Dry Stems Question: What are they and how to prepare and present them?

[DogsoverLava]

I'm getting ready to launch my Stairway to Heaven Project website and as part of it I am going to post/provide dry stems for people to download and play around with if they like. I recorded everything with lots of headroom so my understanding here is that I need to render each stem at the null point (zero) on the mixing channels which will give them all the same headroom I had, with all my FX turned off (certainly any 'verbs etc). Should I equally turn EQ off? How dry should my stems be? Do I also need to center (right/left) them before I render them? (my understanding is that a mono track is a mono track regardless and the left/right has to do with the send to the master bus or playback.

I'm a little scared to post such naked tracks for people to have access to (still quite self conscious) - particularly with no EQ at all. Is there a downside to rendering the stems with EQ beyond the obvious?

Many thanks

 

[pcrecord]

If you want people to be able to mix their own version, the tracks should be dry, no effect no eq nothing.. what the converter grabed. (of course if you did something in the way in that couldn't be removed.. but that's ok
Yes mono tracks should not be rendered as stereo tracks.
The only important rule is that every track should start from the same point in time in the song. If you have a solo at the end, it should be a track with silence up to that point.
That way people don't have to time align your tracks

In Sonar, there is a fonction called Bounce to clip, which makes this easy.. I can deactivate all effects, select all the tracks and bounce them at once. this creates a series of files I can copy somewhere else and then I don't save when I exit Sonar, everything goes back how it should be...

You mind telling why you want to do this ?

 

[kmetal]

To add to what Marco said which was well put, you may want to print a couple of 'efx' busses and for stems just in case.

 

[DogsoverLava]

If you want people to be able to mix their own version, the tracks should be dry, no effect no eq nothing.. what the converter grabed. (of course if you did something in the way in that couldn't be removed.. but that's ok
Yes mono tracks should not be rendered as stereo tracks.
The only important rule is that every track should start from the same point in time in the song. If you have a solo at the end, it should be a track with silence up to that point.
That way people don't have to time align your tracks

In Sonar, there is a function called Bounce to clip, which makes this easy.. I can deactivate all effects, select all the tracks and bounce them at once. this creates a series of files I can copy somewhere else and then I don't save when I exit Sonar, everything goes back how it should be...

You mind telling why you want to do this ?

I have found that I learn best when I view each project I do as a tutorial (for myself) and I document what I've learned through the process. Because Stairway is so beloved, and because I benefited from other people's work and resources on the web while putting it together, I decided to give-back by way of sharing my process so that others could benefit from what I did. I thought I'd create a website documenting the project itself and to serve as a hub for others looking to record their own Stairway Projects (where they can gather) and as a central location to find resources that would similarly help them. In this way it's sort of supposed to be a multi-faceted Stairway to Heaven celebratory and resource hub. It's in live alpha/beta stage right now as I flesh it out and populate it with content. (thestairwayproject dot com).

Sharing the stems (wet & dry) with those that might want them is part of the learning ethos I wanted the project to have. The only thing I wont share at this stage is the dry vocal stems as I have not sought permission to share those from everyone who cut a vocal for me (I had several). I can imagine that someone might want to try their hand at mixing them, or possibly replace some of the stems and incorporate them into their own version or maybe just use them as a time base or guide tracks for their own version.

Thanks PC - you confirmed what I thought about the stems.

 

[DogsoverLava]

To add to what Marco said which was well put, you may want to print a couple of 'efx' busses and for stems just in case.

I can think of a couple of places where this would be good - particularly with what I did for reverb on the recorders and acoustic guitar -- Though those two parts are panned hard left and right, each sends a bit of the reverb to the opposite side so I can see myself printing a few fx buses as tracks for them to use.

 

[dvdhawk]

You might not want to invite Randy California to participate, it might come out sounding just slightly different.

 

[DonnyThompson]

You might not want to invite Randy California to participate, it might come out sounding just slightly different.

 

[Brother Junk]

I'm a little scared to post such naked tracks for people to have access to (still quite self conscious)

I feel ya. I have a song right now that is still in the rough stages and would like to show it to some people who know music better than me (like here) but, it's a bit scary. I know I'm not that good, so the cricitism part can be scary. I'll get it a little but further and then man up.

In Sonar, there is a fonction called Bounce to clip, which makes this easy.. I can deactivate all effects, select all the tracks and bounce them at once. this creates a series of files I can copy somewhere else and then I don't save when I exit Sonar, everything goes back how it should be...

