• 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!

Apple AirPods Max Review (Noise Cancelling Headphone)

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
Yeah, or when spatializing 3D audio. I think the HRTF personalization also applies to stereo?

My guess is that it's a partial HRTF individualisation that plays a role whenever the spatial audio engine is used.
I don't think that it has any effect on stereo content played without spatialisation enabled, this is something that I think I'll be able to check soon.
 

Mauro

Member
Joined
Nov 23, 2019
Messages
92
Likes
81
Not sure if this has been covered before but there's a very easy way to PEQ APM if you're using iOS.

Go to Settings > Accessibility > Audio/Visual > Headphone Accommodations

Select 'Custom Audio Setup' > Continue > Add Audiogram > Continue > Camera

Now take a picture of anything that *does not* look like a FR graph

Select 'Keep Scan' > Save

Now it will say 'Unable to Import Audiogram'

Select 'Add Values Manually' > set 2k to 5dBHL and 4k to 8dBHL and everything else to zero for both channels

Select Done > Next > Save > Use Audiogram

This will give you a response like (or better than) 'Balanced Tone' without the DRC
I wish that some measures the effect of this. I have made several tries but I am quite confused by the impact of the numbers put as inputs. Maybe @MayaTlab can help?
 

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
I wish that some measures the effect of this. I have made several tries but I am quite confused by the impact of the numbers put as inputs. Maybe @MayaTlab can help?

I've also been quite confused by the effects of the headphones accommodation settings, and I'm not certain that they only impact FR (compression as well ?). I have not investigated it much given that I don't like their effect in the slightest, but Oratory tested the impact of manually entering the audiogram's values via a third party app (not following the instructions above) and got nonsensical results - perhaps PM him on Reddit ?
 

Mauro

Member
Joined
Nov 23, 2019
Messages
92
Likes
81
I've also been quite confused by the effects of the headphones accommodation settings, and I'm not certain that they only impact FR (compression as well ?). I have not investigated it much given that I don't like their effect in the slightest, but Oratory tested the impact of manually entering the audiogram's values via a third party app (not following the instructions above) and got nonsensical results - perhaps PM him on Reddit ?
I can’t find these Oratory measurements but I will try reaching out to him.
Yeah. I like the balanced mode but I guess you are right pointing the finger to compression or some aggressive feedback mechanism in the upper registers. That’s why I was trying to use an audiogram.
 

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
I can’t find these Oratory measurements but I will try reaching out to him.

Not shared publicly on Reddit / Twitter. He spent a lot more time than me on that question so indeed good idea to ask him directly :D.

BTW, Watch OS 9.2 now enables for the APP1 and APM this feature, already available for the APP2 : https://audiosciencereview.com/forum/index.php?threads/airpods-pro-2.37164/post-1330715
No idea whether it is accurate or not...
 
Last edited:

barreleye

Member
Joined
Nov 19, 2022
Messages
6
Likes
4
I also wonder if the tonality changed with firmware updates from whenever this review was done? Or some other sneaky poorly-documented DSP playing tricks on me? (these headphones are a little too "smart" for their own good) I don't put great stock in such memories I remember these sounding considerably darker when I demoed them last year vs the pair I got on black friday, and Amir's boosts to the pinna gain are absolutely unlistenable with my set.
I've had the same experience. I've owned my APM for six months, and I don't use them with my iPhone, but I recently did connect them and updated the firmware. I was two versions behind. From the beginning, I've happily used oratory1990's PEQ with Equalizer APO, a Qudelix-5K, and a miniDSP 2x4 HD for wired listening, but since updating, it's been uncomfortably shrill. I have a playlist for testing purposes, and I found myself actually wincing at times during songs like "Birdland" by Manhattan Transfer and even "Aja" due to the knife-like mid/treble. Even halving oratory's mid/treble boost was still too much, and it was much less than Amir's boosts to begin with. I'm not using EQ at all now. Either my hearing has changed, or the APM sound signature has.
 

barreleye

Member
Joined
Nov 19, 2022
Messages
6
Likes
4
Unlikely out of a firmware update at the least : https://audiosciencereview.com/foru...noise-cancelling-headphone.25609/post-1396880

