• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

The state of lossy audio in 2022.

If compatibility is a concern (since not all decades old devices support Opus decoding), MP3 highest quality VBR is good enough (actually better than 320kbps MP3 I recall reading due to allowing whatever bitrate the file calls for during playback).
320Kbps CBR with LAME just a V0 with the bit rate capped to 320k for all frames It doesn't have quality tuning that V3 ~ V0 have. Pretty sure VBR mode actually abuses the bit reservoir allowing up to 580kbps but the decoders thinks It seeing a 320kbps frame. Also I've noticed that with both Helix 5.1.1 ~ 5.1.2 & LAME when set to 48KHz SOX(max quality) it legit helps with pre echo triggering content.
 
OPUS actually seems to be the poised successor to MP3 (and AAC if Apple would just stop being Apple for two seconds). I saw one guy on Reddit though challenging this notion with Monkey's Audio Encoder. I remember being convinced in a serious aspect I can't recall though (might be filesize with respect to preserving most of the spectrum). But I personally lean on Opus since it's bandwidth savings are huge given how good the results are. It's deployed on platforms like Twitch and Discord for good reason, unrivaled in voice communications I believe, so not just music frequencies. But yeah, almost all these codecs are good enough to use and never worry about lossless differences during listening (when there are audible differences, you have to be listening for it, and be trained, and listening to specific tracks). It all comes down to compatibility and bandwidth savings and I suppose conversion speeds at the end of the day. I think OPUS achieves this better than any other.

If compatibility is a concern (since not all decades old devices support Opus decoding), MP3 highest quality VBR is good enough (actually better than 320kbps MP3 I recall reading due to allowing whatever bitrate the file calls for during playback).

Without wanting to get into the "Facts, or just an Apple Fanboy" discussion :) , I do think it's worth noting that Apple chose AAC when it first set up the iTunes Store because AAC is open-source and at the time mp3 was not. (And to my knowledge AAC was/is just as good or better at equivalent bitrates.)

I understand that Apple has a habit of only supporting its formats or the formats it likes, but I would also say that whatever walled-garden stuff Apple has pulled with its music ecosystem, landing on AAC and sticking with it was not an example of that.
 
iTunes does play MP3. :)
When we were talking about lossy bit rates 96kbps or 128kbps etc three decades ago then AAC probably did sound better than MP3 at those lossy bit rates.

But bandwidth is better and drive space is cheap. Anything in the realm of 150kbps and better probably sounds transparent. And LAME at V3 (the worst variable bit rate setting) is probably transparent for 99% of recorded files and even if there was a difference, the listener probably won’t be able to tell which file he prefers. I choose V2 because it’s one step up and an additional “level of assurance” of complete transparency.

For those anal retentive, any time you rip a CD in standard fashion, you are introducing ERRORS from the CD-ROM’s misread of data from the optical CD. The only way out is to use an error-free reader when converting to wav (Exact Audio Copy).
 
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.
Big +1 to this. Storing music in FLAC vs. MP3 makes good sense. It's like "why keep your negatives when you have these good prints"? I don't know, in case you want to make more prints? Maybe you can't easily see /hear the difference, but I don't like throwing away perfectly good information without a good reason.

If you ever want to start using a different lossy format someday in the future, it's unquestionably best practice to transcode from a lossless original.

Also, BT codec compression is one of my great grievances with the BT SIG and industry in general. There's usually plenty of bandwidth for a FLAC stream in a BT connection, they just haven't specified a universal lossless codec for BT yet. In general BT compression sounds like sweaty ass and is a lot easier to hear than even mediocre MP3. As a result an entire class of otherwise decent headphones and speakers are consigned to sweaty-ass-level sound, forever. It bugs me.
 
Last edited:
iTunes does play MP3. :)
When we were talking about lossy bit rates 96kbps or 128kbps etc three decades ago then AAC probably did sound better than MP3 at those lossy bit rates.

But bandwidth is better and drive space is cheap. Anything in the realm of 150kbps and better probably sounds transparent. And LAME at V3 (the worst variable bit rate setting) is probably transparent for 99% of recorded files and even if there was a difference, the listener probably won’t be able to tell which file he prefers. I choose V2 because it’s one step up and an additional “level of assurance” of complete transparency.

For those anal retentive, any time you rip a CD in standard fashion, you are introducing ERRORS from the CD-ROM’s misread of data from the optical CD. The only way out is to use an error-free reader when converting to wav (Exact Audio Copy).
Hydrogenaudio gets mad at the thought of Lossy audio at from 192 ~ 256kbps hence LAME & Musepack dev's gave up being active there. As threads would just turn into "Use lossless" then just locked before you can say anything.

This is why I don't see the point with touching Opus/AAC at 256kbps since LAME V0 is also transparent. Caring about under 160kbps in this day & age is quite silly It a mindset that only worked in 1997 ~ 2007 where the best for portable was 64GB we have 256GB micro SD's those are enough for throwing in a V0 MP3 flies.
 
Also, BT codec compression is one of my great grievances with the BT SIG and industry in general. There's usually plenty of bandwidth for a FLAC stream in a BT connection, they just haven't specified a universal lossless codec for BT yet. In general BT compression sounds like sweaty ass and is a lot easier to hear than even mediocre MP3. As a result an entire class of otherwise decent headphones and speakers are consigned to sweaty-ass-level sound, forever. It bugs me.
Hmm.... can't complain about my lil' BT transmitter that I use on the go. Fat chance that I can hear the compression in a noisy bus/city anyway. :'D

Have you tried the new aptX "lossless" version yet? From what I've seen that can go up to 1200kbps.
 
