Apple AirPods Max Review (Noise Cancelling Headphone)

[crinacle]

Hey, so coming in to clarify some stuff on request: yes, I use the 43AG-7 which is basically the single channel version of what Amir and oratory1990 uses, but the components are all the same (RA040X coupler, KB500X pinna). The difference between other 43AG users and I is that I've actually built a custom mounting solution for mine to properly space out a headphone like it would on a human head, roughly the same distance as a 45CA so the clamp force should be nearly identical.

I've actually shown what the APM looks like on my rig in my review itself:

And here is another more recent picture of my custom mounting solution with another headphone:

Reason for this design is so that it is collapsible for portable use. Like so:

Hope that clears things up.

 

[Robbo99999]

Hey, so coming in to clarify some stuff on request: yes, I use the 43AG-7 which is basically the single channel version of what Amir and oratory1990 uses, but the components are all the same (RA040X coupler, KB500X pinna). The difference between other 43AG users and I is that I've actually built a custom mounting solution for mine to properly space out a headphone like it would on a human head, roughly the same distance as a 45CA so the clamp force should be nearly identical.

I've actually shown what the APM looks like on my rig in my review itself:

View attachment 146379

And here is another more recent picture of my custom mounting solution with another headphone:

View attachment 146380

Hope that clears things up.

The one variable I've thought of just now as I see your HD580 on the rig, I guess you'd have to decide how much to extend the earcups down from the headphone band, as normally when you put the HD580 or other similar headphones on your head you would pull out the earcups from the headband to allow for enough reach to your ears based on the headband resting on the top of your head. With that measurement device you don't have the head inbetween so in headphone designs like the 580 you'd have to decide how much to extend the driver cups out of the headband, which in turn would effect mounting pressure? I suppose that comes down to the "measurement protocol" variable that I mentioned in an earlier post to another member. So even though these GRAS devices are compatible between each other, there is still "measurement protocol" as a variable of difference between headphone measurements (to mention just one variable, with others being unit to unit variation & pad wear).

 

[GaryH]

Unless I'm mistaken in the ear canal gain region (1-4kHz or so) HPs still operate under "pressure conditions" (I don't know the right terminology, ie basically SPL is the same at any point in space inside the front volume) - that's what that article suggests for example : https://www.aes.org/e-lib/browse.cfm?elib=20605

Pressure chamber conditions

So anything that affects the front volume's, well, volume, may affect the response, and perhaps in the case of the APM disproportionally so in the ear canal gain region. And the APM's headband to cup attachment + spring loaded earcup may add quite a bit of difficulty to the equation. Perhaps the ways pads compress as well : on a rig with a flat plate around the pinna it's probably quite a lot more evenly compressed than on a real human's head (particularly since the APM's pivot + spring cannot by design ensure an even pressure around the ear for most people).

Which means heads of all different shapes and sizes will also affect their response, so any differences in pad compression between measurements actually act as a proxy for this. Either way, the APM's frequency response consistency is poor.

As I said I've measured four APMs, that's eight earcups in total (since they're nearly perfectly front to back symmetrical, unlike other headphones, it's somewhat valid to reverse them and measure them on the same ear), and the response didn't vary that much (if taking into account the typical seatings to seatings variation I see in the ear canal region on my head with the APM there may have been no more than around 1dB of sample variation in that range, with one cup being an outlier and the others even more tightly grouped). Cf graph for four earcups earlier.

Why did you only choose to post 2 units' measurements instead of showing all 4, including the 'outlier'? And no, I don't think it is valid to measure earcups on the wrong ear, because the front volume of the APM is not horizontally symmetrical due to the ANC mic being on one side of the inner, which may result in acoustical differences when switched sides on the same ear. Not to mention doubts over the validity of your probe mic measurements in the first place (located 'somewhere near the DRP' ), which may also affect the acoustic impedance of your ear canal, meaning not only will they be an inaccurate representation of what you hear without the probe, but even relative differences between these probe measurements may be inaccurate too, for the same reason any measurement rig with an acoustic impedance differing from the human ear will result in inaccurate measurements and comparisons, as Oratory explains here.

