When I first started encoding MP3s I used a 128kbps rate which is noticeably inferior to the original CD. I noticed this in the early 2000s when I would up listening to a CD of some music I usually listened to as a 128kbps MP3 and was blown away with how much more I heard.
I'd say that 192kbps is much better and the 320kbps that the author advocates is basically transparent.
Correctly identified with 100% accuracy. The author said they can't, but for me the mp3 versions have noticeable high frequency artifacts that make the recording sound slightly less clear. Using Sony XM5
Acoustic guitar, drums are a good signal - lower quality just sounds hollow / spacey. The most obvious a/b was the Gamma Ray sample, imo (with mid-range Beyer headphones, wired). It's easiest to tell with recordings you know well, for me Steely Dan is a good reference. I rip to FLAC for archiving even though 320 or 250+ VBR is probably 'close enough' unless I'm scrutinizing.
Part of that might be if you're using them wireless because then you're double compressing the audio which amplifies the artifacts (mp3 -> Bluetooth compression).
Pretty great demo! It'd be great to see a 128/192 comparison.
I had Tidal many years back, and from the Lossless v Regular I only ever noticed a difference when it came to breathy sounds/etc. I did see that Tidal would burn through like 50GB of data monthly though.
Also - you may want to test some more modern recordings, the microphone/mastering quality of things nowadays is far better than what it was 2 decades ago (despite what some audiophiles may claim)
As the author points out, it's not really a "MP3 vs Uncompressed" conversation, it's a "which encoder are you using" conversation ...
because any of us from the late 90s/early 2000s who used the early versions of LAME will tell you in a second how easy it was to pick MP3 over raw, even at 320kb/s
I remember this repeated with the opensource AAC encoders. We had pretty decent LAME MP3 output by then, but everybody wanted to squeeze bytes and suddenly we were hearing a lot of terrible artifacts again.
Few audio things bug me more than the kind of tinkly pre-echo effects that were pervasive for a while.
It was really easy to tell which is which for the vocals.
On the other hand, the only sample in which I didn't hear ANY difference is Ennio Morricone's, to the point where I couldn't really tell it apart from its 56kbit/s version.
Can the hearing be selectively bad for some frequencies within the standard 20-20000 range, and normal for the others?
I was right for all but one. High frequencies give it away. I can tell the difference, but it was certainly close enough that I am not sure I care anymore.
I wonder how likely it is that the people who are posting that they got most of them correct are just the people who happened to randomly guess correctly with 50/50 chance each time - people who guessed wrong or thought they couldn’t tell probably aren’t going to post…
It could often depend on the encoder - things like lame have a hard low-pass filter even on the "insane" settings [0]. This can often mean, if you're someone who can detect that high frequency (probably not most adult), you may pretty easily be able to tell the difference if those frequencies are present in the recording.
Additionally, a lot of audio pipelines (even beyond the DAC - like amplifiers and similar) can end up with artifacts and harmonics in more audible frequencies - this is often more notable at extremely high frequencies (like 96khz and similar) - there's honestly nothing any human can actually hear near that range - but that doesn't mean it doesn't then affect audible ranges when actually played back on real equipment.
The big point is that "Being Able To Tell The Difference" isn't always the same as "Better Quality". You're often just replacing one artifact of the playback pipeline with another. Neither may truely match the original performance.
first: I have done this test myself many times in various ways, including recreating albums as a mix of 16bit FLAC and v0 MP3 (track by track, not within tracks), putting them on, and listening on speakers. I can tell sometimes, but the v0 still sounds great.
I was able to distinguish the 3 rock recordings with confidence, high frequency transients sounded more impactful in WAV. The Queensryche in particular has a lot of (well applied!) dynamic compression on the acoustic guitar and vocal which really brings out those transients.
However, if I heard the MP3 in isolation I would not detect anything was off. They all sounded good.
The Morricone and Vangelis I had no conviction either way and I guessed wrong both times. I suspect in their recording/mixing/mastering a lot of high frequency sound was lost anyway. In either case, I don't know if the CD master was made from original tapes or not. I know the Blade Runner OST has had a convoluted release history. Morricone has a 2004 CD master which is pretty well liked.
"Moving Pictures" was recorded to tape, but was notably an early digitally mastered album. Maybe that has resulted in preserved high frequency sound.
Compressed audio is great, I love it and I use it a lot.
I use CD Quality for archival purposes and my home library.. for most of the past decade hard disks have been inexpensive. I convert to Opus 192 for mobile devices.
Another reason for CD Quality archiving - I have a long term idea of recreating a CD collection. I want to get printable CDs and burn the audio/print the art because I want my children to have the experience of going thru a shelf or flipping through a binder, putting the disk in the tray, pressing play. I always loved doing that.
Again, could I tell if I transcoded a well encoded mp3 back to redbook? Maybe not consistently, but it's more likely the transcode of mp3 -> CD would introduce more audible problems than the encoding of WAV -> mp3.
Huh, I guessed all correct. Random guess would have a 1/(2^5) = 1/32 chance of being correct.
I don't make any claim to any special hearing or expertise. I've been listening to practically only lossy music since around '98, ripping from CDs at that time.
Morricone and Vangelis have been especially hard for me to tell apart, could have been a random guess on my part (I listened to those ~20 times).
When I read the title I expected to hear the actual _difference_ between the lossless and lossy waveform - i.e. only the actual artifacts. Could be a fun exercise.