My understanding is that aptX does increase the bluetooth latency. Not an issue when just listening to pure audio/music but annoying when you are watching a video.

With drive space cheap, lossless storage is not an issue like it was when I recall paying 400 dollars for a 100 gigabyte hard drive in the prehistoric ages when I first started ripping CD’s. As a poor student, $ savings on hard disks was a major factor. :)
 
My understanding is that aptX does increase the bluetooth latency. Not an issue when just listening to pure audio/music but annoying when you are watching a video.

With drive space cheap, lossless storage is not an issue like it was when I recall paying 400 dollars for a 100 gigabyte hard drive in the prehistoric ages when I first started ripping CD’s. As a poor student, $ savings on hard disks was a major factor. :)
Well, if you want to store your CD-collection on a storage in the cloud, to have it accessible wherever you happens to be in the world, then donwloading or streaming over the internet can still be an cost related issue.
 
My understanding is that aptX does increase the bluetooth latency. Not an issue when just listening to pure audio/music but annoying when you are watching a video.

With drive space cheap, lossless storage is not an issue like it was when I recall paying 400 dollars for a 100 gigabyte hard drive in the prehistoric ages when I first started ripping CD’s. As a poor student, $ savings on hard disks was a major factor. :)
When I first started with digital audio for broadcasting, a 4GB drive was a full 5 1/4" bay, and cost around £2000!!! This was 1992/3, and we couldn't believe that it was possible to get 4,000,000,000 bytes in something we could hold in our hands. Floppy drives were 1MB, and hard-drives in PCs were around 40MB. GB drives were almost science-fiction.

S.
 
Last edited:
My understanding is that aptX does increase the bluetooth latency. Not an issue when just listening to pure audio/music but annoying when you are watching a video.
Easy to solve: Just use a video player that allows for manual audio-delay settings.
 
No idea why Helix MP3 5.1.1 ~ 5.1.2 Isn't more popular since It outperforms LAME at 192 ~ 256kbps. I get zero pre-echo at at 192 ~ 256kbps at 48KHz on pre-echo triggering samples.
 
No idea why Helix MP3 5.1.1 ~ 5.1.2 Isn't more popular since It outperforms LAME at 192 ~ 256kbps. I get zero pre-echo at at 192 ~ 256kbps at 48KHz on pre-echo triggering samples.
I'm not sure it's better or worse. LAME's psychoacoustic filter appears to be more cautious below 16KHz, only omitting noise between strong harmonics. That's what I highly prefer. Helix appears to pull a lot from 12-16KHz, and does more noise filtering below 12KHz. Whether that's audible or not, I'm not sure. Do what makes you happy. I'm sticking with LAME V0 for my car and FLAC on my home network.

Helix -V100 -HF2
1678816453403.png


LAME VBR V3
1678816474249.png
 
If your doing V0 it better to have force -Y & set a 20KHz lowpass as the 16 ~ 22KHz area will steal bits from under 16KHz area causing issues.
 
Tried QAAC at 160kbps VBR It amazes me how It chokes on synth heavy Dark ambient or Dark ambient with acoustic instruments noticed one Noise album just fails unless the --quality is set to 32. Basically It could match Opus/Musepack but no encoder makes use of It feature set but then again I don't buy the idea MP3 being 576 x 192 Is the issue AAC/Opus devs claim. HA already tried to hide the fact Lame 3.99+ uses a trick to allow frames over 320kbps since on Foobar I get VBR files being 332kbps for 1 second?.
 
Honestly because all my CDs are ripped to alac
And here I thought I was the only one! After all these years I still haven’t found anything better than iTunes for getting music onto my iPhone.

I reripped all my CDs to ALAC (previously MP3) when it became available. Not because I had a problem with lossy but just to not have to worry about generational loss when the next sexy lossy codec came along. I assumed at some point I’d batch convert to FLAC but it never happened because as Apple relaxed its walled garden streaming took over and nobody else bothered to make it easy to transfer a music collection to an iPhone. As terrible as iTunes is, I appreciate how easy it is to manage a portable library through smart playlists and transcoding. They did a few things right.

The downside is that I have to use iTunes to manage my digital library, but luckily Foobar doesn’t mind at all. The only annoying thing is having to convert the occasional FLAC to ALAC just to appease the ghost of Jobs.
 
Just an update to my apple aac thoughts earlier. Somehow, the decoding of aac's inside deltawave with a full match was causing the downsloping matched spectra and odd artifacts that aren't visible when the aac is decoded to a wav externally. Hey @pkane does this sound like a windows decoder issue? I can reproduce it at home, but not at work for some reason. I tried installing the latest k-lite codec pack, but no change.
 
because AAC is open-source and at the time mp3 was not. (And to my knowledge AAC was/is just as good or better at equivalent bitrates.)
How can you type so much wrong information? Like audiophiles who claim that Apple invented AAC.

AAC is still covered by patents while the patents on MP3 expired in 2017.
 
Just an update to my apple aac thoughts earlier. Somehow, the decoding of aac's inside deltawave with a full match was causing the downsloping matched spectra and odd artifacts that aren't visible when the aac is decoded to a wav externally. Hey @pkane does this sound like a windows decoder issue? I can reproduce it at home, but not at work for some reason. I tried installing the latest k-lite codec pack, but no change.
AAC reader in DeltaWave will use the available Windows decoder, so likely that's the culprit.
 
How can you type so much wrong information? Like audiophiles who claim that Apple invented AAC.

AAC is still covered by patents while the patents on MP3 expired in 2017.

AAC is open source; 2017 was many years after Apple set up the iTunes Store.
 
Back
Top Bottom