compensation curves when I received my miniDSP EARS due to the limitations of the miniDSP EARS never being able to faithfully replicate a GRAS measuring device in terms of creating a headphone target that would be applicable to all headphone models - my understanding is that this is because the impedance of miniDSP EARS is too far off the impedance of a human ear to be able to have a target curve that is valid between different headphone models. If you were to measure a specific unit of a headphone on the GRAS though, and then measure that very same unit on your miniDSP EARS, then you could create a compensation curve that could turn your miniDSP EARS measurement into a GRAS measurement, but that would only be valid for that particular headphone model,(+1) it wouldn't translate to other headphone models, it would be a different compensation curve for each different headphone model, as shown by the evidence of Oratory's graph which I showed in an earlier post: .

Lets just rise above it for a second.
Of course you are correct in your points, and in no way am I comparing the EARS to GRAS. But the fact is, that some basic information is being obtained using either graph. Where the approximate peaks and dips are, not their levels though. The two graphs are more similar in that respect than different.
My sample is different to Oratory's sample, my gear is vastly different, conditions of the test are not even comparable, yet, EARS can come up with an approximate general shape of a professional rig. To this end, EARS has proven itself useful!
Add to that, the level at which I conducted my test was different too.
Don't worry, I have heard you loud and clear. I am not directly comparing the two curves.
I was just praising my new toy! now I know, it is more than just a toy.

 

[sharock]

Where in London do you live @Ken Tajalli ?

I'm considering some LCD-XC but have no way to audition them!

 

[Robbo99999]

Its a free country, I can superimpose my own face on Trumps body, if I feel like it! (joking)

Lets just rise above it for a second.
Of course you are correct in your points, and in no way am I comparing the EARS to GRAS. But the fact is, that some basic information is being obtained using either graph. Where the approximate peaks and dips are, not their levels though. The two graphs are more similar in that respect than different.
My sample is different to Oratory's sample, my gear is vastly different, conditions of the test are not even comparable, yet, EARS can come up with an approximate general shape of a professional rig. To this end, EARS has proven itself useful!
Add to that, the level at which I conducted my test was different too.
Don't worry, I have heard you loud and clear. I am not directly comparing the two curves.
I was just praising my new toy! now I know, it is more than just a toy.

Because we know it's not valid to compare the miniDSP EARS measurements with the GRAS measurements directly as you have done so, then any similarities can be placed down to coincidence, or certainly shouldn't have much weight placed on them - there's not really any defence for displaying invalid comparisons/information. I mean we're not gonna shoot you, but there's nothing wrong with pointing out the invalidity of such comparisons. (Your logical thinking is very confused in your sentence where you state "My sample is different to Oratory's sample, my gear is vastly different, conditions of the test are not even comparable, yet, EARS can come up with an approximate general shape of a professional rig. To this end, EARS has proven itself useful!", that's some very confused thinking to even begin to point out where the faults lie in the logic & understanding behind such a statement, instead it's far better to just to put a line through that & pretend it never existed.....if you're an arts based person with zero scientific background and without inclinations to attempt to step a toe into scientific rigour/thought then this is forgivable.)

 

[Ken Tajalli]

The more I say I am not comparing them, the more you insist that doing so is invalid.
Let's put it to rest for now.

 

[Robbo99999]

(there's no point in wriggling) (but I don't mind it being put to rest)