 

[Maiky76]

This is a review and detailed measurements of the Apple AirPods Max wireless Bluetooth noise cancelling headphone. It is on kind loan from a member. The retail cost is US $549 but I see it on Amazon for $479 including Prime shipping.

I am used to shocking good looks in Apple products but this one with dual kidney shape did not do it for me:

View attachment 146187

The mesh band was stylish though as was the design of the woven pads. What I absolutely loved was the rotary volume control. Shame on all others that make you deal with gestures and such with coarse volume adjustment to boot. I also liked the solid, tactile button for noise cancelling. Alas, that button is way overloaded, requiring you to decode Morse code light at the other end to figure out what you are doing with it. Why not voice feedback? Anyway, you get the job done.

Noise cancelling works well but has that "suction" effect to the max. It literally feels like there is a vacuum cleaner inside this thing at times! Didn't notice it during use though, only when I put it on.

There is adaptive EQ for seal. Pull the cups away a bit and you can instantly hear it boosting the bass. This made measurements a bit tricky but I got through it.

The headphone is a bit on the heavy side:

View attachment 146188

The cups are deep at 30 mm which is nice. The cups are oval with dimensions of 60x43 (heightxdepth).

The hinge pressure is pretty high and combined with the texture of the pads, I feel them on my face.

Back to the unboxing experience, I found the magnetic "bra," I mean soft wrapper a bit odd looking. And the fact that it is the only way to force it into deep sleep a bit odd. Why not add yet another mode to the darn NC button to do that on demand? I know, Apple knows best.

Note: The measurements you are about to see are made using a standardized Gras 45C. Headphone measurements by definition are approximate and variable so don't be surprised if other measurements even if performed with the same fixtures as mine, differ in end results. Protocols vary such as headband pressure and averaging (which I don't do). As you will see, I confirm the approximate accuracy of the measurements using Equalization and listening tests. Ultimately headphone measurements are less exact than speakers mostly in bass and above a few kilohertz so keep that in mind as you read these tests. If you think you have an exact idea of a headphone performance, you are likely wrong!

Fitment on the fixture was somewhat challenging as eluded to earlier. Bass response would have ringing in it depending on how you fitted the headphone. Having to stream content to it over Bluetooth made this fair bit more work than testing wired headphones. Frequency response was the same with or without noise cancelling unlike any other NC headphone I have tested.

Apple AirPods Max Measurements
As usual we start with the headphone frequency response as comparison to our preference curve:

View attachment 146189

That is a first: full compliance with our reference curve all the way to 1 kHz but then the wheels fall off. Why so little amplitude there? This is going to create a dull headphone with little spatial qualities.

View attachment 146190

So bad news on frequency response but check this distortion graph out:

View attachment 146191

Distortion was so low that I provide a zoomed version. Distortion at 94 dBSPL is so low that it is polluted now with some noise causing the wiggliness.

View attachment 146193

So we have plenty of headroom to EQ. BTW, the max volume as about 113 dBSPL. It would not go any higher which is fine.

Apple AirPods Max Listening Tests and Equalization
My main listening station is a Windows desktop. I have not used Bluetooth in it and when I turned it on, it was quite flakey due to lack of external antenna. So I dragged an external no-name bluetooth transmitter. That worked but volume was very low. The volume control on AirPods was no longer active either. I switched to a Fiio BT transmitter which has its own volume control. That gave me a bit more gain but still nothing like what I could get out of my Android phone. Speaking of Android phone, my Roon player repeatedly hung when playing tracks to AirPods Max. This seemed to be related to high sample rate/bit depth content. Reducing everything to 16 bits, 44.1 kHz helped but not completely.

I used some digital gain to get reasonable volume so that I could develop my EQ and arrived at this simple two filter setup:

View attachment 146198

The sound opened up and clarity much improved. Spatial qualities went from nothing to good. I listened for a few tracks and then turned off the EQ. Man, it sounded like someone had turned off the tweeter and maybe even the midrange on a speaker! It was far more drastic going from EQ on to off, making you realize how bad the stock tuning really is.