Perhaps test for positional variation or pad wear ?
Pads are still like new. I'm always skeptical of claims like mine, so I tested this over several sessions over the last couple of weeks, because I really don't want to believe it. I can't imagine positioning of the headphones is a factor. I mean, I'm wearing them like I always have, and the new observations have been consistent. The testing is with the same material and equipment as when I first got them, and I'd like to think I'm pretty rigorous at how I approach it. There are 23 songs in my test playlist from (prog) rock/pop/jazz/acoustic/classical, and there are specific sections in most songs that I pay special attention to. I use the Qudelix-5K and Equalizer APO for gain-matched A/B testing between EQ and flat, because they switch between profiles instantly. The miniDSP has like a 10 second transition, but it sounds the same as the others as best I can tell. Something has changed, I just don't know what. If I've recently become more sensitive to mids/treble, I'm not hearing it in my AirPods Pro 2, Bose QC II, and Sony MDR-1000X headphones.
 

lindijones

Member
Joined
Mar 9, 2023
Messages
9
Likes
3
It is also possible on iOS in the Headphones Accomodations settings to use an audiogram as input for customisations. After selecting you need to upload a picture of your audiogram, but if iOS does not recognise the values you can input them manually. You have to input dBHL values for 125, 250, 500, 1000, 2000, 4000, and 8000 Hz. I wonder if this could be used as a 'poor man's fixed band EQ'. I've played around with this a bit, but was not able to get very satisfactory results. Anybody tried this yet?
Yes. I made an app for creating fully customized audiograms.


As you said it‘s the „poor man‘s eq“ and the only way to get a kind of „flexible eq“ on iPhones. And only on iPhones as you only have HealthKit with audiogram data on iPhones.

Unfortunately you cannot see the resulting filter curves/values which Apple creates according to the used audiogram.
As far as i can say it is not a 1:1 compensation.
Means if you add a (virtual) hearing loss of e.g. 10db at 10kHz then the result is NOT a 10db boost at this frequency.

And as far as i can say it makes sense not to add too many frequency points for compensation.



Play around and let me know!
 

Zensō

Major Contributor
Joined
Mar 11, 2020
Messages
2,740
Likes
6,740
Location
California
I had a pair of AirPods Max when they first came out. The FR was deficient in the ear gain region, but I really liked the clean, low distortion bass. With Oratory’s EQ they sounded good, but I ended up selling them because I was primarily listening on mobile where I didn’t have a convenient way to implement the EQ.

Fast forward to this week and I ended up with another pair. Much to my surprise, these are clearly brighter and more balanced, and when I use the same EQ as before they are far too bright. I don’t know how to explain this other than to think Apple has changed the tuning, most likely through a DSP/firmware update. Has anyone else experienced this recently?
 

oleg87

Senior Member
Joined
Sep 26, 2022
Messages
302
Likes
540
Location
California
I had a pair of AirPods Max when they first came out. The FR was deficient in the ear gain region, but I really liked the clean, low distortion bass. With Oratory’s EQ they sounded good, but I ended up selling them because I was primarily using them on mobile where I didn’t have a convenient way to implement the EQ.

Fast forward to this week and I ended up with another pair. Much to my surprise, these are clearly brighter and more balanced, and when I use the same EQ as before they are far too bright. I don’t know how to explain this other than to think Apple has changed the tuning, most likely through a DSP/firmware update. Has anyone else experienced this recently?
Yes, I had the same experience that I brought up some pages ago. I auditioned these when they first came out and had the same impression about the ear gain.
I bought a pair early this year (and returned them after a week due to comfort issues) and they seemed reasonably balanced tonally with a warm and slightly U-shaped signature, I didn't try Oratory's preset but Amir's EQ preset was absolutely unlistenable (and I'm quite familiar with what the Harman target sounds like on my various headphones).
 

Zensō

Major Contributor
Joined
Mar 11, 2020
Messages
2,740
Likes
6,740
Location
California
Yes, I had the same experience that I brought up some pages ago. I auditioned these when they first came out and had the same impression about the ear gain.
I bought a pair early this year (and returned them after a week due to comfort issues) and they seemed reasonably balanced tonally with a warm and slightly U-shaped signature, I didn't try Oratory's preset but Amir's EQ preset was absolutely unlistenable (and I'm quite familiar with what the Harman target sounds like on my various headphones).
Cool, thanks. I wish Apple would be more transparent about these running changes, but we all know that’s never gonna’ happen.
 

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
I had a pair of AirPods Max when they first came out. The FR was deficient in the ear gain region, but I really liked the clean, low distortion bass. With Oratory’s EQ they sounded good, but I ended up selling them because I was primarily listening on mobile where I didn’t have a convenient way to implement the EQ.

Fast forward to this week and I ended up with another pair. Much to my surprise, these are clearly brighter and more balanced, and when I use the same EQ as before they are far too bright. I don’t know how to explain this other than to think Apple has changed the tuning, most likely through a DSP/firmware update. Has anyone else experienced this recently?

While I haven't measured a pair manufactured in the last several months, by now I've measured a dozen or so in situ with in-ear mics and I don't think that I've found anything that would make me suspect a deliberate change in tuning over time. The most noticeable difference so far that I've noticed has been brand new pads vs. broken in.

With my current pair, successive firmware updates have done nothing to the tuning that I've been able to measure. Once broken in the pads have also remained stable over time.
 

Zensō

Major Contributor
Joined
Mar 11, 2020
Messages
2,740
Likes
6,740
Location
California
While I haven't measured a pair manufactured in the last several months, by now I've measured a dozen or so in situ with in-ear mics and I don't think that I've found anything that would make me suspect a deliberate change in tuning over time. The most noticeable difference so far that I've noticed has been brand new pads vs. broken in.

With my current pair, successive firmware updates have done nothing to the tuning that I've been able to measure. Once broken in the pads have also remained stable over time.
That is very interesting. I’m always one to defer to measurements, but in this case the difference is so dramatic and unexpected it’s hard to believe something hasn’t changed. Regarding pads, have you found they cause the response to get darker or brighter as they break in?
 
Last edited:

GaryH

Major Contributor
Joined
May 12, 2021
Messages
1,348
Likes
1,804
Yep places like Rtings have measured differences after firmware updates. Best to ignore hobbyist DIY in-ear mic measurements (especially in the treble), it's just inaccurate, imprecise pseudoscience really.
 
Last edited:

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
That is very interesting. I’m always one to defer to measurements, but in this case the difference is so dramatic and unexpected it’s hard to believe something hasn’t changed. Regarding pads, have you found they cause the response to get darker or brighter as they break in?

You can see an example of it a few posts earlier, but I would not presume that my experience of pad aging with the APM would be the same for you, and all in all it's not been, comparatively to other headphones, a particularly strong factor.

I’m not surprised that @GaryH mentioned Rtings’ measurements but I think that this is a misinterpretation to deduce from them that a firmware update changed the FR. I guess that he’s referring to this graph :

firmware-3c39-large.jpg


These measurements were performed weeks or months apart and not with the headphones fixed in place on the fixture. Rtings’ usual modus operandi for FR measurements higher than a few hundred Hz is to perform five seatings on their HMS fixture and average the results, but I don't know if the same method was used for the measurements corresponding to the latter firmwares. The AirPods Max having quite poor consistency with that method, it could be a case of averages to averages differences given the magnitudes involved. Or a different operator positioning them differently on the fixture. I don’t know to which firmware the “Sanity check” trace corresponds to, but the fact that it’s different from any of the others should raise questions.

The difference between the curve labelled “3B71” and the one labelled “3C39” is this :

Rtings APM diff.jpg


You can also use a different data point. Oratory released a pdf for the APM quite soon after release, in December 2020. The one currently available was published during February 2022, after accumulating more samples :

Screenshot 2023-07-13 at 21.10.26.png
Screenshot 2023-07-13 at 21.10.13.png


This is how they differ :

Orato APM diff.jpg


While this is a rolling average that will include in the February pdf in the earlier samples from the December pdf, I don’t see here anything that would strike me as evidence that a firmware update or a deliberate change in production happened.

In both cases the differences observed can arise either from the methodology resulting in averages to averages variation, or in the case of Oratory sample variation or an increase in sample size as well.

In situ measurements can be performed in a way that produces less seatings to seatings variation than the sort of variation you see when performing spatial averaging on ear simulators, you can see an example here.
The main drawback from that method being that seatings to seatings variation of the mics themselves can be important. This can be reduced by designing the mics so that they always take the same place, and avoiding comparing measurements using different mic designs or placements.

I don’t want to deny someone else’s experience, far from it, and the differences between the different traces from either Rtings, Oratory or mines are what I’d consider audible, but I’ll reiterate that I don’t think that these variations are evidence that a deliberate change in tuning happened either from a firmware update or a change in manufacturing, and I think that they can be solely explained by other factors first and foremost. If such a firmware update or change in manufacturing happened, it’s small enough to be buried under the noise from averages to averages variation, or sample variation.

