• TalkBass has been independent since 1998. Add your voice.
    Create a free account to reply to discussions, view embedded media, and browse with fewer display ads.
    Join freeLog in
    Want zero display ads or expanded classifieds tools? Compare plans.

Logic Pro 11 and Logic Pro iPad 2

Well the Stem Splitter is pretty darn good. Very happy with it. You can hear some low volume artifacts (clicking sounds) from the drums in the vocal stem but I could clean it up with iZotope RX. I use reference tracks for my projects, so having stems of the those tracks are going to be very useful.

Screenshot 2024-05-15 at 4.32.24 PM.png
 
Well the Stem Splitter is pretty darn good. Very happy with it. You can hear some artifacts (clicking sounds) from the drums in the vocal stem but I could clean it up with iZotope RX. I use reference tracks for my projects, so having stems of the those tracks are going to be very useful.

View attachment 5448696
Which version of RX do you have? I’m curious how stem splitter stacks up against RX’s music rebalance tools which have been the industry standard for this kind of work for several years now. I haven’t yet upgraded to the new Logic, so I haven’t had a chance to do this test for myself.
 
RX 11 Standard. I was using Music Rebalance in RX 10.

Which version of RX do you have? I’m curious how stem splitter stacks up against RX’s music rebalance tools which have been the industry standard for this kind of work for several years now. I haven’t yet upgraded to the new Logic, so I haven’t had a chance to do this test for myself.
 
Just downloaded this yesterday. The software drummers are much better than before both in sound quality and control. Not that excited about the bass session player though, but it's inevitable since there's been other software bassists around for some time. I was disappointed that Apple got rid of the 'personalities' for the drummers (Anders etc.) as I was hoping to see what they had for the bass players.
 
But in post production mix houses for tv and film and the like, they still call everything "stems". For example, 99.9% of the time I send a "drums stem" - a stereo mixdown of the whole kit. But even if I did break the entire kit out and bounce down a stereo file for each piece, they'd still call them "stems". You'd have "hi hat stem", "kick stem" and so on. That's just how the term is used in that industry at that stage of production. All just depends on how much flexibility you want to provide the mix engineer. Usually they want as few stems as possible, but sometimes there's elements in a track that could cause issues at the final mix stage with how it works VO, sound design, etc, and they need to be able to adjust it.


That really depends on the age, and experience of people you're working with.

If they got into the industry after about 2005~ (when all digital workflows really started to become the norm), and they weren't properly mentored by experienced veterans, they're probably using the term incorrectly.

Generally the stems (complete mixdowns of groups of sounds i.e. All the drums as a stereo file / all the guitars as a stereo file / all the Foley as a stereo file / all the music as a stereo file, including all automation) should combine with the faders at zero, and no additional processing, to exactly match the final mix.

There are some exceptions, when delivering M&E / mix minus for a foreign language dialogue track, you wouldn't want automation included that helped the English dialogue sit in the mix (like ducking ambience etc) that would have no bearing on the mix for a foreign language.

Probably worth looking up the SMPTE standards and delivery specs, for a complete set of guidelines re. terminology, as well as DAW session hand off, and archiving (although last time I looked, which was about a decade ago, it was entirely focused on Pro Tools, but the principles would still apply to Logic Pro).
 
  • Like
Reactions: JRA
That really depends on the age, and experience of people you're working with.

If they got into the industry after about 2005~ (when all digital workflows really started to become the norm), and they weren't properly mentored by experienced veterans, they're probably using the term incorrectly.

Generally the stems (complete mixdowns of groups of sounds i.e. All the drums as a stereo file / all the guitars as a stereo file / all the Foley as a stereo file / all the music as a stereo file, including all automation) should combine with the faders at zero, and no additional processing, to exactly match the final mix.
This is how I've been doing it for the last 20 years for TV, film, and advertising work. But as I say, sometimes we break out specific individual things if we want the flexibility at the final mix--like a certain element that might conflict with voiceover or whatever. But we still call that a "stem". So you might have "drums stem", "guitars stem", and also "lead guitar melody stem", or something like that, if that lead part might need to come down (or out).
I always provide all that--the stems and a final mix representing the sum of the stems with no additional processing. I also typically send a version of the final mix with mastering, as that's usually what the client has approved (after all, these days everything has to "sound like a record" - even initial demos!). Much of the time, my mastered mix is what gets used, but if the mixer needs to futz w the stems, then he at least has that mastered "client version" to refer to. Where it gets tricky, is when there's a project that requires specialized processing on the master buss---for example some lo-fi things, or recreating the sound of old recordings in a period-correct way. I just did some stuff for a Netflix production that had to sound like it was from the 30's, and you have to process the full mix to achieve that authentically. Stems just don't work well in that scenario, and you can't expect the final mix guy to take the time re-invent all the weird tricks you did to get that sound (unless bizarrely he has all the exact same plug-ins you have, also uses Logic(almost never), and you can send him the channel strip settings). So in those cases, I'll usually send alt versions of the full mix that might be handy.
 
