Let's open another can of worms — are music streaming services providing genuinely lossless audio?

[seeteeyou]

First of all, someone posted this back in 2018 and most likely it wouldn't be a pleasant read by any means

 

https://www.head-fi.org/threads/fake-flac-identification.826737/page-2#post-13999511

Quote

Unfortunately there is also a way to fake the spectrum itself, meaning, it is possible to turn an MP3-like spectrogram into a FLAC-like "full" spectrogram. I have seen examples of that and they can even fool a person using graphical spectrum analyzers, if they are not experienced. I won't say more about how that's done since I don't want to be giving anyone ideas (I may only explain how to detect such files or send learning examples, if anyone is interested). A well made fake from an already good lossy file may be literally impossible to detect though.

Web-shops selling lossless music and websites streaming lossless music are also not perfect. They rely on label supplying their releases and all the websites (shops and streaming sites) get the same files. And there are labels which sell fake WAV / FLAC files (lossy re-encodes). And I'm not talking about a handful of releases. I'm talking about complete catalogues! I keep a list of those and there are already over 20 labels from the electronic music genre (dance, house, trance, etc.) on there that I know about. One huge "offender" is DIY (Do It Yourself Multimedia Group) including all of its sublabels (Major Records, Liquid Sound, Nitelite, D-Lite Records, ...). Then there's Nocolors, Bit Records, Tornado, and the list goes on. And those are just from the genres I'm interested in. I have friends that worked in the music industry that told me that for older releases it is possible that the labels simply "lost" the digital masters after they went out of business or were acquired by other labels. But, there are even instances of digital downloadable releases dated 2017 that are being sold in lossy WAV/FLAC on web-shops. Most of the releases I checked are clearly unjustifiably fake, because I know that they are lossless on CD, but lossy on WEB WAV/FLAC. Web shops should definitely not be selling those to people who pay premium for lossless files but get exactly the same quality as MP3, or slightly better according to spectrogram (whatever lossy codec the label used), but definitely not lossless. By the way, again, all web-shops carry the same files since the labels send them all the same release. So if you get a lossy file from one web-shop, don't expect to get a different file from another web-shop. Just stick with the shop that makes less fuss when it's time to get a refund.

I have written to several bigger web-shops known in the electronic music genre and they are mostly unaware or unprepared for this (or they are playing stupid and waiting for people to start complaining). Once confronted they do issue refunds though, and I have seen releases taken down. Still, they don't automatically check all releases they receive from labels so there are A LOT of fake WAV/FLAC files being sold to gullible customers at this very moment. It is definitely a MUST to have Spek on your PC if you're buying lossless music on-line, even from legitimate web-shops offering digital downloads. Oh, and I also know of retail CDs that some labels mastered from actual MP3 files. So not even ripping a physical CD can be a sure way of getting a real lossless track. Again, use Spek for everything if you're dealing with lossless music files!

 

And then @The Computer Audiophile also mentioned this last year

 

https://audiophilestyle.com/ca/bits-and-bytes/is-it-time-to-rethink-lossless-r1231/

On 11/7/2023 at 10:38 PM, The Computer Audiophile said:

On the other hand we have files and streams at 16/44.1, 24/48, 24/96 and 24/192 from almost every record label. The likelihood that these are truly lossless to the original source is, unknown. It’s likely unknown to the artist, the label, and most everyone involved in creating the album.

 

---

 

I wouldn't talk about where it's found and how do we get access for obvious reasons, though 1:1 rippers are actually available for these guys listed below

 

• "BRIDAL" Music (FLAC)
• "Freezer" (FLAC)
• "Gobus" (FLAC)
• "Red Delicious" Music (decrypted ALAC)

 

Here's the thing, they would only rip the cover of an album if no lossless tracks were actually available. Then I also tried ripping a single track in order to confirm the fact that only lossy ones could be found on the server, that's exactly what I saw and therefore I would simply forget about it.

 

At some point I just stumbled upon this particular album

 

https://www.discogs.com/master/651826-Various-Music-From-The-OC-Mix-6-Covering-Our-Tracks

 

Even if there were a dozen tracks, I could only rip 2 tracks (i.e. 2 lossless + 10 lossy) off some music streaming services while getting only 1 track from the others.

 

Wouldn't that be more like a big fat red flag then? How on Earth could there be both lossless and lossy tracks on the very same album?

 

In addition, we don't have any software to verify lossless files with 100% accuracy + reliability. Both false positives and negatives should be expected given the fact that just about anything could be faked.

 

And then I could also find more cheesy examples. Disc #1 of a particular 12-CD box set seemed to be fine. Then I started ripping disc #2 but unfortunately only the cover could be ripped. Same deal for disc #3 and I had to give up afterwards.

 

All albums of this artist could be ripped off "Freezer" successfully since everything turned out to be lossless so far

 

https://musicbrainz.org/artist/a16ed3e1-1fab-4b18-8dfa-941eab07f373

 

OTOH, "Red Delicious" Music only provided a handful of albums in lossy format (i.e. no ripping was done at all except for the cover) while the rest of them should be "supposedly" lossless.

 

However, nobody knows if any "supposedly" lossless tracks were genuine or not.

 

---

 

Personally I'm not interested in discussing the ethics of ripping, please don't even bother

 

https://old.reddit.com/r/Piracy/comments/1625i64/does_anyone_really_pirate_music_anymore/jxwwd1l/

Quote

Spotify is piracy. They pay artists nothing.

 

https://musicalpursuits.com/music-streaming/

Quote

Another part of the music streaming equation is the artist revenue policies. Unfortunately, there are no official numbers available on how much each platform pays for its artists and which is the highest paying music streaming service.

According to the Trichordist’s artist streaming stats research (2020), the average pay-per-stream rate across all platforms is approximately $0.00173. (9)

 

For some reasons they could always advertise whatever they want by claiming that we're indeed paying for lossless quality, at least for an undetermined percentage of the entire catalog that is.

 

If restaurants weren't able to tell the differences between real and fake olive (i.e. 100% olive-free) oil to begin with, could we trust the providers of music streaming services without the means to verify anything whatsoever?

 

https://www.youtube.com/watch?v=kwyrangVexI&t=8m

[MarcelNL]

won't it be possible to determine if a FLAC album that you ripped yourself is identical to the streamed version, using something like paralell waveform analysis?

 

That said, I think we should set up a '100% genuine lossless file'certification system, for a fee per album and a general subscription fee per annum per provider once the initial audit is completed ;-)

[botrytis]

22 minutes ago, MarcelNL said:

won't it be possible to determine if a FLAC album that you ripped yourself is identical to the streamed version, using something like paralell waveform analysis?

 

That said, I think we should set up a '100% genuine lossless file'certification system, for a fee per album and a general subscription fee per annum per provider once the initial audit is completed ;-)

 

It may not be that but hopefully, it won't be like Radio, where compression is/was used extensively to make it 'sound better' than another radio station.

[MarcelNL]

compression on lossless streaming? I'd be interested to hear how that would work and end up being lossy?

I mean Radio sound quality is usually screwed up, DAB is already better but I fear it's banged up before being broadcast digitally...but FLAC files from a serious content provider like f.e. Qobuz, if those are messed up I'm done with them.

 

Albums being available in multiple formats make me weary, as it;s easy to reformat a dataset, especially with older stuff, yet newer recordings can be mastered using DXD to various formats.

[botrytis]

1 hour ago, MarcelNL said:

compression on lossless streaming? I'd be interested to hear how that would work and end up being lossy?

I mean Radio sound quality is usually screwed up, DAB is already better but I fear it's banged up before being broadcast digitally...but FLAC files from a serious content provider like f.e. Qobuz, if those are messed up I'm done with them.

 

Albums being available in multiple formats make me weary, as it;s easy to reformat a dataset, especially with older stuff, yet newer recordings can be mastered using DXD to various formats.

 

I trust the industry as far as I can throw it. I am not that strong.... 

 

If you get my meaning. 

[MarcelNL]

26 minutes ago, botrytis said:

 

I trust the industry as far as I can throw it. I am not that strong.... 

 

If you get my meaning. 

trust me, I do! I'm wary of the marketing lingo that floats around....been part of the industry but never could pusk myself to the point where the messaging outperformed the gear.

[Norton]

The question is only meaningful if you can define what “lossless”  means.

Bearing in mind that it will not be possible to reconstruct exactly the original recording from the vast majority of consumer distributions of music in any format , lossless is a meaningless marketing term. In practice , it just means that it is possible to take a consumer distribution and reconstruct  the file that immediately preceded it in the distribution chain.

 

 

[danadam]

I have an idea. We could come up with a scheme that allows labels to embed a cryptographic signature in the lowest bit of audio. Then a DAC could verify if the signature is correct. If it is, it could maybe light up a LED. I think blue would be nice. What do you think? And how would we call it?

