Apple's AirPods Max 2 are finally here, and after spending time with them, I can say that this release is the definition of an iterative one, rather than a meaningful rework.

Since my last video on the original AirPods Max, we’ve adopted new testing methods that let us see how headphones behave on real human heads, not just standardized rigs. This will allow us to do a bit less head-scratching about how these devices actually behave on a person.

Before we get to that though, allow me to re-emphasize: the AirPods Max 2 doesn’t reinvent anything. Apple advertises a list of "new" improvements that came from the AirPods lineup offerings prior—Conversational Awareness, Voice Isolation, Adaptive Audio, Live Translation—but not much of it has anything to do with sound quality. They barely mention changes to sound at all, and even when they do, it’s vague enough to dismiss without issue. 

However, there is actually one big change that everyone seems to have missed. We'll get to that.

For context though, the original AirPods Max followed a predictable consumer-friendly tuning with a reasonably elevated bass response, elevated treble, and a mostly clean (if slightly muffled) midrange. The big dealbreaker for me (and others) was a slightly choppy transition between mids and treble that often caused the spicy upper treble to come across as even more spicy.

The AirPods Max 2 keeps that same general shaping to the frequency response, but before we talk more about the sound, we should touch on the one thing Apple did here that is a genuinely very advanced piece of engineering, new for the AirPods Max lineup: they’ve dramatically improved consistency in the upper midrange across different placement scenarios, which means their entire frequency response from 20-3000 Hz will be much more similar on my head, and Griffin's head, and Cameron's head, and the measurement rigs we use, than the original AirPods Max would've been.

And you might think, "well why does that matter?"

Well that's a fair point! There's a reason why Apple hasn't included this in any of their marketing; its not immediately obvious to consumers why this matters. 

But if you're a headphone nerd, you may know that the variability (specifically in the upper midrange) of what ANC headphones will produce in a given circumstance tends to be incredibly extreme from one position—or one person—to the next.

Fixing this, in our opinion, represents a crucial step forward for Apple's headphones in this class, finally putting them on parity with (or perhaps even better than) Bose in this realm. Make no mistake, this is not only very cool, but makes it one of the only closed back headphones on the planet that can actually deliver virtually the exact same sound to everyone for more than 2/3rds of the frequency response.

Why? Well, if we're only focusing on similar closed back headphones—ones that have ANC systems/microphones—most of the feedback systems used in ANC/transparency-enabled devices don't have the engineering refinement/processing power (or both) to adequately control frequencies above around ~800 Hz on a moment-to-moment basis.

This means the 800-2 kHz region in particular can often have massive differences of behavior from seating to seating on lesser ANC headphones; things as simple as different amounts of clicks on the headband on a single person, or different head widths across multiple people, can actually produce markedly different responses. 

But Apple has seemingly figured out how to near-entirely fix this issue in the 800-2000 Hz range.

Let me show you one example: If you look at the measurement below around 1 kHz-4 kHz versus AirPods Max OG, the new one (blue) exhibits reduced placement variation versus the original (orange).

But also, if you compare the AirPods Max 2 to the original model (measured on two different fixtures), you'll see a much bigger difference between the responses on two fixtures for the old AirPods Max—which seems to have its feedback mechanism optimized until around 800 Hz, where it stops being able to consistently produce the same frequency response—than we see for the new AirPods Max 2, which is consistent up to the aforementioned 3 kHz point.

The older GRAS fixture produces a big difference in the 800-3000 Hz range between these two heads... so do they really sound this different? Nope.

The newer 5128 fixture shows the headphone a more human-like ear load, which causes it to display the behavior more accurately. This shows that what we're seeing here is not actually a big tuning difference, but instead a difference in how consistently these tunings can be achieved on different ears.

Moving on from what I'd say is the only "actually very cool and interesting tech upgrade, done in a way only people with Apple-level R&D Budgets can achieve" here, we can see there are minor differences with the newly-released version: seemingly, a smoother mids-to-treble transition, and slightly reduced treble aggression. But they're not really that different.

And that trend continues throughout most of the sound quality-relevant things here. Apple’s usual dynamic volume-dependent EQ is still present, so while louder listening levels shift bass and treble behavior downward, at the volumes most people use, the overall sound is still likely to be heard as somewhat V-shaped. 

Now, to capture the behavior of the AirPods Max 2 on my head as accurately as possible, I measured them on my actual head using in-ear microphones. 

Apple’s engineers are undoubtedly iterating based on data measured on real human heads as well, so this perspective is vital... and also unique to us: you will not find data like this anywhere else, because Apple doesn't publish theirs.

On my head, the general shape is very similar to what we see on the rigs: elevated bass, elevated treble, and a sensible midrange tuning. But the specific treble features differ dramatically from what you see on a measurement rig. The 5 kHz dip becomes more of a narrow peak followed by a dip, and the highest treble arguably becomes even spicier than the rig implies in some parts.

Glasses would typically introduce leakage, but because Apple's feedback systems used in the ANC/transparency mode have leakage compensation as an inherent benefit (most ANC systems do this, error correction tends to correct for errors), the response stays remarkably consistent up to about 2 kHz. Even the remaining differences could easily come down to very small shifts in position or seal.

Conclusion

In short, the AirPods Max 2 are not a radical rethinking of Apple’s over-ear platform. They are a subtle refinement with better consistency, slightly smoother transitions in key regions, but the same core tuning philosophy and the same vexing industrial design.

Unfortunately, the bigger story here is likely to be what Apple didn't change: most of it, and that's a shame. People who hate the headband, the case, the lack of a power button, the lack of foldability, or the stock tuning approach they're going for should just look elsewhere.

If you for some reason value ANC/transparency above all else and use Apple devices exclusively, sure, this is a better headphone overall than the AirPods Max 1, but its basically just one or two things about the sound quality and features that makes me say so.

Otherwise, it's not really exceptional or new in any meaningful sense... unless you're a hyper-nerd who cares about jargony alphabet soup like "rHpTF consistency in the upper midrange of the audio band'.

Sound Score: 5.1/10

Overall Score: 4.7/10