Last edited:
As an amateur in a dad band, the new stem splitter is a killer feature for a couple reasons:

Cleaning up live recordings. If the band is working on a new original song, I'll record a basic multi-track during rehearsal. Usually there's a ton of mic bleed in the vocal tracks, especially since our drummer also sings. The stem splitter does a great job removing the drums from his vocals and makes it way easier to create a decent demo without having to re-track vocals.

Learning new songs. If we're working on a new cover, I can import an mp3 into Logic, run the tempo analyzer, run the stem splitter, add markers outlining the song structure, and mark up the chord progressions (another great new feature). This takes maybe five minutes and then I have a fully annotated track that I can play along with, either muting the original bass or soloing it to work out tricky passages.

Screenshot from a learning track I put together for Just What I Needed. Took me probably ten minutes to put this together in Logic and now the song just clicks in my mind.
Screenshot 2024-05-23 at 8.26.51 AM (2).png
 
I updated last weekend on my Intel Mac, which means I don’t get StemSplitter or ChromaGlow, which are by far the most talked about features currently. That’s fine, as I have third party tools that do these things already.

The rest of the update is fine. No new bugs or crashes. I think the UI might be slightly darker. They still haven’t fixed PDC or automation accuracy despite many of us having requested it for literal years.

The new session bassist and keyboardist being added alongside the existing session drummer are incredibly white bread. It’s basically just a selection of pre-written MIDI phrases that can be made more complex or dumbed down with a slider and a few articulation tools. Despite Apple’s insistence in marketing these as some sort of AI-driven tool, I don’t think that’s the case. They’re fine for scrapping together some ideas but I suspect the limits of their vocabulary will become apparent very quickly as we start hearing corporate backing music in commercials on YouTube, et al having been made from this will all start sounding even more similar than it already does.

So, in summary. Logic is still my favorite DAW, if you don’t have an Apple Silicon Mac you can look to third party developers for the most highlighted features that are unavailable to you, don’t expect magic from the session player tools, Apple remains unwilling to fix a couple of fundamental problems with Logic that have known about for years, and all told I will continue making music with Logic for, seemingly, years to come :smug:
 
Any thoughts on Stem Splitter versus Music Rebalance in terms of quality, ease of use, etc.?
I like Music Rebalance for removing stuff. Especially when I want to practice with an existing track, it's usually easy to remove the bass. Then I'd just export an MP3 for headphone practicing.
But Music Rebalance seems relatively slow to me; RX10 Advanced on MacBook Air M2.

Earlier this week I tested Stem Splitter for the first time by splitting our live recording from last weekend, recorded with the cheap small Zoom H1n using its built-in stereo mic only.
The Stem Splitter workflow is amazing, and considering this was 1 hour worth of music, also surprisingly fast. I haven't tested it, but it likely would have taken over an hour to split using RX10 in the best quality setting.
This band is a trio, drums, bass (me) & guitar/vocals, so I ended up with individual instrument tracks.
Now… my individual solo'ed tracks didn't sound "great" because the source material isn't good and clean; it's just a stereo room recording. But playing them back together sounded the same as the source while I was able to eq, compress and balance the individual parts.
Now I'm seriously considering to rework many of my old live (re)masters using Stem Splitter…
 
Here's a another review from Why Logic Pro Rules. He compares LP Stem Splitter with iZotope RX, Acon Digital Acoustica - Remix, and the original stems from the project. At 16:00 he does a direct comparison with LP Stem Splitter and the original stems from the project. The conclusion is that there's not a lot in it between Acoustica Remix and RX, and LP Stem Splitter. The obvious advantage of LP Stem Splitter is that's it's free and a LP plugin. You can rent RX for a monthly subscription for $19.99. For me as a learning tool, these stem splitters are invaluable. I always add reference tracks and stems to my projects. As for the Bass and Keyboard Session Players, not that impressed by them. Would be useful in making demos if you're a solo composer/producer though. ChromaGlow saturation tool sounds pretty good. I've started adding it to my MixBus.

 
Last edited:
  • Like
Reactions: loukash
ChromaGlow saturation tool sounds pretty good.
Yeah, I just started to experiment with it yesterday after I've finally received the multitrack recording from our show two weeks ago, and I began with the first "quick'n'dirty headphone mix draft" last night. For the start, I put one instance on the drums bus and another one on the "submaster" bus. I liked what I heard! (Until now, usually I would be using the Retro Reel block in the Helix Native plugin for a similar effect.)
 
  • Like
Reactions: Krisfield
Here are a few clips of the stem splitter in action. With the drummer singing lead vocals, the splitter removes the drum bleed in the original recording and makes it way easier to process.

I'm just making rough demos for an amateur band, but it makes it more enjoyable for me, which means I'm more likely to work on the recordings and maybe even practice my parts.

Vox as recorded with drum bleed:
Vox run through stem splitter:
Vox in the full mix: