AirPods Max
Jan 11, 2021 at 1:01 PM Post #2,371 of 5,629
In Canada sales tax is Province and Federal based. For example, in Ontario HST is 13%, of which 8% is the Province component and 5% is the Federal component. We’ve got free universal healthcare, though. Not a valid reason, however, for us to blow out our ears with loud headphones. :wink:

That’s nuts! I’m not an American citizen or resident, so my US tax knowledge is extremely limited; I thought it was by state.
 
Jan 11, 2021 at 1:06 PM Post #2,372 of 5,629
How does this feature work exactly?

My phone would probably revert me to 50% volume after about 40-60 minutes of listening at full volume with the built in protect hearing settings that are implemented in the iPhone. 43.75% tends to peak me out around 80-85 dB, but averages around 76 dB (iTunes+ 256 AAC). I normally listen a little quieter with other headphones, but the midrange drop does need that additional 5dB :/
 
Jan 11, 2021 at 1:15 PM Post #2,373 of 5,629
How does this feature work exactly?

The iPhone calculates listening level based on volume you have it set at and what it thinks the volume being played back is at for that given volume level. At times, this calculation is very accurate (AirPods, EarPods, and Beats lines) and in other's it's not. It then uses a lookup table to see what the maximum listening time for that volume level is (either by day or week). It likely uses a weighted average to determine if you've listened too loudly. If it feels you have, it reverts the volume back to something it feels is safe (50% in my case). You can manually override it after this is done though. So basically Apple will tell you it thinks you're blowing out your ears, but won't stop you from doing it (but will try to prevent you from doing it). This data (volume level listening) is available in the Health app of the iPhone.

I've had it invoked once. I had my iPhone connected to the Topping DX3 Pro Bluetooth DAC/amp which was outputting to my HiFiMan HE-560. My iPhone's volume was set to 100% while my amplifier was set to like 30%. Because the iPhone was transmitting at 100% it assumed I was playing at 120 dB. So after 25 minutes it lowered the volume to 50% and gave a warning in notification center about it. My actual listening volume on my HE-560 is about the same as my APM (around 70 dB average, maybe a little less).

The EU spec iPhones also have another volume limiting feature which limits the maximum volume that the iPhone can put out. I believe this can manually be turned off and is defaulted to off in the US.
 
Jan 11, 2021 at 1:16 PM Post #2,374 of 5,629
In Canada sales tax is Province and Federal based. For example, in Ontario HST is 13%, of which 8% is the Province component and 5% is the Federal component. We’ve got free universal healthcare, though. Not a valid reason, however, for us to blow out our ears with loud headphones. :wink:

Cook County (Chicago, IL) has a 10% sales tax, and we don't get universal healthcare :p 7% goes to state and 3% goes to county. Federal taxes are taken directly from your pay check (state tax too). I wouldn't mind paying 13% (maybe higher) if it meant universal healthcare and other benefits.
 
Jan 11, 2021 at 2:53 PM Post #2,375 of 5,629
The iPhone calculates listening level based on volume you have it set at and what it thinks the volume being played back is at for that given volume level. At times, this calculation is very accurate (AirPods, EarPods, and Beats lines) and in other's it's not. It then uses a lookup table to see what the maximum listening time for that volume level is (either by day or week). It likely uses a weighted average to determine if you've listened too loudly. If it feels you have, it reverts the volume back to something it feels is safe (50% in my case). You can manually override it after this is done though. So basically Apple will tell you it thinks you're blowing out your ears, but won't stop you from doing it (but will try to prevent you from doing it). This data (volume level listening) is available in the Health app of the iPhone.

I've had it invoked once. I had my iPhone connected to the Topping DX3 Pro Bluetooth DAC/amp which was outputting to my HiFiMan HE-560. My iPhone's volume was set to 100% while my amplifier was set to like 30%. Because the iPhone was transmitting at 100% it assumed I was playing at 120 dB. So after 25 minutes it lowered the volume to 50% and gave a warning in center about it. My actual listening volume on my HE-560 is about the same as my APM (around 70 dB average, maybe a little less).

The EU spec iPhones also have another volume limiting feature which limits the maximum volume that the iPhone can put out. I believe this can manually be turned off and is defaulted to off in the US.

If someone is annoyed by this notifications, here is a short cut to change it
https://www.reddit.com/r/shortcuts/comments/kbzdi4/workaround_for_the_new_headphone_audio_level/
 
Jan 11, 2021 at 7:19 PM Post #2,376 of 5,629
when you get it check that both ear pads are properly manufactured and remove them to see if they forgot or not to add the acoustic felt sticker that is supposed to cover the driver holes. The first copy I tried had that sticker on one side but not the other.
Thanks. Will check.
 
Jan 11, 2021 at 8:00 PM Post #2,377 of 5,629
Please take a look below at the answer rtings gives to a user asking why they tested the APM with SBC and not with the higher quality AAC:

“Posted 15 hours ago by Shea Angus (RTINGS.com)
Hi StormyGuy!
Thanks for being patient as we worked on this.
We didn’t expect there to be a difference between between SBC or AAC and all of our testing is done using the SBC Codec so we use that for consistency across all of our reviews. That being said, because of the concerns raised here, we retested the AirPods Max with the AAC Codec and the results were the same. We have updated our review to clarify this and show the results of the AAC Codec testing. You can view the graph here NOTE: Ignore anything below the 200-300hz range on this graph, as this was entirely measured on our testing equipment.
We do recognize that different codecs might have an impact on the overall listening experience for some people, but our current test bench doesn’t support that right now & we recognize that this is a downside of our current methodology. That being said, we expect that our results are accurate and should be applicable for people who will be using these headphones.”

and now i just wonder; if they admit by themselves that there is a downside in their methodology, then how accurate all their measurements can be, when most of them are done with SBC? And generally how reliable their whole system can be when there is a downside built into it?

link: https://www.rtings.com/headphones/discussions/QKwnocO6_yTjH8SR/sbc-vs-aac-codec
 
Last edited:
Jan 11, 2021 at 8:08 PM Post #2,378 of 5,629
Please take a look below at the answer rtings gives to a user asking why they tested the APM with SBC and not with the higher quality AAC:

“Posted 15 hours ago by Shea Angus (RTINGS.com)
Hi StormyGuy!
Thanks for being patient as we worked on this.
We didn’t expect there to be a difference between between SBC or AAC and all of our testing is done using the SBC Codec so we use that for consistency across all of our reviews. That being said, because of the concerns raised here, we retested the AirPods Max with the AAC Codec and the results were the same. We have updated our review to clarify this and show the results of the AAC Codec testing. You can view the graph here NOTE: Ignore anything below the 200-300hz range on this graph, as this was entirely measured on our testing equipment.
We do recognize that different codecs might have an impact on the overall listening experience for some people, but our current test bench doesn’t support that right now & we recognize that this is a downside of our current methodology. That being said, we expect that our results are accurate and should be applicable for people who will be using these headphones.”

and now i just wonder; if they admit by themselves that there is a downside in their methodology, then how accurate all their measurements can be, when most of them are done with SBC? And generally how reliable their whole system can be when there is a downside built into it?

link: https://www.rtings.com/headphones/discussions/QKwnocO6_yTjH8SR/sbc-vs-aac-codec

It should also be noted that the Imaging test may not be accurate:

“Unfortunately, we couldn't test the left and right drivers at the same time, as we normally do. These headphones use Apple's Adaptive EQ, which adjusts its sound profile to the fit and seal of the headphones on your head. This feature can't be turned off unless ANC or Ambient mode are turned off as well. We ran the left driver separately from the right driver to test with the ANC on. While we don't know how comparable the results may be due to this change in our methodology, the results of this test are still valid.”

Still, I expected slightly more out of AirPods Max from a technical perspective even if the bass and mid range are practically flawless.
 
Last edited:
Jan 11, 2021 at 8:25 PM Post #2,379 of 5,629
Please take a look below at the answer rtings gives to a user asking why they tested the APM with SBC and not with the higher quality AAC:

“Posted 15 hours ago by Shea Angus (RTINGS.com)
Hi StormyGuy!
Thanks for being patient as we worked on this.
We didn’t expect there to be a difference between between SBC or AAC and all of our testing is done using the SBC Codec so we use that for consistency across all of our reviews. That being said, because of the concerns raised here, we retested the AirPods Max with the AAC Codec and the results were the same. We have updated our review to clarify this and show the results of the AAC Codec testing. You can view the graph here NOTE: Ignore anything below the 200-300hz range on this graph, as this was entirely measured on our testing equipment.
We do recognize that different codecs might have an impact on the overall listening experience for some people, but our current test bench doesn’t support that right now & we recognize that this is a downside of our current methodology. That being said, we expect that our results are accurate and should be applicable for people who will be using these headphones.”

and now i just wonder; if they admit by themselves that there is a downside in their methodology, then how accurate all their measurements can be, when most of them are done with SBC? And generally how reliable their whole system can be when there is a downside built into it?

link: https://www.rtings.com/headphones/discussions/QKwnocO6_yTjH8SR/sbc-vs-aac-codec

Did they not... Let's see, put the headphones on their head and listen? Some people have even seen varying results using Spotify on the phone vs Spotify on a MacBook (both AAC). Going from SBC to AAC should have a pretty decent affect if you're listening to music. A lot of the "sound impressions" they provide on many of their other reviews (and this one) actually have me questioning whether or not they actually take a listen to the product they review or if they try to do a fully "objective" assessment of the headphone.

I will say this though, as far as measurements are concerned, the frequency response of the headphones won't be affected by the codec being used; they're likely right about this. At worse, it might affect distortion readings... I'm still a little confused how they managed to get higher distortion at 90 dB than they did on 100 dB; on top of that they were able to get above the 0.1% that Apple quoted by about 0.5% (though Apple quoted < 0.1% @ 100 dB, they went above that @ 90 dB).
 