Conclusions
Apple with more R&D and manufacturing resources than any headphone company gets a few things really right such as extra low distortion and excellent bass response. Alas, they either listened to Beats people, marketing or who knows who to through out all that energy in the 1 to 7 kHz. Don't they know our ears naturally amplify that region for a reason? What book did they read that said you want to have flat response there? I am amazed how many online reviewers said this thing sounds great. Really? What is your reference? The old wired home phones?

Without EQ, the Apple AirPods is one terrible sounding headphone. As much as I like bass, I hate it without balancing high frequencies. I paid for the full spectrum of the music I consume, not just the bass. So please give it all to me. With EQ, the sound becomes very good prompting me to recommend it such. How good, I can't quite tell because I can't get enough volume on Windows. It can range from good to great.

----------
As always, questions, comments, recommendations, etc. are welcome.

Have 30 pounds of cucumbers from the garden to can and here I am testing headphones! Talk about priorities.....

Any donations are much appreciated using: https://www.audiosciencereview.com/forum/index.php?threads/how-to-support-audio-science-review.8150/

Hi

Here are some thoughts about the EQ.

Notes about the EQ design:

• The average L/R is used to calculate the score.
• The resolution is 12 points per octave interpolated from the raw data (provided by @amirm)
• A Genetic Algorithm is used to optimize the EQ.
• The EQ Score is designed to MAXIMIZE the Score WHILE fitting the Harman target curve with a fixed complexity.
  This will avoid weird results if one only optimizes for the Score.
  It will probably flatten the Error regression doing so, the tonal balance should be more neutral.
• The EQs are starting point and may require tuning (certainly at LF).
• The range around and above 10kHz is usually not EQed unless smooth enough to do so.
• I am using PEQ (PK) as from my experience the definition is more consistent across different DSP/platform implementations than shelves.
• With some HP/amp combo the boosts and preamp gain need to be carefully considered to avoid issues
• Not all units of the same products are made equal. The EQ is based on the measurements of a single unit.
• YMMV with regards to the very unit you are trying this EQ on.

Incredible L/R match.

I have generated one EQ, the APO config file is attached.

Score no EQ: 39.4
Score Armirm: 67.8
Score with EQ: 85.1

Apple Airpods Max APO EQ Flat@HF 96000Hz
August092021-100955

Preamp: -9.3 dB

Filter 1: ON PK Fc 2236.11 Hz Gain 4.95 dB Q 0.91
Filter 2: ON PK Fc 4196.10 Hz Gain 7.00 dB Q 1.89
Filter 3: ON PK Fc 7352.39 Hz Gain 6.93 dB Q 3.79
Filter 4: ON PK Fc 13267.20 Hz Gain -4.66 dB Q 3.47

 

[ouimetnick]

Haven’t read through all 7 pages yet, but these headphones have an adaptive EQ that changes based on the source material. I tried them out and liked them. I should note that I have the AirPods Pro as well. I wear hearing aids so I take those out when using the AirPods Pro (the Pro stay in my ears, the basic AirPods do not. I don’t use the ANC or Transparency benefits of the AirPods Pro).

The AirPods Max were open box from Best Buy for $499 and I returned them because I felt that was too expensive. This was back in January, so now that you can buy them brand new for $479 is nice. I’ll wait until Black Friday and hope I can snag them for even less. They are so bulky and heavy that they are only good if you are relatively still. I wear my AirPods Pro at the gym and sweat all over them. I couldn’t do that with $500 headphones. Should point out I paid $220 for my AirPods Pro in March 2020, and snagged another pair for a family member for $189ish recently from Amazon.

 

[Tom C]

On iOS, using “Headphone Accommodations” can fairly dramatically improve the sound quality of AirPods Max. The “Balanced Tone” setting at the “Slight” level boosts levels above 2 kHz and really wakes up the headphones. More here: https://9to5mac.com/2021/01/28/how-to-customize-iphone-headphone-audio/

Thank you! That helped a lot! Amir’s right. Sounds much better with a treble boost, and when you turn the compensation off, it really sounds dull.
I still prefer the sound of my JBL in-ear Club Pro though.

