Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

tripleh3lix

macrumors 6502a
Original poster
Jun 17, 2014
564
375
For any music quality increase; if the dongle and internal codec for the AirPods Max limit the ability to play higher music quality?
 
No, it will not allow the AirPod Max to play lossless. The wire transcodes the analog signal to AAC so the headphones can play them. APMs don't support lossless and they can't be made to do that.

The good news is that AAC 256 VBR is audibly transparent. You can't tell the difference between that and lossless or HD quality using human ears. All the music in the Apple store sounds exactly the same.

The VP that oversees Apple Music was interviewed in Billboard recently and he said that Apple isn't focused on supporting lossless. They will give it to people if they ask for it, but since it sounds exactly the same, there is no point expanding the data stream just for that. He says Apple is more focused on Spatial Audio because that is an area where sound can actually be improved, not just made inefficient for no real benefit.
 
No, it will not allow the AirPod Max to play lossless. The wire transcodes the analog signal to AAC so the headphones can play them. APMs don't support lossless and they can't be made to do that.

The good news is that AAC 256 VBR is audibly transparent. You can't tell the difference between that and lossless or HD quality using human ears. All the music in the Apple store sounds exactly the same.

The VP that oversees Apple Music was interviewed in Billboard recently and he said that Apple isn't focused on supporting lossless. They will give it to people if they ask for it, but since it sounds exactly the same, there is no point expanding the data stream just for that. He says Apple is more focused on Spatial Audio because that is an area where sound can actually be improved, not just made inefficient for no real benefit.
I'm a life long Apple user, but I'd take Cue's dismissal of lossless with a grain of salt. He also said the reason Apple didn't support Blu Ray was because Apple's HD standard was good enough. I believe he also crowed about removing ports from the 2016 MB.

You can definitely tell the difference in rock music between 256 ACC and lossless/AIFF. Digital stuff recorded in 16 bit Garageband? Not so much.
 
There have been dozens if not hundreds of controlled listening tests to discern differences between high data rate lossy and lossless. Overwhelmingly, they show that human ears cannot hear a difference.

If you think you hear a difference, it is likely due to one of three things...

Different mastering: You have to compare apples to apples. Apple Music may have a different master than Spotify or your particular CD. The way to eliminate this is to down sample a single track to the various codecs you're comparing and not assume that if it's the same song from the same album it will sound the same.

Expectation bias: Every human being that lives and breathes is subject to bias. It is how we make decisions in everyday life. If we stopped and gathered info and formed theories and tested them for each and every decision we make, we would never get out of bed in the morning. Bias is hard wired into us. It isn't conscious. The way to minimize it is to conduct a blind test.

Perceptual error: When given two similar audio samples to compare, humans will perceive the one that is slightly louder to be of higher quality, even if the two samples are identical in every other way besides the volume. A difference as little as 1dB can cause perceptual error to skew results. The solution to this is to precisely level match the samples by normalizing or calibrating with tones. Also, Auditory memory is notoriously short. If you listen to two similar sounds with as little as a three second gap between them, it can be impossible to consistently discern the difference. The solution to this is to do a direct A/B switched comparison.

If you can consistently discern the difference between lossless and AAC 256 VBR with identical mastering in a blind, level matched, direct A/B switched controlled test, I would suggest that you immediately contact Apple and submit yourself to prove your claim. They would be very interested in finding out what you are able to discern, because you would be one listener in a million. They can improve the AAC codec if they know what a person with extremely non-standard hearing is cueing into.

Lossless is not the direction to go if you want to audible improve sound quality. Apple is right. DSPs are the future of home audio.
 
There have been dozens if not hundreds of controlled listening tests to discern differences between high data rate lossy and lossless. Overwhelmingly, they show that human ears cannot hear a difference.

If you think you hear a difference, it is likely due to one of three things...

Different mastering: You have to compare apples to apples. Apple Music may have a different master than Spotify or your particular CD. The way to eliminate this is to down sample a single track to the various codecs you're comparing and not assume that if it's the same song from the same album it will sound the same.

Expectation bias: Every human being that lives and breathes is subject to bias. It is how we make decisions in everyday life. If we stopped and gathered info and formed theories and tested them for each and every decision we make, we would never get out of bed in the morning. Bias is hard wired into us. It isn't conscious. The way to minimize it is to conduct a blind test.

Perceptual error: When given two similar audio samples to compare, humans will perceive the one that is slightly louder to be of higher quality, even if the two samples are identical in every other way besides the volume. A difference as little as 1dB can cause perceptual error to skew results. The solution to this is to precisely level match the samples by normalizing or calibrating with tones. Also, Auditory memory is notoriously short. If you listen to two similar sounds with as little as a three second gap between them, it can be impossible to consistently discern the difference. The solution to this is to do a direct A/B switched comparison.

If you can consistently discern the difference between lossless and AAC 256 VBR with identical mastering in a blind, level matched, direct A/B switched controlled test, I would suggest that you immediately contact Apple and submit yourself to prove your claim. They would be very interested in finding out what you are able to discern, because you would be one listener in a million. They can improve the AAC codec if they know what a person with extremely non-standard hearing is cueing into.

Lossless is not the direction to go if you want to audible improve sound quality. Apple is right. DSPs are the future of home audio.
I've passed those tests. But then again, I listen to music on this, so what do I know? Still don't know why some people just can't accept there is a difference and some people are willing to pay a premium. Apple has allowed people to rip in ALAC/AIFF since iTunes came out. Besides, what's the better look for Apple re: lossless: (1) we support superior formats for people with the tastes and equipment to enjoy them or (2) we just did it for a gimmick but get excited anyway?
 

Attachments

  • D1E3FC79-0060-4B13-8767-675CB208B841.jpeg
    D1E3FC79-0060-4B13-8767-675CB208B841.jpeg
    293.3 KB · Views: 108
  • Like
Reactions: rumz
AM is in a tough spot because while they know their competitors are now offering lossless, Apple has no inherent advantage with that because lossless is lossless, generally speaking. Spatial/Atmos is how they are differentiating AM from other services, in an attempt to grow it. Whether that will work, time will tell.

As The Verge and other outposts have noted, the benefits of spatial on catalog music recordings in AM have been widely varied and uneven in quality. Given the furor from legacy artists over their perceived financial misfortune at the hands of streaming services (see all the rights deals theyre all signing up for with Irving Azoff and et al), I don't see many of them going out of their way to go back to master tapes and remix for AM's spatial feature properly.
 
There have been dozens if not hundreds of controlled listening tests to discern differences between high data rate lossy and lossless. Overwhelmingly, they show that human ears cannot hear a difference.
This is surely the case for pop, rock, hip hop, and most other forms of radio-friendly music. But in more complex compositions, such as classical/romantic, jazz, and electronic music the differences in separation and fullness is very apparent.
 
  • Like
Reactions: ntlman and rumz
I've passed those tests.
Cool! I'm always interested in comparing notes with other people who set up controlled listening tests. How did you convert your test files, and what codecs and data rates did you use? How did you do your level matching? How many trials did you do and what was your percentage of right answers to wrong at different data rates?
 
This is surely the case for pop, rock, hip hop, and most other forms of radio-friendly music. But in more complex compositions, such as classical/romantic, jazz, and electronic music the differences in separation and fullness is very apparent.
Separation isn't an issue because codecs haven't done joint stereo since the days of 128 MP3s long ago. And fullness is frequency response which isn't affected above 192. Complex music doesn't mean complex to compress. The hardest thing for codecs to compress without artifacting is massed sounds like choruses, crowds applauding and random noise like the surface noise in a 78rpm record.

I have a test that I put together where I took ten samples of choral and orchestral music and encoded them at 192, 256 and 320 in three different codecs, AAC, MP3 LAME and Frauenhofer MP3. The tenth sample is lossless. The object is to take the ten samples and listen carefully and rank them from best sounding to worst. When you do this, you can find the approximate point where compressed audio becomes transparent and indistinguishable from lossless. If you or any other readers would like to try it, let me know and I will send you either a lossless ALAC or a FLAC file with the samples contained in it. You rank the ten samples and send me back your list and I'll let you know how you did. It's an eye opening (and ear opening!) experience. You'll definitely learn a lot. Just let me know.
 
  • Like
Reactions: motulist
With headphones, that would be improved with equalization to eliminate masking issues in the response. Distortion is rarely an issue with headphones.
 