I also don’t find my own experience of the APM to align well past 1Hz with what I’d expect given the measurements seen so far, but as far as I’m concerned that was from day one.
 

Attachments

  • Apple AirPods Max.pdf
    415.4 KB · Views: 46
  • Apple AirPods Max.pdf
    417.6 KB · Views: 38

MayaTlab

Addicted to Fun and Learning
Joined
Aug 15, 2020
Messages
924
Likes
1,512
Yep places like Rtings have measured differences after firmware updates.

Nah not really.

Best to ignore hobbyist DIY in-ear mic measurements (especially in the treble), it's just inaccurate, imprecise pseudoscience really.

Is it pseudoscience just because you don’t like it and behave in your usual keyboard warrior fashion, or do you actually have something cogent and intelligent to say ?
 

Zensō

Major Contributor
Joined
Mar 11, 2020
Messages
2,740
Likes
6,740
Location
California
You can see an example of it a few posts earlier, but I would not presume that my experience of pad aging with the APM would be the same for you, and all in all it's not been, comparatively to other headphones, a particularly strong factor.

I’m not surprised that @GaryH mentioned Rtings’ measurements but I think that this is a misinterpretation to deduce from them that a firmware update changed the FR. I guess that he’s referring to this graph :

View attachment 298994

These measurements were performed weeks or months apart and not with the headphones fixed in place on the fixture. Rtings’ usual modus operandi for FR measurements higher than a few hundred Hz is to perform five seatings on their HMS fixture and average the results, but I don't know if the same method was used for the measurements corresponding to the latter firmwares. The AirPods Max having quite poor consistency with that method, it could be a case of averages to averages differences given the magnitudes involved. Or a different operator positioning them differently on the fixture. I don’t know to which firmware the “Sanity check” trace corresponds to, but the fact that it’s different from any of the others should raise questions.

The difference between the curve labelled “3B71” and the one labelled “3C39” is this :

View attachment 299000

You can also use a different data point. Oratory released a pdf for the APM quite soon after release, in December 2020. The one currently available was published during February 2022, after accumulating more samples :

View attachment 299008View attachment 299009

This is how they differ :

View attachment 299013

While this is a rolling average that will include in the February pdf in the earlier samples from the December pdf, I don’t see here anything that would strike me as evidence that a firmware update or a deliberate change in production happened.

In both cases the differences observed can arise either from the methodology resulting in averages to averages variation, or in the case of Oratory sample variation or an increase in sample size as well.

In situ measurements can be performed in a way that produces less seatings to seatings variation than the sort of variation you see when performing spatial averaging on ear simulators, you can see an example here.
The main drawback from that method being that seatings to seatings variation of the mics themselves can be important. This can be reduced by designing the mics so that they always take the same place, and avoiding comparing measurements using different mic designs or placements.

I don’t want to deny someone else’s experience, far from it, and the differences between the different traces from either Rtings, Oratory or mines are what I’d consider audible, but I’ll reiterate that I don’t think that these variations are evidence that a deliberate change in tuning happened either from a firmware update or a change in manufacturing, and I think that they can be solely explained by other factors first and foremost. If such a firmware update or change in manufacturing happened, it’s small enough to be buried under the noise from averages to averages variation, or sample variation.

I also don’t find my own experience of the APM to align well past 1Hz with what I’d expect given the measurements seen so far, but as far as I’m concerned that was from day one.
Thanks for the detailed explanation, it’s much appreciated. Your arguments are convincing, which leaves the question unanswered for me and others who have had similar experiences. I would almost always chalk this up to expectation bias, but in my case there was such a clear lack of expectation, and the difference is so dramatic, that I’m left feeling puzzled...
 

oleg87

Senior Member
Joined
Sep 26, 2022
Messages
302
Likes
540
Location
California
Thanks for the detailed explanation, it’s much appreciated. Your arguments are convincing, which leaves the question unanswered for me and others who have had similar experiences. I would almost always chalk this up to expectation bias, but in my case there was such a clear lack of expectation, and the difference is so dramatic, that I’m left feeling puzzled...
When you're dealing with headphones that are so "smart", trying to quantify all of this is an exercise with a lot of unknowns. We know the Apple stuff does real time auto-EQ, and I doubt anyone except the engineering team at Apple knows the full details about how all that stuff works.
 
Top Bottom