The state of lossy audio in 2022.

[fieldcar]

Lately there have been a few threads complaining about the lack of fidelity for lossy audio. I figured that it may be worthwhile to better illustrate where we are actually at with a few samples. Given that our auditory memory is only a matter of seconds and most tests play out a long chunk of a song, I decided to prepare a perfect loop track from a random song. What better random track than a random song from Daft Punk's Random Access Memories? I've formatted the samples as shown below, and it will be best to play in a player capable of gapless playback like foobar2000. Make sure you go to Playback > Order > Repeat (playlist) if you'd like the playlist to loop.

I'm running the test through my focusrite 2i4 > balanced > JBL 306pmkii's and I'm having a really hard time telling them apart, even the lower bitrate tracks. I'm really curious to hear what everyone thinks. What are your thoughts?



Download samples here:

Lossy Test - Google Drive

drive.google.com

foobar2000:

foobar2000

foobar2000

www.foobar2000.org

Study comparing 192Kbit MP3 and AAC vs wav:
https://downloads.hindawi.com/journals/ijdmb/2019/8265301.pdf

Study comparing lower bitrate Opus, AAC, Ogg Vorbis and MP3 to a lossless source track:
Results of the public multiformat listening test (July 2014)

 

[Zensō]

Thanks for doing this. I already know the answer, at least for myself: in normal listening, high bitrate lossy (256 AAC, 320 Ogg Vorbis) is indistinguishable from lossless and hi-res. Given enough time and focused effort I might be able to pick out some minute differences in certain types of isolated high frequency sounds, but these would be irrelevant when listening to music for enjoyment.

Here are a couple of blind tests that support my findings:

Blind Test Results

The file size of an iTunes song is only a fraction of its CD counterpart because it has been compressed discarding data in the process. Is the audio quality degraded? More than a year ago I devised…

cdvsmp3.wordpress.com

High Bitrate MP3 Internet Blind Test: Part 2 - RESULTS

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

 

[ahofer]

Yeah, I already did a bunch of these and know my limits here. I do better with classical music differentiating 320 from redbook, but not perfect (like 7/10). With pop I am just guessing.

 

[fieldcar]

Yeah, I already did a bunch of these and know my limits here. I do better with classical music differentiating 320 from redbook, but not perfect (like 7/10). With pop I am just guessing.

I did try the same a while back with the ET orchestral OST and found flaws in some instruments. I feel like strings that pitch bend are usually the easiest "tell" for compression artifacts. Maybe I'll throw those in if I've got some time.

 

[Jimbob54]

Thanks for doing this. I already know the answer, at least for myself: in normal listening, high bitrate lossy (256 AAC, 320 Ogg Vorbis) is indistinguishable from lossless and hi-res. Given enough time and focused effort I might be able to pick out some minute differences in certain types of isolated high frequency sounds, but these would be irrelevant when listening to music for enjoyment.

Here are a couple of blind tests that support my findings:

Blind Test Results

The file size of an iTunes song is only a fraction of its CD counterpart because it has been compressed discarding data in the process. Is the audio quality degraded? More than a year ago I devised…

cdvsmp3.wordpress.com

High Bitrate MP3 Internet Blind Test: Part 2 - RESULTS

A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.

archimago.blogspot.com

This is the point and one that comes across loud and clear in the many ABX tests littering this site and the web in general as well as all the hand wringing about audible differences between components. DACs especially. Its not whether one can discern file A from B or component A from B in a quick swap blind test, its whether the difference is noticeable to the extent you would spot it in your normal listening conditions. Which invariably it isnt.

By all means go lossless or higher if it prevents the nagging doubts that you could get better, especially if it doesnt cost much/ anything. By all means choose uber DAC X for your system because it has unimpeachable performance compared to the middling DAC Y built into your amp etc but again. don't pretend its making a significant difference to your listening.

Peace of mind has a value, no doubt. But to me, its marginal.

 

[Digital Mastering System]

Years ago I did lossy codec tests by encoding and decoding a music sample 10 times. This accentuated the errors and made them obvious enough so you could be 'trained' to hear them. I found MP3 seemingly adds noise and distortion. The vector compressor (Yamaha VQF) had less noise and distortion, but exhibited terrible gain errors and sounded much worse.

 

[DVDdoug]

For me, good-quality lossy is good-enough. I don't hear a difference, and I don't try to hear a difference. I think the main argument for lossless these days is that storage & bandwidth is plentiful and cheap, so "why not' use lossless?

I have a 160GB iPod almost-full of "V0" MP3s (the "highest quality" LAME variable bitrate). My vehicles have iPod dock connectors and the iPod lives in my car. Sometimes I'll listen to the MP3s on my computer plugged into my home stereo or with headphones.

I've never done any ABX listening tests and I've intentionally avoided training myself to hear compression artifacts, (Of course we've all heard poor-quality MP3s.)

The few times I've thought I was hearing a compression artifact, the "defect" also appears on the original CD.


...I grew-up with vinyl and I didn't have to train myself, or listen carefully, to hear the "snap", "crackle", and "pop" (which I found VERY annoying), or the inconsistent/mediocre frequency response, or the occasional distortion.

 

[sergeauckland]

For me, good-quality lossy is good-enough. I don't hear a difference, and I don't try to hear a difference. I think the main argument for lossless these days is that storage & bandwidth is plentiful and cheap, so "why not' use lossless?

I have a 160GB iPod almost-full of "V0" MP3s (the "highest quality" LAME variable bitrate). My vehicles have iPod dock connectors and the iPod lives in my car. Sometimes I'll listen to the MP3s on my computer plugged into my home stereo or with headphones.

I've never done any ABX listening tests and I've intentionally avoided training myself to hear compression artifacts, (Of course we've all heard poor-quality MP3s.)

The few times I've thought I was hearing a compression artifact, the "defect" also appears on the original CD.


...I grew-up with vinyl and I didn't have to train myself, or listen carefully, to hear the "snap", "crackle", and "pop" (which I found VERY annoying), or the inconsistent/mediocre frequency response, or the occasional distortion.

Exactly this. When I started as an audio design engineer, I was trained in listening for things like wow & flutter, treble crushing and all the other ills that beset tape recordings. Partly for that reason, I gave up on tape as soon as I could. The faults on vinyl are pretty obvious, so never had to be especially trained on these , and W&F affects both tape and records. When digital came along in the early 1980s, it was great, and bit by bit, my early training became obsolete and unpracticed, so when data compression came about, I didn't have a professional need by then to identify faults, and happily never learnt. I can now listen to MP3s, AAC, Ogg etc etc quite happily. Yes, there's something 'wrong' about very low bit rates, but I'm not bothered by even 128k MP3, I can quite happily listen to Venice Classic or France Musique at 128k, and of course BBC Radio 3 at 320k is fine.

Sometimes, ignorance is indeed bliss.

S.

 

[TheBatsEar]

B-but it's lossy, my gents!

Doesn't matter if you hear the difference in every track or not, it's enough if you hear it in certain tracks under certain conditions.

If in 20 years you get a strange new device as a player and need your stuff converted to its format, and you have stored lossy, then you are, pardon my german, gefickt.

There is one case where lossy makes sense, and that is in case you are a multinational megacap ultragiant corporation, using lossy as a form of weak DRM, keeping the good stuff in your vault, so you can milk it longer.


You should fight lossy! Like, always!

 

[anotherhobby]

Doesn't matter if you hear the difference in every track, it's enough if you hear it in certain tracks under certain conditions.

It'd say this pretty much nails my thought process. It's not hard for me to hear most compression. No, I can not always hear it with every kind of music, and not at all volumes, but absolutely sometimes, so why even bother? If you can hear it, it sounds gross. I just use lossless streaming services, keep my personal digital music collection as lossless FLAC, and I don't worry about it. I just want CD quality. I could care less about hi-res audio.

 

[Zensō]

It'd say this pretty much nails my thought process. It's not hard for me to hear most compression. No, I can not always hear it with every kind of music, and not at all volumes, but absolutely sometimes, so why even bother? If you can hear it, it sounds gross. I just use lossless streaming services, keep my personal digital music collection as lossless FLAC, and I don't worry about it. I just want CD quality. I could care less about hi-res audio.

Multiple well-designed blind tests have shown that an overwhelming majority of people (like in the >99% range) cannot hear the difference between high bitrate lossy (256 AAC and 320 Ogg, not low bitrate MP3) and lossless (this includes a wide range of subjects from “normal” people to musicians, audio professionals, and self-described audiophiles). You may very well be one of the lucky few (or unlucky depending upon how you look at it) that can easily hear the difference, but unless you’ve done so blind and level matched you may be experiencing at least some expectation bias. The reason I like to talk about the facts on this question is because there’s already enough FUD and nervosa among audiophiles, and I’d rather not add to the mountains of misinformation we’re already dealing with.

 

[pablolie]

Nothing is new under the sun. It's well known from countless scientifically conducted tests the vast majority of people can't tell apart HD from redbook from quality lossy bitrates in most great recordings (and you waste your time with ho-hum ones, which covers 80% of recordings).