Logic does the same, you just select 'bypass all plug-ins' I actually dislike the PT process for exporting stems.

 

[DonnyThompson]

I feel ya. I have a song right now that is still in the rough stages and would like to show it to some people who know music better than me (like here) but, it's a bit scary. I know I'm not that good, so the cricitism part can be scary. I'll get it a little but further and then man up.


Logic does the same, you just select 'bypass all plug-ins' I actually dislike the PT process for exporting stems.

Don't be afraid to post your mixes. While some critiques may seem a bit harsh, ( nobody ever likes to be told that their baby is ugly, LOL) it's never intended as a personal offense, so you can't take it as such.
RO is a problem-based learning forum. That's what we're all here for. It's the difference between sites like GS, where unless you have a half-million dollar rig, nobody wants to talk to you or to listen to your work.
At RO, we learn by analyzing problems and then helping each other to solve them... and BTW, if you think that veterans don't need the same assistance from time to time, you're wrong.
I've been at this craft since 1979, and I still turn to my colleagues here for suggestions and analysis.
You are never too old to learn, and you can't improve if you don't let others with experience in the craft tell you what they hear, with potential issues, and suggestions for improvement. The problem may be in the mix, or it may be in the initial recording, it may be in the gear, it may be in your room...or it may be something missing in your knowledge. Until you are willing to grow a thick skin and lay it out for others to hear, you'll only ever improve "so" much.

Exporting stems/tracks for remix:
Don't forget to properly name or number your stems/tracks, and make sure that you render the entire track into one corresponding file so that it can be easily imported into another workstation, without the engineer having to load different sections or objects, and having to guess where things go on the timeline.

I know these things sound obvious, but you'd be surprised at how many times I've received stems and tracks that weren't what they were labeled as, or were all over the place on the timeline.
I've received kick drum tracks that were labeled "piano"... bass tracks labeled "guitar"... horns labeled as "synth"...
I've also had tracks that were one instrument for a certain length of time, and then inexplicably, out of nowhere, turned into another instrument entirely.

Dry/raw tracks are always best. If you add processing, you are taking away the remix engineer's ability to do what they want to do.
Output the tracks without any GR, no effects - unless a particular track was recorded with an effect; such as a guitar with delay, or chorus... and if this is the case, make sure you notate it as such so that the mix engineer knows this.

 

[Brother Junk]

I've been at this craft since 1979, and I still turn to my colleagues here for suggestions and analysis.
You are never too old to learn, and you can't improve if you don't let others with experience in the craft tell you what they hear, with potential issues, and suggestions for improvement. The problem may be in the mix, or it may be in the initial recording, it may be in the gear, it may be in your room...or it may be something missing in your knowledge. Until you are willing to grow a thick skin and lay it out for others to hear, you'll only ever improve "so" much.

That makes me feel better (that you still ask for opinions). It's like you said, no one wants to be told their baby is ugly. But for someone with very little experience like myself, I'm kind of resigning myself to asking, "How ugly is my baby?" lol

Don't forget to properly name or number your stems/tracks, and make sure that you render the entire track into one corresponding file

When I do it , I typically do this "Good Fight, Kick 1 Dry, 80 bpm, C#m" Is there anything else that would be useful info in my labels? I also go through and check every track and label at the end bc I also hate getting tracks that are all mislabeled.

so that it can be easily imported into another workstation, without the engineer having to load different sections or objects, and having to guess where things go on the timeline.

I really hate that. I don't want the parts to rebuild your whole song, just make every track start at 0:00 If I want pieces I'll cut them up myself. I've never said anything though bc I wasn't really sure what the "norm" is. But it sounds like the way I do it is right.

unless a particular track was recorded with an effect; such as a guitar with delay, or chorus... and if this is the case, make sure you notate it as such so that the mix engineer knows this.

In those examples, how would you note it? Is "dry" insufficient?

Or is it just so that he doesn't contact you looking for the dry?

 

[DonnyThompson]

When I do it , I typically do this "Good Fight, Kick 1 Dry, 80 bpm, C#m" Is there anything else that would be useful info in my labels? I also go through and check every track and label at the end bc I also hate getting tracks that are all mislabeled.

I would label simply:
01:kick
02:kick trig ( or whatever)
03: snare mono
04: side stik
05: hh
06: toms stereo

etc.

You don't have to name all the tracks with the title or the BPM, rather, put them all into a file named for the song, and tell the remix engineer what the key sig, time sig and BPM are in a side note. You could notate 'wet" tracks as such, but most mix engineers are gonna want dry stuff anyway. If you've got verb or delay or even compression printed on the track, the remix engineer can't take it away... it's like trying to un-bake a cake.