 

[MayaTlab]

Pressure chamber conditions

Which means heads of all different shapes and sizes will also affect their response, so any differences in pad compression between measurements actually act as a proxy for this. Either way, the APM's frequency response consistency is poor.

That's indeed something that I'd find interesting to learn (and why I'd love to see more measurements done on real humans as Rtings does it at lower frequencies). What I notice is that at least as far as I'm concerned the APM is nowhere near as inconsistent from individual traces to individual traces in the ear canal gain region (not above 4-5kHz though), even when I deliberately make no efforts whatsoever when I put them on.
It's more aligned with Jude's impressions here : https://www.head-fi.org/threads/airpods-max.949152/post-16091082

Why did you only choose to post 2 units' measurements instead of showing all 4, including the 'outlier'?

Because I prefer to stick to the ones measured during the same session and they were measured by two. You actually get to see the outlier (the two traces slightly below the others in the ear canal gain region), and to some degree the seatings to seatings variation (which is not insignificant past 5kHz or so) with the two averages (the one lonely trace at 5kHz). I didn't keep the individual traces.
The other sessions are similar.

And no, I don't think it is valid to measure earcups on the wrong ear, because the front volume of the APM is not horizontally symmetrical due to the ANC mic being on one side of the inner, which may result in acoustical differences when switched sides on the same ear.

It is in the dead centre, not to one side :

The only asymmetric parts are the vents' exits (the two narrow slots on the outside in one corner for both the front and rear volume), the window for the optical sensor, and the L/R markings for the pads.

Not to mention doubts over the validity of your probe mic measurements in the first place (located 'somewhere near the DRP' ), which may also affect the acoustic impedance of your ear canal, meaning not only will they be an inaccurate representation of what you hear without the probe, but even relative differences between these probe measurements may be inaccurate too, for the same reason any measurement rig with an acoustic impedance differing from the human ear will result in inaccurate measurements and comparisons, as Oratory explains here.

It's always been a source of concern indeed and I had read that post a long while ago .
That's exactly why I run over and over graphs like these, whereby one session's relative measurements are plotted against another, which shows you how they differ in absolute or relative terms :

Or these, where the difference between HPs and a reference HP are plotted against each other, and for different types of mics to know how they differ (here two probes sessions and two blocked ear canal sessions, for example) :

Or these, where closed ear canal (blocked ear canal entrance mic, in concha mic with clocked canal) and open ear canal measurements (in concha mic with open canal, probe) are compared, during the same day :

To assess how much confidence I can have in them indeed.
I'd like to investigate a bit more what happens between 2 and 4kHz indeed but in general the open canal measurements tend to be the closest to each others vs the blocked canal ones and veer away from the latter in the same direction.
Oh, and I went to see an audiologist to get a sense of the intertragic notch / DRP distance BTW .

Besides it's one of those things where you can just listen to it in edge cases given the degree. Applying "counter EQ" to match a target according to a particular measurement of the APM makes it plainly obvious that I'm not getting results that match superbly well most of the APM's measurements (applying 10dB at 4800Hz as Rtings would suggest is murder for my ears and results in a very audible peak running sweeps). We're not talking about 1-2dB here, we're talking about 4, 5 - heck even 8-10dBs at times.

So that's the thing that puzzles me a little bit : whether in terms of seatings to seatings on the same test rig, or in terms of differences between measurements, it varies by quite a good deal more than what I can personally experience either in terms of individual seatings to seatings on my head, or sample variation. I find the situation @eddantes highlighted (https://www.audiosciencereview.com/...cancelling-headphone.25609/page-6#post-873495) unsatisfying.

 

[mononoaware]

I don't just perceive it, I measure it

Ok sorry I must have misunderstood what you meant in your previous comment.

With headphones such as the HD 650, HD560S, Sundara, etc. in general the available measurements gave me the "gist" of what I was about to experience listening to them, but the APM (and more recently the Hi-X65) are headphones for which that didn't quite work for me.