 

😉😉😉

[MarcelNL]

1 minute ago, danadam said:

I have an idea. We could come up with a scheme that allows labels to embed a cryptographic signature in the lowest bit of audio. Then a DAC could verify if the signature is correct. If it is, it could maybe light up a LED. I think blue would be nice. What do you think? And how would we call it?

 

😉😉😉

we'd call it an XYZ certified stream ;-)

[danadam]

3 hours ago, MarcelNL said:

compression on lossless streaming? I'd be interested to hear how that would work and end up being lossy?

In case it's not clear (I can't tell), he meant dynamic compression, not data compression. AFAIU radio broadcaster use audio processors to create a "signature sound", e.g. https://www.orban.com/radio-audio-processors

 

Some time ago I captured "Luka" by Suzanne Vega played on Polish radio https://radio357.pl/ and compared it with my CD and with what Spotify streams. CD and Spotify were very similar and the radio version quite different:

 

 

[MarcelNL]

O you were quite clear, yet there is a major difference between radio (FM / AM) and data streaming.

for FM/AM radio all sorts of compression and sound manipulation is used to deal with a limited bandwidth, whereas DAB (Digital Audio Broadcast) SHOULD be capable of transmitting higher quality osund than FM but in reality are not.

https://en.wikipedia.org/wiki/Digital_Audio_Broadcasting#:~:text=The typical bitrate for DAB,-4 HE-AAC standard.

 

Digital streaming OUGHT to be capable of delivering FLAC, effectively using far more 'bandwidth' than any radio (digital or analog).

 

 

[yamamoto2002]

I'm not surprised if they alter original PCM data to add audio watermark on PCM stream to track down which user illegally redistribute it to pirate sites

[MarcelNL]

I dunno, they could but that should not change anything beyond the CRC checksum or similar ...the waveforms should be identical. How IS DRM done these days?

[botrytis]

3 hours ago, MarcelNL said:

I dunno, they could but that should not change anything beyond the CRC checksum or similar ...the waveforms should be identical. How IS DRM done these days?

 

Microsoft Word - dsp_project_final.doc (columbia.edu) This is how. Actually, a great paper.

 

I hope this helps. 

[Currawong]

On 3/8/2024 at 5:14 PM, yamamoto2002 said:

I'm not surprised if they alter original PCM data to add audio watermark on PCM stream to track down which user illegally redistribute it to pirate sites

The major labels used to, and it was audible. I don't know what caused them to change it, but probably that TIDAL and Qobuz couldn't advertise their music as lossless if the originals had been altered.

[PeterG]

I think part of this needs to include the definitions of "lossless" and "providing".  With a Bluesound Node 2 streamer going into my DAC, I have never been able to get sound that matches a ripped CD going into the same DAC.  I understand there are different versions, etc; but the overall sonics of a significant number of albums were all second best via streaming.  So I am certain I am not able to provide a lossless redbook stream into my DAC with my well-respected, value-priced streamer.  My conclusion was that lossless streaming may be possible at a higher price point, but definitely not my price point.

 

Are others here able to stream without sonic compromise vs CD?

[MarcelNL]

lossless does not mean that quality of sound cannot suffer...a better network, better streamer and the gap becomes smaller

[PeterG]

5 hours ago, MarcelNL said:

lossless does not mean that quality of sound cannot suffer...a better network, better streamer and the gap becomes smaller

 

Yes, this is exactly what I meant by the need to define lossless.  Separate from the numbers, I lost quite a bit of music

[Currawong]

Lossless has a hard definition: That the music hasn't been altered irreversibly from the master recordings (of which there may be multiple)* at the bit-level.

 

That's as succinctly as I can put it, so it bears a bit of expansion. 

 

Irreversible alteration would mean that either lossy compression has been applied (for MP3, AAC and MQA), watermarking, or that some kind of post-processing or EQ is applied, such as volume levelling. This can equally be applied to the original digital file, or done on the fly by software or hardware.

 

So, the question really is: Are music services using the same digital music files as you'd get on, say, a CD, or buy from HDTracks, Linn, NativeDSD or elsewhere?  Then, given peoples' experiences with the sound seeming to be different from a CD than from a streamed track, is that really the result of the files being different, or is something going on with the software or hardware in their system (eg: volume levelling, re-sampling, or hardware noise levels) to make an audible difference?

 

If we consider the former, that a streamed track is not a bit-perfect match to a CD, is it because it's at a different sample rate? If it is 16/44.1, then is it from a newer or older master of the same album? Are both the CD and streamed versions exactly the same length? If not, that would cause a checksum comparison to fail, but still result in a flat output from an audio diff comparison if the samples are aligned correctly. 