I do enjoy listening to my fav recordings in my FLAC local library, but that's to a large degree for peace of mind. I did spent too many hours doing tests between high bitrate lossy and lossless, and with some of my well-recorded favs I am 90% accurate in telling the difference (I have been trained on what to listen for, though), but honestly I do not care about the difference.

What I care about is the integrity for archiving purposes. I can definitely 100% enjoy great music as Spotify streaming stuff. Recall my audiophile card if you want.

 

[Blumlein 88]

Flac vs high rate lossy is only a bit over a 2:1 advantage. I agree with why go lossy? The thing is I'd like some non-lossy Bluetooth to become common.

 

[anotherhobby]

Multiple well-designed blind tests have shown that an overwhelming majority of people (like in the >99% range) cannot hear the difference between high bitrate lossy (256 AAC and 320 Ogg, not low bitrate MP3) and lossless (this includes a wide range of subjects from “normal” people to musicians, audio professionals, and self-described audiophiles). You may very well be one of the lucky few (or unlucky depending upon how you look at it) that can easily hear the difference, but unless you’ve done so blind and level matched you may be experiencing at least some expectation bias. The reason I like to talk about the facts on this question is because there’s already enough FUD and nervosa among audiophiles, and I’d rather not add to the mountains of misinformation we’re already dealing with.

Yes, I'm well aware, and yes I've done several DBT on myself, starting back in 2016 when I first noticed it in some Apple AAC files. As I said in my comment "No, I can not always hear it with every kind of music, and not at all volumes, but absolutely sometimes, so why even bother?" But I appreciate all of your skepticism, it's healthy. Like with all this stuff, people just need to test themselves and determine what matters to them.

 

[Zensō]

Flac vs high rate lossy is only a bit over a 2:1 advantage. I agree with why go lossy?

No reason not to. That said, it’s always good to dispel myths in this hobby that is so overloaded with misinformation. The reality is that in normal listening, for most people, it just doesn’t matter. And in most cases, when people think they hear a difference it is highly likely to be expectation bias, not actual sonic differences. I’m the first to admit I’ve experienced this many times myself with codecs, bitrates, DACs, and so on. Blind testing always cuts through the confusion.

 

[pablolie]

I agree with why go lossy?

Convenience. Ability to discover new music quickly. There is power in that.

I also distinctly recall you saying elsewhere your internet is not ready for lossless, so infrastructure shortcomings may be a reason for some (not me).

 

[anotherhobby]

Convenience. Ability to discover new music quickly. There is power in that.

Sure, back before almost everybody started offering lossless streaming. Apple, Tidal, Amazon, etc all (except Spotify) have lossless audio now for less than the price of an album a month. While that was true for quite a while, I just don't agree that's a tradeoff anymore.

 

[tmtomh]

I've had a strong bias in favor of lossless high-res until recently. The high-res bias I let go of - in fact, I'm now sort of the other way, preferring red book to high-res because the latter is just a waste of space.

When it comes to lossless, though, I still prefer it - it just makes me feel better, and in most cases I find it trivially easy to get ahold of CD-quality lossless.

Now, if I subscribed to a streaming service, I don't think I'd pay extra for a lossless tier if the regular tier gave me 256 or 320k lossy. But I still collect CDs, and for most back-catalogue stuff, I can buy a used CD, rip the lossless files onto my computer server, and get the entire album in lossless with a robust physical backup for 2/3 or in some cases as little as 1/10th the cost of buying mp3s from Amazon or AACs from Apple.

So for me it's 16/44.1 lossless, but I'm certainly not averse to lossy and there's no way in the world I would pass a blind ABX test between lossless and 256k or higher lossy.

 

[TheBatsEar]

Convenience. Ability to discover new music quickly. There is power in that.

You could do that with lossless.

I also distinctly recall you saying elsewhere your internet is not ready for lossless, so infrastructure shortcomings may be a reason for some (not me).

They should buy better internet. Why should i use lossy because their internet is bad?

 

[pablolie]

If this was a discussion about which online streaming service was best, I'd go in that direction. And I'd state they all have strengths and weaknesses... size of library... UI... But I'll keep my comments to the title line of the state of lossless.

As I said, to me it's about convenience and size of library... and suggestions. Spotify help me discover new music quite successfully. What I used to do in Tower Records, Spotify now does for me quite well. I have read between Spotify and Apple they always claim the leads for the largest library. But I'll openly state to me Apple has become a part of the evil digital empires, and I try to avoid their ecosystem.

The rest have smaller libraries so are not as effective.

The SQ is good enough for me for the purpose of what I use Spotify for. I have said this else where but I support my fav artists by actually buying their music, and not just streaming it. SQ solved.