Why go to a lot of trouble in a studio to master a track just the way you want it only to have some ham-fisted DJ mash it up like Khaled?
“… and another one!”
That’s why it didn’t work with the open NI file format and that’s why AI stems are doomed.
Enjoy them while they last because someone out there (and it could be me) is thinking of marketing an audio watermark to defeat stems just like disc copy protection ruined your fun already.
I have no interest in stems but I don’t agree with your sentiment.
If they were doomed, then all of these big software/hardware companies clearly have no clue considering the time, money and effort they are putting into them.
For me stems appear niche to a particular type of DJ like house or hip hop, for all the DJs that I follow and watch I’ve never seen any of them talk or hint at using stems, but then perhaps they do but don’t advertise it.
A post to entice debate. Me likey! Let’s just hope it stays sensible.
So some of my exclusive promos were audio watermarked with personal IDs via Music House a few years ago. They warned you it could affect the audio quality on some sound systems. No thanks.
What would the pay-off from a watermark be? To disable or to garble the sound? Trigger a flag in software to disable the feature? Without the help of software developers it’s dead in the water as they would be needed to be onboard. Copy protection is (mostly) defeated. The labels would have to be onboard to carry any weight.
Playing stems is as legal as looping tracks and FX layering. You’re EQing the sound in a new and creative ways and although works have to be unmodified to play out, they are playing back in software as an unmodified file but with filters and stem isolation FX added.
The only way to shut it down is to do what Tidal did and ask the developers of DJ software to remove stems for their services. You won’t get them willingly take it out.
Besides, stem separation tools are getting better as the years pass. They’ll hit a limit to what they can do but we’re not quite there yet.
Oh and it’s not a fad. Stems are actually usable and enjoyable, unlike 3D TVs which were crap.
Video DJs still play out but lack of interesting content pushed many to LED screens and motion graphics. Not a bad thing.
It’s perfectly understandable from the musician’s point of view, that they don’t want their hard work tampered with outside of their control.
That’s why Native Instruments’ idea of artists releasing pre-separated stems never went anywhere. The artists/labels are not going to just hand over the component parts of their mix to people - DJs or not.
The idea of them having to dig out all the master tapes of their back catalogue, have some engineer separate out the parts then repackage them in another format is unlikely too - it would cost a fortune.
This new on-the-fly method is a little safer from their POV as the full quality isn’t there, and if it’s only done while the track plays, and not stored, I’m sure they would be happier.
There are already so many tracks being sold online which consist of the original track in its entirity, with maybe some extra drums or percussion added - released under a new name, with no credit to the original artist.
That’s bad enough, but then if people can separate out just the drums, just the vocals and so on, the thing could end up being a hybrid of work from multiple artists and labels.
This is why De La Soul’s album was not available digitally for so long. Numerous samples = nightmare to license.
Stems are fun for us DJs but there are way too many people thinking that they can just do whatever they want with someone elses copyrighted work, it’s unbelieveable.
Putting disc copy protection on a Red Book Audio CD prevents them from legally being allowed to put “Audio Compact Disc” on it, and therefore is easy to spot and not widespread. I never expected the NI format of actual stems to get huge, but I don’t see how anyone’s going to outright prevent essentially glorified vocal-remover processing. That’s been in DJ gear since at least old Tascams. We’ve already seen certain stipulations for streaming content DJ services, but that’s it, and those are entirely contrived between the streaming service and your DJ hardware manufacturer.
Lining up those acapellas on vinyl were so sweet when everything just fell into place beautifully. When the pitch, key and phrasing are just on point there was nothing better.
Yeah i like the idea of it, but it’ll be the same processing as the various apps that do it, and frankly the sound of separated stems is generally wishy-washy like a low bitrate mp3. I can’t see many sound engineers being happy with a DJ playing AI-stems on decent sound systems.
I’ve used the apps to create them for my own remixes etc, but i’m also processing them and fixing parts, I cant see how you can really do that live in the mix
Stems are far from doomed Dj Pro has shown the next phase of what can be done, hardware limitations are a factor but are not likely to be the case for much longer.
Interestingly at the time of writing Phil from DDJT expressed Engine Dj stems maybe ‘parked’ due to not being able get stems to sound good on the current Denon Gear, not sure if that is true as the P4+ main selling point is stems.
Maybe Denon need more time to refine it in order to catch up to where the current standard is. However given Denon/Engine have brought features that were said previously could not be done e.g sampler I can’t see Stems being a stumbling block. Denon/Engine surely had/have done RnD to know if it is possible or not on the current gear. Anyone who has currently purchased a P4+ may be very unhappy if Stems do not arrive on that device and are not to a decent standard.
People in the live comment feed claiming Stems on Denon (InMusic) standalone being ‘dead’ is not so great though. Looking around the WWW it looks like that is fast becoming the opinion being spread around
No - the reason Prime 4+ users got access to the stems first is that Denon DJ wanted a smaller user group for the testing. Denon did not introduce stems to sell the Prime 4+.
If some DJs only bought a 4+ because of the stems beta, that’s their own fault.
We all know that stems are coming to the Akai MPC series…and they have the same processors as the Prime products. Currently the plugin is only for the desktop MPC software, and has good feedback on the quality.
Akai must be aware that if the MPC standalone separation is worse, their users won’t be happy, so I’d assume their aim is to make it as good - and if they can manage that on the MPCs then it should translate straight over to the Prime stuff.
Denon specifically choose the P4+ as it’s product to get/launch it’s version of stems it is the only product Denon has confirmed that will get stems (currently) having a small beta sample pool is by the by it could have been the either of the Sc live products but they choose the P4+
so by default it is a selling point and is the main selling point given the only other key difference between the P4+ and the original is having amazon music the rest is aesthetic.
If InMusic wanted a small group to carry out tests on stems, they would have done better to choose to do so on SC6000/M or on prime 4, as there are a lot of really active long-time users here who could have experimented and given feedback.
By announcing it as a new feature specific to prime 4+, InMusic only succeeded in attracting curious newcomers who just wanted to try it out.
But in any case, I’ve always said that current prime hardware would never have the computing power needed to generate a decent stem separation without artifacts, at least equivalent to that of serato.
In a closed beta, this would have involved only a small group. Beta users on Sc6000 or prime 4 are not so numerous, but they are users who invest a great deal of time and effort in feedback on feature development.
The exact opposite of what people who downloaded the 4+ stems Beta have done, and my number one issue with seeing complaints on here about the feature.
I know Stu, and I totally agree with you on that. That’s why I say a closed beta on a small group of long-time beta users would have been much more effective. Especially since everyone knows that in reality, apart from the amazon chip, the hardware is identical in the sc6000 / prime 4 compared to the prime 4+.
As you can imagine, if prime4+ had a specific hardware for stems, the R&D team would have implemented an adapted processor in the specifications to ensure artifact-free stems.