 

[DonnyThompson]

addendum..

if You've got things like guitar tracks that were done ITB using a VSTi amp sim; like Amplitube or Scuffman, or Slate , make sure you also send a clean version of that guitar track; so that the engineer could re-amp through another amp sim ( or even a real amp) if they feel it would be better for the song...

If you have bass or guitar that was recorded with deley, chorus, etc., send them that, but send them a dry version of it as well.

-d.

 

[audiokid]

I receive stems in a zip package. Inside the main package are folders which may or may not have notes about rules, tempo's whatever in each folder. The folders are usually broken into drums, bass, vocals, strings, guitars, keyboards, background vocals, percussion, effects etc. Just like we organize our groups in a mix.
The stems are usually in line and can be dry or processed. They import into most professional DAW platforms without issue.

If you want to get into stem mixing, I have been part of Indaba Music for years. Its all about stem mixing and participating in some really fun contests.

Once you are a member, you can receive packages from indie clients and record labels, you might even enjoy some of the music too.

I like stem mixing and remixing.
Stem mixing other peoples music can be fun, educational, sonically wonderful or horrible. Its a great way to improve your skills through problem based learning. You learn a lot about sound mixing other people's stuff.
I have done remixes from really great unknown groups, Telefunken givaways to platinum artists like Christina Aguilera.

 

[audiokid]

I had a great Indaba deal setup for our members a few years back... I think 2 out of about 25,000 members saw it as a good way to learn more about mixing and production.

If we got the numbers of interests up here again, I could possibly inspire Indaba to offer something up for us again.
There is a big stem mixing world out there. It makes for some really good discussion here too. What better way to learn about mixing music, especially when the songs are of the same level as hits being produced on the radio.

 

[kmetal]

I never mixed stems before. Any of the tracks I've mixed that haven't been by me have been full multitrack.

Does anyone find stems limiting relative to the full multitrack?

I feel like I'm behind the times or missing out on something here.

 

[Brother Junk]

I had a great Indaba deal setup for our members a few years back... I think 2 out of about 25,000 members saw it as a good way to learn more about mixing and production.

If we got the numbers of interests up here again, I could possibly inspire Indaba to offer something up for us again.
There is a big stem mixing world out there. It makes for some really good discussion here too. What better way to learn about mixing music, especially when the songs are of the same level as hits being produced on the radio.

That sounds extremely interesting to me. However this...

I never mixed stems before. Any of the tracks I've mixed that haven't been by me have been full multitrack.

Does anyone find stems limiting relative to the full multitrack?

I feel like I'm behind the times or missing out on something here.

Maybe I don't understand truly understand what stems are? Vs. Multitrack?

 

[kmetal]

Stems are 'sub mixes' or groups of instruments bounced down to a single track.

So the engineer would mixdown all the drum tracks to a single stereo track. Mix all the guitars down to another single stereo track. Background vocals same thing.

This way instead of passing on each track in the Session to the mix/mastering engineer, you pass along a smaller set of tracks.

This keeps things concise and allows the mastering guy or whoever to adjust levels of certain things. So if a track gets to mastering and the ME feels the vocals are to soft he can raise them up with a volume fader on the vocal stem track. This is easier and allows for less afrtifacts than say trying to use eq or compression or Mid/side techniques to bring the vocal up.

The ME may not have control over each thing like say overhead or room mics, becuase those would/may be part of the drums stem. But that depends on how things are arranged and organized.

Stems aren't just for mastering.

You may want to consolidate 14 background vocal tracks down to a single stereo track once they are edited and balanced with each other. Perhaps to keep things organized in your session or to save CPU resources.

 

[DogsoverLava]

Thanks for your help guys - I followed all your good advice.

I now have all the stems (wet and dry) available for download (including DI guitars/bass as well) for anyone that wants to use them at the project website (thestairwayproject Dot com). They may never ever get downloaded by anybody, and no one may ever visit the site, but it was all part of the project ethos so I had to make them available.

 

[DonnyThompson]

I never mixed stems before. Any of the tracks I've mixed that haven't been by me have been full multitrack.

Does anyone find stems limiting relative to the full multitrack?

I feel like I'm behind the times or missing out on something here.

Well, one of the reasons, I think, is that stems can be more challenging, Kyle... and it's in that spirit that they can be more fun to mix, because you have to use your skills, knowledge and experience, and push your talents to get good mixes because you don't have the luxury of having each individual track to work with.