Last edited:
No, it will not allow the AirPod Max to play lossless. The wire transcodes the analog signal to AAC so the headphones can play them. APMs don't support lossless and they can't be made to do that.
I keep hearing people say that they’re “told” the APM’s DAC can only accept AAC. So apparently by extension the 3.5mm cable is encoding an analog signal to AAC? It doesn’t make sense to me but a) I could easily be wrong and b) I wouldn’t put it past Apple to do something like that, I guess.

I know the only way you get sound out of the APM is through feeding its DSP a digital signal and so that analog connection needs to be converted to 1’s and 0’s. Encoding it to AAC just seems unnecessary extra work (and unnecessary lossy compression). Maybe this 3.5mm cable was an afterthought? Or there’s something about the design / signal path in the APM that makes this compromise worth it? Still sounds odd to me. But since these were explicitly designed as wireless headphones, it’s possible 🤷🏻‍♂️

(don’t mind me. I just can’t wrap my head around this one, even though I’d not be likely to use these in wired mode anyways— I bought them to be wireless.)
 
I tried feeding it high resolution lossless files through bluetooth using Amazon Unlimited. The iPhone converted it to 44.1/48. The phone won't hand the cans a file it can't play. And yes, the cable has a ADC built into it that converts analog to a codec the APMs accept. It's wee wee tiny and in the plug. That's why it costs $30 smackers. The cable was definitely an afterthought. Apple was just giving people a reason not to complain. It doesn't matter because it's all audibly transparent anyway.

By the way, the APMs should be able to handle any codec that is suited for bluetooth, not just AAC. AAC is just the best one and the one Apple favors.
 
I tried feeding it high resolution lossless files using Amazon Unlimited. The iPhone converted it to 44.1/48. The phone won't hand the cans a file it can't play. And yes, the cable has a ADC built into it that converts analog to a codec the APMs accept. It's wee wee tiny and in the plug. That's why it costs $30 smackers.

By the way, the APMs should be able to handle any codec that is suited for bluetooth, not just AAC. AAC is just the best one.
Sure— even a MacBook Pro does the same thing unless you connect an external DAC capable of handling high res audio.

thanks — I learn something every time I engage in these threads. I often think of only compressed audio having a codec, but all digital audio has some kind of codec, right? Such as linear PCM for a non-lossy, “uncompressed” codec. So when Apple’s support page says the APM can’t do lossless with the 3.5mm cable it’s not simply due to an analog to digital conversion happening in the cable, it’s also that the codec used / output by that ADC is also a lossy codec negating any advantage of a lossless analog source, right?

Again… makes sense since these were clearly not built to ever need anything more than lossy BT-appropriate codecs.
 
I often think of only compressed audio having a codec, but all digital audio has some kind of codec, right? Such as linear PCM for a non-lossy, “uncompressed” codec.
Codec means compress-decompress, so no, linear PCM is not a codec.
 
This is a helpful thread. Thanks @bigshot.

I've caught a bit of 'lossless' fever and was about to plunge into the mora$$ of equipment upgrades, but quite honestly, I flunk every lossless/lossy comparison test I've taken, so what's the point?

I may eventually pick up a <$100 DAC just for the heck of it but I'm pretty sure it's going to end up in my pile of unused gadgets.
 
I was asked by an equipment manufacturer to evaluate their flagship headphones and give suggestions for improvement to the designer. I did a bunch of tests and wrote up some notes and chatted with their lead engineer and they were pleased. They asked me if I would do the same with their high end ($1200) DAC/headphone amp. I said sure. They sent me this fancy amp and I did a bunch of tests and I couldn't find any indication that it added anything at all to the sound quality. I wrote them back and told them that and I got stony silence. No more stuff to evaluate for me!
 
I can definitely tell the difference between lossy/lossless. Your average listener listening on earbuds won’t. You would need an audio system that is resolving enough so you could hear the difference provided you know what to listen for. I work in the music industry as a composer/sound designer so I’ve had the pleasure of listening with a multitude of playback systems. Both in studios and in homes as I also appreciate a good hi-fi system. I however prefer to listen to music rather than the equipment, which is a problem I feel with audiophiles. Compressed mp3s that have been limited to death in the Mastering process appeals to many as its louder. Louder equals better in most circles. There’s little to no dynamic range in a lot of music being put out these days. It’s ok for things like hip-hop and heavy dance music but unfortunately the same processes are applied to all genres to help them compete.
 