Yes I guess it comes down to predictable consistency. . .

 

[the_brunx]

There are fake AirPods Max going around too

 

[Robbo99999]

Hi

Here are some thoughts about the EQ.

Notes about the EQ design:

• The average L/R is used to calculate the score.
• The resolution is 12 points per octave interpolated from the raw data (provided by @amirm)
• A Genetic Algorithm is used to optimize the EQ.
• The EQ Score is designed to MAXIMIZE the Score WHILE fitting the Harman target curve with a fixed complexity.
  This will avoid weird results if one only optimizes for the Score.
  It will probably flatten the Error regression doing so, the tonal balance should be more neutral.
• The EQs are starting point and may require tuning (certainly at LF).
• The range around and above 10kHz is usually not EQed unless smooth enough to do so.
• I am using PEQ (PK) as from my experience the definition is more consistent across different DSP/platform implementations than shelves.
• With some HP/amp combo the boosts and preamp gain need to be carefully considered to avoid issues
• Not all units of the same products are made equal. The EQ is based on the measurements of a single unit.
• YMMV with regards to the very unit you are trying this EQ on.

Incredible L/R match.

I have generated one EQ, the APO config file is attached.

Score no EQ: 39.4
Score Armirm: 67.8
Score with EQ: 85.1

Apple Airpods Max APO EQ Flat@HF 96000Hz
August092021-100955

Preamp: -9.3 dB

Filter 1: ON PK Fc 2236.11 Hz Gain 4.95 dB Q 0.91
Filter 2: ON PK Fc 4196.10 Hz Gain 7.00 dB Q 1.89
Filter 3: ON PK Fc 7352.39 Hz Gain 6.93 dB Q 3.79
Filter 4: ON PK Fc 13267.20 Hz Gain -4.66 dB Q 3.47

View attachment 146418

Hi, I wish you'd make super super clear what Target Curve you're using in your EQ's, because most people are gonna assume it's the Harman Curve, but I notice you're not using the Harman Curve.......you're using instead something you've created & named yourself as the Amirm Curve, which has less treble and/or more bass than the proper Harman Curve.......so I think you actually need to say that you're not using the Harman Curve rather than just sneaking in a little reference to Amirm Curve. Where did you even get this Amirm Curve? It's not a target curve that Amir has created because his Target Curve he uses in his diagrams is the Harman Curve. (And sometimes in previous reviews you use instead the proper Harman Curve for your EQ's, so there's no consistency - people don't know what they're getting, and they probably don't even know that sometimes you're using a different Target Curve.....and people should know this if they're gonna be using your EQ's).

On a separate topic (which is not a "complaint"), what is meant by a Genetic Algorithm that you mention?

 

[restorer-john]

There are fake AirPods Max going around too

Heaps of them. All over Ali.

https://www.aliexpress.com/item/1005003013885637.html?spm=a2g0o.detail.1000014.24.624e7d75pwCa8O&gps-id=pcDetailBottomMoreOtherSeller&scm=1007.13338.183346.0&scm_id=1007.13338.183346.0&scm-url=1007.13338.183346.0&pvid=684de6d7-1be4-4497-8743-970141aec592&_t=gps-idcDetailBottomMoreOtherSeller,scm-url:1007.13338.183346.0,pvid:684de6d7-1be4-4497-8743-970141aec592,tpp_buckets:668#0#131923#76_668#888#3325#11_3338#0#183346#0_3338#3142#9890#8_668#2846#8115#2000_668#5811#27177#30_668#6421#30830#840_668#2717#7566#841_668#1000022185#1000066058#0_668#3422#30698#883&&pdp_ext_f={"scene":"23416"}

 

[amirm]

Heaps of them. All over Ali.

Amazing. I have to assume it is the same factory apple is using producing these after dark....

 

[Ein]

t doesn't make sense to test Apple products using Android and Windows. Sea fish will not swim in fresh water

Mhm, it works with isomething, macs (now ATV too). Thats it. With Win or Android you have blown egg. It works really nice in ecosystem, you can "trim" cans for your pleasure, its little bassy - yes - but with apple software and eq (not only hardware, but also software, audio software is crucial! Best option Apple Music here) APM works fine. Of course ANC on/ very DSPlish sound here...