I think it also harkens back to when track counts were limited, and for some, there's an attraction to that limitation, because more attention was paid to capturing and mixing performances as opposed to overdubbing each and every single take on every instrument.
If you think about it, the engineers who were working with The Beatles, The Stones, The Who, Steely Dan, Alice Cooper, The Police ... ( Geoff Emerick, Ken Townsend, Glyn Johns, Roger Nichols, Tom Dowd, Andy Johns, Sam Phillips, Eddie Kramer, Alan Parsons...) of all those classic bands/artists... were all very often working with "stems". They weren't called that at the time - back then they were referred to as "comp" ( compilation) tracks - and while it was indeed more challenging to get good mixes that way, so often the magic of the performances were preserved... there was a "gel" to comp tracks like that... (in some cases, even the bleed could be advantageous).

I look at today's modern recording and mixing methods as kind of a double-edged sword... the benefits are that we have this marvelous technology.... we have an unlimited number of tracks, copy-paste editing, the ability to correct mistakes, EQ that often borders on "forensic" style filtering, giving us the ability to zone in on just one frequency, anywhere in the bandwidth, and adjust...

The other side of that blade though, is that having limitations often forces the bands and engineers to get the best performances possible, because when you have those boundaries, there really is no "fix it in the mix". Sometimes, having limitations can be a good thing - because you don't ( can't) spend hours and hours removing the human element of the art... getting rid of every single breath on a vocal track, wiping out fret noises, or editing with fades and cuts for "the perfect take", which, while on the surface may be technically attractive to us as engineers, can so often suck the life and the human element out of the performances.

Speaking with full honesty here - I consider myself to be a fairly competent engineer; I know which side of a fader is up, I can tell the difference between 100hz and 1k, I can set up an XY coincidental pair over a drum kit without giving it a whole lot of thought... I can get audio from Point A to Point Z and make it sound pretty good; so I'm probably about "par for the course" in terms of ranking as a studio engineer. I'm "average".

But... I'm nowhere near as good as I could have been. Like many engineers, I've often thought about how cool it would have been for me to have been an active, working engineer back in the 50's and 60's, at places like Western, Goldstar, Abbey Road, Trident, Chess, Sunn, Sigma Sound ... But ya know, I honestly don't think I could have held my own against those engineers I mentioned above. I'd have been lucky to achieve the rank of "Gopher". Those guys mentioned above were ( are) serious bad-asses when it came to recording and mixing records... and for as much as I know now, digital recording has made me... hmmm... how should I say this... well... "lazy".
These days, the DAW does so much of the work for us. I don't believe that I now use as many of the skills I was taught back in the days of LF consoles and tape machines, because in so many cases, I don't have to anymore. I'm not as challenged as I once was... and I think my mixes have suffered because of this.

Stem mixing brings back those challenges a bit. The fun is in pushing yourself, and getting the best results possible with far less options than you would prefer to have.

IMHO of course
-d.

 

[Brother Junk]

Stems are 'sub mixes' or groups of instruments bounced down to a single track.

So the engineer would mixdown all the drum tracks to a single stereo track. Mix all the guitars down to another single stereo track. Background vocals same thing.

This way instead of passing on each track in the Session to the mix/mastering engineer, you pass along a smaller set of tracks.

This keeps things concise and allows the mastering guy or whoever to adjust levels of certain things. So if a track gets to mastering and the ME feels the vocals are to soft he can raise them up with a volume fader on the vocal stem track. This is easier and allows for less afrtifacts than say trying to use eq or compression or Mid/side techniques to bring the vocal up.

The ME may not have control over each thing like say overhead or room mics, becuase those would/may be part of the drums stem. But that depends on how things are arranged and organized.

Stems aren't just for mastering.

You may want to consolidate 14 background vocal tracks down to a single stereo track once they are edited and balanced with each other. Perhaps to keep things organized in your session or to save CPU resources.

Ok, that's about what I assumed. So, stems are when I split my drum kits midi into 8 tracks and send those individually, snare, kick 1, kick 2, hh etc.

Multitrack would be if I sent the entire drum kit?

The reason this confused me is because I've never seen it done any other way. I've never gotten a drum track that was ALL the drum sounds in one track, it's always split up. Vocals, I make them give me every track separate. But this is highly likely bc you guys are RE's and a lot of stuff I get is midi, so it's easy to split.

Sorry for the pause in conversation, I just wanted to make sure I was talking about the same thing you guys were.