  • Like
Reactions: ntlman
I can definitely tell the difference between lossy/lossless. Your average listener listening on earbuds won’t. You would need an audio system that is resolving enough so you could hear the difference provided you know what to listen for. I work in the music industry as a composer/sound designer so I’ve had the pleasure of listening with a multitude of playback systems. Both in studios and in homes as I also appreciate a good hi-fi system. I however prefer to listen to music rather than the equipment, which is a problem I feel with audiophiles. Compressed mp3s that have been limited to death in the Mastering process appeals to many as its louder. Louder equals better in most circles. There’s little to no dynamic range in a lot of music being put out these days. It’s ok for things like hip-hop and heavy dance music but unfortunately the same processes are applied to all genres to help them compete.
This is a really good point that is somewhat overlooked. The extra loudness in more recent mastering or re-masters appeals to cheap earbuds/headphones that don't seal off outside noise well without active noise-cancellation. But listening to well balanced music from vinyl or original issue CDs let users crank up the volume without getting unbearable levels of distortion.

Here's a couple visual samples for comparison. The track is "Changeling", from the 1979 album Real to Real Cacophony by Simple Minds. It was originally issued on CD in 1985, and later remastered in 2002. The original track is well balanced and within a healthy level range for volume control. The "remaster" is nearly brickwalled, and has limited volume range before distortion creeps in. Because of the high-frequency filter applied for the remaster, the natural reverb of the instruments and voice is lost, resulting in a more brittle, harsher sounding mix.
Changeling-ORIG.png
Changeling-RMSTR.png
 
Last edited:
  • Like
Reactions: Gonzbull
It's pointless to say that you can hear the difference between lossy and lossless without specifying the codec and data rate. Anyone can tell the difference with fraunhofer MP3 below 128. No one I have ever encountered has been able to tell the difference with AAC 320 VBR.

And you can't judge sound quality by looking at waveforms, especially when the level of the two waveforms is different. If you are comparing apples to apples, you have to level match and compare with a listening test.

I have a listening test that compares three codecs at three different data rates along with lossless. If anyone is interested in seeing if they can actually discern a difference, and at what point they reach their level of transparency, I would be happy to administer it to you.
 
It's pointless to say that you can hear the difference between lossy and lossless without specifying the codec and data rate. Anyone can tell the difference with fraunhofer MP3 below 128. No one I have ever encountered has been able to tell the difference with AAC 320 VBR.

And you can't judge sound quality by looking at waveforms, especially when the level of the two waveforms is different. If you are comparing apples to apples, you have to level match and compare with a listening test.

I have a listening test that compares three codecs at three different data rates along with lossless. If anyone is interested in seeing if they can actually discern a difference, and at what point they reach their level of transparency, I would be happy to administer it to you.
As I wrote:

The "remaster" is nearly brickwalled, and has limited volume range before distortion creeps in. Because of the high-frequency filter applied for the remaster, the natural reverb of the instruments and voice is lost, resulting in a more brittle, harsher sounding mix.

This observation is not based on the visual sample I provided. This is based on my own listening experience. I'm not sure why you're so determined to prove others wrong by asserting that differences are undetectable when several people on this thread alone have pointed out that's not the case in their experience.

FWIW, my earlier comments have nothing to do with codecs. The visual samples are from uncompressed AIFF rips direct from the CD. My point is that that because modern mastering standards are not what they used to be, the final track's audio limitations can become more apparent in the ACC/MP3s that they are generated from.
 
I was replying to the same person you were.

Yes, I agree that different mixes and different masterings are going to be different. That has nothing to do with whether someone can hear the difference between high data rate lossy and lossless.

A lot of people who claim to hear "night and day" differences between lossy and lossless have either never bothered to carefully compare, or they are operating on impressions they had long ago of codecs and data rates that have become totally obsolete. There's an abundance of evidence from controlled listening tests that modern lossy codecs can be completely transparent for just about everyone.
 
  • Like
Reactions: motulist
I tried feeding it high resolution lossless files through bluetooth using Amazon Unlimited. The iPhone converted it to 44.1/48. The phone won't hand the cans a file it can't play.

By the way, the APMs should be able to handle any codec that is suited for bluetooth, not just AAC. AAC is just the best one and the one Apple favors.
AptX is the industry standard for high quality/hi-res bluetooth over standard bluetooth, but LDAC is becoming more and more popular as well. Apple is the outlier to not support either.

 
  • Like
Reactions: motulist
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.