You tested it wrong Amirm ;-)

 

[amirm]

The manual clearly says you can use it with non-Apple products:

Until they change that, that is how I treat it.

BTW, I forgot to gripe about the readability of the manual in the review. Why on earth do they use that gray, low contrast font for the manual? My eyes were hurting trying to read the stupid thing. Of all companies, I expect Apple to get this right. Instead it seems they went for style as opposed to functionality.

 

[MayaTlab]

it seems they went for style as opposed to functionality.

I'm tempted to think that the manual isn't the only aspect of the APM for which that applies.

With Win or Android you have blown egg.

As Crinacle showed it has no significant bearings on the measurements.
You can't update the firmware though, but so far I don't think I've seen any report that a firmware update for the APM changed the sound.

 

[restorer-john]

BTW, I forgot to gripe about the readability of the manual in the review. Why on earth do they use that gray, low contrast font for the manual? My eyes were hurting trying to read the stupid thing. Of all companies, I expect Apple to get this right. Instead it seems they went for style as opposed to functionality.

Low contrast writing on stuff really gets me too. White on anodized silver is one of the worst.

Even ingredients on food packaging- light yellow on yellow, light blue on blue. How about black on white? It works for books.

 

[Ein]

The manual clearly says you can use it with non-Apple products:

View attachment 146498


Until they change that, that is how I treat it.

BTW, I forgot to gripe about the readability of the manual in the review. Why on earth do they use that gray, low contrast font for the manual? My eyes were hurting trying to read the stupid thing. Of all companies, I expect Apple to get this right. Instead it seems they went for style as opposed to functionality.

Typical Apple. You read it wrong Amirm ;-)

Of course you can use it with "other" but you know... it's low level experience. Buy Apple! All magical things works here in ecosystem. Only! Dolby/Spatial (yea Android now have apple av software but nothing change for APP/APM here), full support of DSP=ANC/transparent mode, EQ support, AI, upgrades - they do some works for sound etc etc. And one more thing - they measure 24/7 (if you have them on your head ), measure users ears, measure environment, audio material depends... so I think that you use their potential for 20% or less when connect to other devices. Apple do this for PERSONAL experience. This is so special and original but not good for precise review.

 

[restorer-john]

Typical Apple. You read it wrong Amirm ;-)

@amirm held his iPhone 4 the wrong way back in 2010.

 

[dwaveIX]

Spot on review. The lack of meaningful mid range and treble destroys what could be a decent if expensive bluetooth headphone. Unlike their main competitors you can't run them without internal power. Using the (custom) 3.5mm lead connected to a DAC doesn't give any improvement to the SQ.

 

[Maiky76]

Hi, I wish you'd make super super clear what Target Curve you're using in your EQ's, because most people are gonna assume it's the Harman Curve, but I notice you're not using the Harman Curve.......you're using instead something you've created & named yourself as the Amirm Curve, which has less treble and/or more bass than the proper Harman Curve.......so I think you actually need to say that you're not using the Harman Curve rather than just sneaking in a little reference to Amirm Curve. Where did you even get this Amirm Curve? It's not a target curve that Amir has created because his Target Curve he uses in his diagrams is the Harman Curve. (And sometimes in previous reviews you use instead the proper Harman Curve for your EQ's, so there's no consistency - people don't know what they're getting, and they probably don't even know that sometimes you're using a different Target Curve.....and people should know this if they're gonna be using your EQ's).

On a separate topic (which is not a "complaint"), what is meant by a Genetic Algorithm that you mention?

Hi,

I just double checked, I used the Standard Harman target. What makes you think it is not?
When I don't, I usually make it clear.
Amirm preferences are roughy +3dB at LF -1.5dB at HF.

For GA very good explanation, although a bit involved if you are not familiar with optimization:
https://en.wikipedia.org/wiki/Genetic_algorithm