Last edited:
Jan 11, 2021 at 8:26 PM Post #2,380 of 5,629
I compared AirPods Max to a few different headphones, and found it interesting that it follows a similar treble curve to Audeze’s LCD-2 Classic.

33486A10-25F4-4388-B3CF-90C2CDCF37FF.jpeg


The LCD-2 Classic (light gray) has an even steeper drop-off in the low-treble.

Had it not been for the lower score in the treble region, AirPods Max would have had a more neutral sound signature than HD 800S. However, the HD 800 S is still stronger in a few key areas including THD.

Even with the poorly-scoring treble range, AirPods Max still delivers a more neutral sound than its closest competition, the Sony XM4 as well as the Bose 700.

Side note: for a more budget-friendly version of AirPods Max with a similar frequency response and features, consumers should look at the Beats Solo Pro.
 
Last edited:
Jan 11, 2021 at 11:00 PM Post #2,382 of 5,629
Side note: for a more budget-friendly version of AirPods Max with a similar frequency response and features, consumers should look at the Beats Solo Pro.

I’ve had both and they don’t sound the same to me at all. Neither sounded to me as recessed in the upper mids / lower trebles (APM) nor as sharp (Beats Solo Pro) as Rtings’ measurements, and both are IMO far more listenable than most other ANC BT headphones I’ve tried, but the APM still sounds to me deficient somewhere around 2000-4000 or so and the Solo Pro excessive. While they may have a similar FR curve below 1000hz, there are IMO far too many differences above to change the overall experience - including how you’ll feel the lower frequencies in relation to the rest.

I also couldn’t make the BT or ANC implementation fail by playing single tones with the APM so far while I noticed some artefacts with the Solo Pro (albeit a lot less than with most other high end BT ANC headphones from other companies, some of which are doing an appalling job in that test).

I really wouldn’t read too much into measurements past 1500hz or so with headphones. Above that I think that for now you should limit yourself to the observation of general tendencies - particularly for the APM where there are significant disagreements across tests - and perhaps people (for example unlike the Shure A50 where the location of the peaks in terms of frequency seems constant across tests, suggesting something intrinsic to their design, for the APM the location of the peaks vary quite a bit, let alone the magnitude, suggesting rather that it’s the interaction of the APM with the features of the test system and methodology that is at play).

I also wouldn’t worry one bit about SBC vs AAC as far as FR curve is concerned.
 
Last edited:
Jan 12, 2021 at 12:07 AM Post #2,383 of 5,629
Please take a look below at the answer rtings gives to a user asking why they tested the APM with SBC and not with the higher quality AAC:

“Posted 15 hours ago by Shea Angus (RTINGS.com)
Hi StormyGuy!
Thanks for being patient as we worked on this.
We didn’t expect there to be a difference between between SBC or AAC and all of our testing is done using the SBC Codec so we use that for consistency across all of our reviews. That being said, because of the concerns raised here, we retested the AirPods Max with the AAC Codec and the results were the same. We have updated our review to clarify this and show the results of the AAC Codec testing. You can view the graph here NOTE: Ignore anything below the 200-300hz range on this graph, as this was entirely measured on our testing equipment.
We do recognize that different codecs might have an impact on the overall listening experience for some people, but our current test bench doesn’t support that right now & we recognize that this is a downside of our current methodology. That being said, we expect that our results are accurate and should be applicable for people who will be using these headphones.”

and now i just wonder; if they admit by themselves that there is a downside in their methodology, then how accurate all their measurements can be, when most of them are done with SBC? And generally how reliable their whole system can be when there is a downside built into it?

link: https://www.rtings.com/headphones/discussions/QKwnocO6_yTjH8SR/sbc-vs-aac-codec
Okay. Thanks for sharing.
Rtings is now one of the sites i dont bother to read anymore.
 
Jan 12, 2021 at 3:38 AM Post #2,385 of 5,629
Finally ! I was waiting to see who would be brave enough to use a SIM card extraction tool to try to remove the headband by shoving it in the tiny hole at the top of the earcup, just above the driver. The authors of the video below were :D. Demonstration here :


So yep, that mesh might be more fragile than your typical headband, but the entire headband can be quickly replaced in a jiffy without opening the cups - unlike any other headphones with a wire through the headband that I know of. That joint mechanism truly is a work of mechanical engineering art.

Now it seems that there is a rubber film in that hole (possibly to provide some form of acoustic isolation) so don't be brutal about it and don't tear it open if you feel adventurous - my guess is that Apple technicians will have a specialised tool to that effect.
The remaining question would be : how much will Apple charge you for a headband replacement ?

FR curves (on a B&K 4128 I believe here ? They also seem to have another HATS available but didn't use it ?), compared to the Bose 700, WH1000XM4 and Huawei Freebuds Studio :

Measurements seem somewhat consistent with the other measurements done on a B&K 4128 Jude posted about in this thread, or Rtings', but inconsistent with measurements made on G.R.A.S. fixtures or the B&K 5128.
Several sound, ANC and mic comparisons throughout the rest of the video.
 

Users who are viewing this thread

Back
Top