The author could do a bit more work to make their opinions more valid. First, what speakers were they listening on and in what environment? If attempting to do critical listening, you do not use low quality speakers in a reverberant listening space with room mode colouration & cancellation.
Second, why not do a null test? Invert phase of MP3 and mix it with uncompressed WAV file. The remaining audio IS the difference, and it is audible.. But it is only significant if you have a good sound system & have ears capable of critical listening.
Third, what volume SPL were they listening at? A compressed JPG looks ok on a small screen but project it large and the image becomes blocky. Same applies to audio.
Fourth, tester fails to mention age and results of a recent hearing test. Maybe they have no perception over 12kHz? We do not know.
20 comments
[ 2.6 ms ] story [ 35.1 ms ] threadWhen I first started encoding MP3s I used a 128kbps rate which is noticeably inferior to the original CD. I noticed this in the early 2000s when I would up listening to a CD of some music I usually listened to as a 128kbps MP3 and was blown away with how much more I heard.
I'd say that 192kbps is much better and the 320kbps that the author advocates is basically transparent.
I would never know the difference during casual listening. Only in this setting where I'm told upfront that there is a difference, do I notice it.
Also, you can train yourself for what to listen for, to a point.
I had Tidal many years back, and from the Lossless v Regular I only ever noticed a difference when it came to breathy sounds/etc. I did see that Tidal would burn through like 50GB of data monthly though.
Also - you may want to test some more modern recordings, the microphone/mastering quality of things nowadays is far better than what it was 2 decades ago (despite what some audiophiles may claim)
because any of us from the late 90s/early 2000s who used the early versions of LAME will tell you in a second how easy it was to pick MP3 over raw, even at 320kb/s
Few audio things bug me more than the kind of tinkly pre-echo effects that were pervasive for a while.
On the other hand, the only sample in which I didn't hear ANY difference is Ennio Morricone's, to the point where I couldn't really tell it apart from its 56kbit/s version.
Can the hearing be selectively bad for some frequencies within the standard 20-20000 range, and normal for the others?
Once you hear the difference in sound quality / see difference in image quality you cannot undo it.
I have become very picky with display resolution and text clarity, and it has not served me well. I miss the days I was happy with a 1080p monitor.
Additionally, a lot of audio pipelines (even beyond the DAC - like amplifiers and similar) can end up with artifacts and harmonics in more audible frequencies - this is often more notable at extremely high frequencies (like 96khz and similar) - there's honestly nothing any human can actually hear near that range - but that doesn't mean it doesn't then affect audible ranges when actually played back on real equipment.
The big point is that "Being Able To Tell The Difference" isn't always the same as "Better Quality". You're often just replacing one artifact of the playback pipeline with another. Neither may truely match the original performance.
[0] https://sound.stackexchange.com/questions/38109/lame-why-is-... - while not an explicit "low-pass" filter, the default option of "-Y" does something similar.
first: I have done this test myself many times in various ways, including recreating albums as a mix of 16bit FLAC and v0 MP3 (track by track, not within tracks), putting them on, and listening on speakers. I can tell sometimes, but the v0 still sounds great.
I was able to distinguish the 3 rock recordings with confidence, high frequency transients sounded more impactful in WAV. The Queensryche in particular has a lot of (well applied!) dynamic compression on the acoustic guitar and vocal which really brings out those transients.
However, if I heard the MP3 in isolation I would not detect anything was off. They all sounded good.
The Morricone and Vangelis I had no conviction either way and I guessed wrong both times. I suspect in their recording/mixing/mastering a lot of high frequency sound was lost anyway. In either case, I don't know if the CD master was made from original tapes or not. I know the Blade Runner OST has had a convoluted release history. Morricone has a 2004 CD master which is pretty well liked.
"Moving Pictures" was recorded to tape, but was notably an early digitally mastered album. Maybe that has resulted in preserved high frequency sound.
Compressed audio is great, I love it and I use it a lot.
I use CD Quality for archival purposes and my home library.. for most of the past decade hard disks have been inexpensive. I convert to Opus 192 for mobile devices.
Another reason for CD Quality archiving - I have a long term idea of recreating a CD collection. I want to get printable CDs and burn the audio/print the art because I want my children to have the experience of going thru a shelf or flipping through a binder, putting the disk in the tray, pressing play. I always loved doing that.
Again, could I tell if I transcoded a well encoded mp3 back to redbook? Maybe not consistently, but it's more likely the transcode of mp3 -> CD would introduce more audible problems than the encoding of WAV -> mp3.
I don't make any claim to any special hearing or expertise. I've been listening to practically only lossy music since around '98, ripping from CDs at that time.
Morricone and Vangelis have been especially hard for me to tell apart, could have been a random guess on my part (I listened to those ~20 times).
When I read the title I expected to hear the actual _difference_ between the lossless and lossy waveform - i.e. only the actual artifacts. Could be a fun exercise.
Second, why not do a null test? Invert phase of MP3 and mix it with uncompressed WAV file. The remaining audio IS the difference, and it is audible.. But it is only significant if you have a good sound system & have ears capable of critical listening.
Third, what volume SPL were they listening at? A compressed JPG looks ok on a small screen but project it large and the image becomes blocky. Same applies to audio.
Fourth, tester fails to mention age and results of a recent hearing test. Maybe they have no perception over 12kHz? We do not know.
Failed test, confirms user bias, as expected.