Is Universal Music Group using audible watermarks on digital files?

I had not thought of the automation angle.
As long as they have a good reason...

 

Similar to the Youtube thing, I wonder if they're not concerned about these files showing up in other works - remixed, sampled, or inserted into films/videos, what have you. I'm not familiar with the software they use to identify the watermarks, though, so it's hard to know what's possible. But if you distribute something that's essentially studio-grade without copy protection, it opens up the possibilities of how it can be used, I suppose.

Kind of dumb, sure, but that may be the thought process here. CDs went out without copy protection because, at the beginning, it was basically impossible to make lossless copies of them at the beginning, but as we know, that didn't last. Maybe they're trying to get ahead of that game somehow, at least legally?

 

I assume some people here are buying high res from ponomusic? There are a couple of threads in the community there on this topic. Might be helpful to make some noise over there in the related threads to see if we can get an official pono response (or action) on this.

 

I had quite a few releases in my HDtracks wishlist that simply disappeared from it. Going though and trying to re-add them, I noticed that all the 2015 Rush masters that were previously available in 96/24 are now only available in 192/24. Huh? Does this have something to do with versions that were watermarked being replaced with ones that aren't? Or vice-versa?

 

Still no official response on the PONO community site. And only a few other posts from others on the threads there. Hmmm, doesn't seem to be much of a concern to the community at large unfortunately.

 

They always sell the highest resolution available. With some new releases the 96 and 192 version sometimes show up for a little while before their filters get around to removing the 96kHz versions, so maybe you were remembering those. I've gotten some good deals buying the lower priced 96khz version, only to have it upgraded to 192 for free because of the Pono Promise. So if you see something available at both, you can be pretty sure the lower resolution version will be removed.

I think it is pretty certain it has nothing to do with watermarking.

 

It used to be that they offered multiple sample rates. I have close to zero interest in 192kHz because it is silly, mathematically speaking. 96kHz, on the other hand, my monitors go up to 50kHz, so if there's information up there, I can play it back (you can debate whether that's good, bad, or trivial, but intermodulation is something that happens in the real world and affects what we hear all the time). I'd much prefer to buy the Rush masters at 96kHz, rather than have to downsample them myself. I thought HDtracks was on board with making that an option, but I guess they've changed their policy?

 

Whoops, I had a brain fart there, as I thought we were talking about Pono.

As far as HDTracks goes, yes, they usually offer 96 and 192. Recently there have been some weird cases of the 96kHz versions disappearing for a few days, but they usually come back. Suspect it is just some kind of glitch in their product database/web site.

 

Man, it's annoying to see this thread popping up periodically with people talking about this and that but studiously steering clear of the completely bogus conspiracy-theory topic specified in the thread title.

Yes, watermarking digital files is a thing — but after two years of handwringing and nebulous paranoia can we agree that the encroaching menace of "audible watermarks" is NOT a thing and this entire discussion is looking like wankery?

 

No, don't agree. The listening tests per the site in the OP show its detectable.

It's also not nebulous when it can be objectively shown that watermarking is occurring.

 

I'd say it's dubious (and I'd say the same about the example on Montag's site). A watermark of non-repeating data that runs over nine seconds? Hard to imagine you'd need that much data for a watermark. (If you have studied anything about data security, you'll know that codes or simple messages can be contained in only a few hundred lower-order bits -- barely detectable in a 44kHz stream. Even if they somehow had the technology to encode each download on-the-fly with the individual customer's name and full contact information, that probably would take up less than a fifth of a second.) Either you would find a watermark used once at the head or tail of the file (in which case, it probably wouldn't run more than a fraction of a second), or you'd find one repeating over and over throughout the song (in which case, you'd see a regular repeating pattern that would be clear as day to identify). This looks more like a low level of random noise, which could be caused by any number of reasons, including differences in mastering such as downscaling from one resolution to another.

 

For what it's worth, I got 100% on the listening test at Matt's site. AKG K702s driven by RME Fireface UFX running over FireWire from my MacBook Pro, quiet room. I've noticed these artifacts on Spotify before. Now I know it's not the compression! A friend has paid for Spotify premium and I've been telling him that in many cases, the streaming quality doesn't sound any better. Now I know why!

 

Whether it's a watermark or bad encoding, I noticed it and looked around for a while for discussion about it before this thread resurfaced.

 

Random noise that always begins 1 second into a song, is organized into discrete blocks of fixed time duration, occurs in well defined parts of the differential audio spectrum, and happens to appear in material distributed by Universal? Doesn't seem very random.

Also, no one is suggesting this is per-customer watermarking. It appears to be intended to identity the music as property of UMG. They seem to want something that will work regardless of how a track is sliced or diced or compressed. There is lots of material out there on Spread Spectrum Watermarking and how it works.

 

The links in the thread clear show that: a) the watermarks exist; and b) one of the stated purposes of the watermarks is to combat piracy. There's no conspiracy when everything is in plain sight.

 

That is cause nobody told them to listen for it.

 

Exactly. We can (most of us) hear it now!

 

The watermarks were applied with varying degrees of audibility. With some recordings, you really have to listen quite closely for it to be noticeable while with other recordings (also depending on the type; classical and especially solo piano recordings stick out) it's plain as day. No conspiracy here. No-one ever told me to listen for it. I noticed it all by myself...

 

What do Mulder and Scully think?

 

As far as I can tell this is still ongoing.

Honestly though I don't see any widespread concern about the practice, even among this community. Prove me wrong though, but the low level of concern at Ponomusic.com is a bit disconcerting for example. It's not going to change by itself.

I have tried at Ponomusic (raised support ticket, posted in the community forums) and contacted an impacted label, but really there is very little traction on this. .

If you care, what have you done? Or, perhaps, what can you do

This is a "call to arms".

 

They want to believe...

 

This is from the 2010 Universal The Cure Disintegration 2x12 lp, if Matt's Spectogram is correct this and the compact disc have it also

 

Surely it'd only be logical to watermark each individual disc(or download) so every copy has a unique number linked to an individual person.* That way if a copy appeared on one of the many naughty download sites they'd be able to trace who supplied it. Watermarking them all with the same watermark is totally pointless since all it tells them is that it's pirated - which they know anyway before they've downloaded anything simply because the files are being supplied from an unauthorised source. Under the present system people are going to pirate the files anyway, whether they have a watermark or not.

*Not that I agree with such behaviour, but if you're gonna do it do it right, not in a totally useless half-assed way.

 

That's like saying that barcoding a product doesn't prevent shoplifting... and well that would be right, it doesn't prevent shoplifting - but that is not really the purpose of a barcode in the first place.
Watermarking seems more likely to exist as a way of asserting authorship in a way that cannot be erased (easily). Like assigning a catalogue number to that particular digital release. And it may have some other functionalities for the rights owners and vendors that we haven't contemplated. My point is that watermarking is not necessarily solely about combating online piracy at all.

 

I think those professing that without individual watermarks is useless....miss the power of the watermarks and them being in the audible band means no resampling...no analog recording of....will be able to rid the audio of the mark.

It's enforced my "the player", of which....how many are there, really? Apple, Microsoft, Google. Hardware will be EASY to get on board as long as it's standardized. The three tech giants' mobile OSs are the future. They all force updates regularly now to drive hardware sales. Apple won't let you go backward. You think that the gatekeeper audience is big because you think about the software player, but they all run on very few OSs, none of which have allowed bare metal access in a long time. They could close said loophole tomorrow and no line would know....which is why they need to pollute the ecosystem and spread the files first.