 

I don't think that there will be a straight-forward answer to all these questions, and it has to be considered for each individual album. Some famous albums have numerous versions now. Albums on streaming services seem to be constantly deleted and replaced, without any explanation as to why.

 

*Ie: There may be multiple masters at different sample rates, and also masters for particular formats, eg: radio, though I'm not sure how common this is now.

[yamamoto2002]

8 hours ago, Currawong said:

So, the question really is: Are music services using the same digital music files as you'd get on, say, a CD, or buy from HDTracks, Linn, NativeDSD or elsewhere?

 

On several years ago of Sound & Recording Magazine (monthly magazine published in Japan) there is the article about art of mastering for high-res. They said something like "it is better to hump some audible high frequency for high-res master, and apply more maximizer for CD master" so maybe answer is no.

 

[Norton]

18 hours ago, Currawong said:

Lossless has a hard definition: That the music hasn't been altered irreversibly from the master recordings (of which there may be multiple)*

But is that a hard definition, and if so , whose?  And indeed what do we mean by a Master recording and the implied provenance, if  there can be multiple masters?   For example, if a 70s analogue  recording is transferred by the record company to 24/96 and then distributed as RBCD as a “remaster” and I then make a bit perfect rip of that CD, can I be said to have a lossless copy of the original analogue master?  I’d suggest not and that, far from a hard definition, lossless is just a meaningless  marketing term 

 

 

[seeteeyou]

https://www.iamthegreengoddess.com/blogs/news/the-truth-about-marketing-claims-what-does-natural-mean

Quote

In other words, anyone can say their products are natural, and ultimately, it means nothing whatsoever because no governing body is breathing down their necks, checking in on their claims. For some companies, if an ingredient ever once (in any way) resembled something in the natural world, that's good enough for them. For others, the word is taken very seriously. 

 

https://archive.today/MIs6x

https://www.bloomberg.com/opinion/articles/2023-09-03/ultra-processed-foods-should-come-with-warning-labels

Quote

There’s a critical distinction between processed and ultra-processed foods. Cooking is itself a form of food processing that humans have been doing for about a million years. Ultra-processed consumer foods emerged after World War I. They’re not foods, per se, but what my colleague Fernanda Rauber describes as “industrially produced edible substances.” They often contain emulsifiers, stabilizers, gums, softeners, sweeteners, modified starches, artificial flavorings, preservatives and so forth. Many ingredients are replacements — artificial sweeteners that replace sugars, gums that replace fats and bacterial exudates that create textures in place of the real molecules that should be in food.

 

https://www.cnn.com/2024/02/01/health/predigested-food-wellness/index.html

Quote

“It’s an illusion of food,” he added. “But it’s really expensive and difficult for a food company to make food that is real and whole, and much cheaper for food companies to destroy real foods, turn them in molecules, and then reassemble those to make anything they want.”

 

 

Give us today our daily bread. Forgive us our sins as we forgive those who sin against us. Do not bring us to the test but deliver us from evil. Amen.

 

Quote

I pray you'll be our eyes, and watch us where we go.
And help us to be wise in times when we don't know
Let this be our prayer, when we lose our way
Lead us to the place, guide us with your grace
To a place where we'll be safe

 

That's the problem, buddies.

 

As I mentioned before, they could advertise whatever they want these days while getting away with just about any kinda deceptive measures while laughing all the way to the bank.

 

Plenty of us dunno what we're eating, what they've been adding to those "foods" or even how they're making all that crap in the first place. GIGO all the way.

 

Whenever we're listening to music that actually came from somewhere inside the cloud, what kinda transparency (or lack thereof) could one actually expect?

[Currawong]

On 3/13/2024 at 5:16 AM, Norton said:

But is that a hard definition, and if so , whose?  And indeed what do we mean by a Master recording and the implied provenance, if  there can be multiple masters?   For example, if a 70s analogue  recording is transferred by the record company to 24/96 and then distributed as RBCD as a “remaster” and I then make a bit perfect rip of that CD, can I be said to have a lossless copy of the original analogue master?  I’d suggest not and that, far from a hard definition, lossless is just a meaningless  marketing term 

I probably should have specifically said the digital original in my definition, but it was assumed. Indeed though, analog-to-digital conversion is a major issue. For example, Miles Davis' Kind of Blue was transferred poorly (the timing was wrong, for a start!) and this plagued all the digital masters until HDTracks managed to get hold of the original tapes and do a new transfer.