Use 'Stems'?

[legionserial]

Firstly, I'm going to start with the disclaimer that if I use the word "master" in this post, I am aware that there is more to mastering than what I am describing. I'm aware that I'm only talking about one song as opposed to a whole collection of songs. Bear with me. I'm not always too good with words. Part of me is also wondering whether this should go in the mixing thread, but I'm here now, and it is kinda relevant.

So...I've got this tune I'm working on. It's a sort of death/thrash metal tune (those prefixes have lost all meaning to me at this point really), but it starts with an acoustic intro that the rest fades in over the top of.

This isn't exactly an uncommon thing in metal, aswell as a similar thing for other genres, so I'm wondering what the best way to do this would be. For this type of thing, is it considered normal to mix the acoustic intro as one 'stem', and the rest as another, and fade the second stem over the first in mastering, or would this normally all be done in mixing with individual fades for each instrument?

I ask because I'm aware that standard fade-ins and fade outs of entire songs are best left to the mastering stage post final compression...something I discovered when adding final compression to an already faded mix. Obviously it didn't fade smoothly. Is this likely to be an issue for what I'm trying to accomplish? Is there something that is generally considered an optimal way of doing this?

Thoughts? Ideas? Flames?

 

[karumba]

i think you missunderstood the term "stems" (with respect to mastering). stems are basically sub-groups of your mix. for mastering e.g. this could be:
- drums
- bass
- vocals
- guitars
- rest
all above are stereo-tracks including all FX for theses tracks. if you playback those 5 tracks, you get your mix. in mastering, you have now the possibility to fix problems which you normally can't (without compromises), e.g.
- vox to loud/low
- deessing of the vox
- bassdrum to low compared to bass
- etc.

so in principle level & frequency problems in general. so i would recommend: just mix as you would normally do & don't think about stems during mixing.

 

[Massive Master]

I'd send them as two individual pieces - but with an accompanying 'one-piece' mix that's lined up exactly how you're hearing it in regards to timing and relative volume.

 

[legionserial]

i think you missunderstood the term "stems" (with respect to mastering). stems are basically sub-groups of your mix. for mastering e.g. this could be:
- drums
- bass
- vocals
- guitars
- rest
all above are stereo-tracks including all FX for theses tracks. if you playback those 5 tracks, you get your mix. in mastering, you have now the possibility to fix problems which you normally can't (without compromises), e.g.
- vox to loud/low
- deessing of the vox
- bassdrum to low compared to bass
- etc.

so in principle level & frequency problems in general. so i would recommend: just mix as you would normally do & don't think about stems during mixing.

Yeah, I know what stems are. That's why I put the word 'stems' in inverted commas like I did with 'mastering'. I'm neither using true stems or doing any true mastering.

I'd send them as two individual pieces - but with an accompanying 'one-piece' mix that's lined up exactly how you're hearing it in regards to timing and relative volume.

Thanks, that's along the lines of what I thought. It'll be me that's 'mastering' it so no need to send anything anywhere. Theoretically, anyway. If I can be bothered. The tune itself isn't too great really but functions a good subject for practice in general because it's very dense with lots of changes in dynamics and timbre. It was just something I was musing on this morning. Cheers.

 

[Waltz Mastering]

I think either way would work. It's like 6 of one / half dozen of the other.

Mix the intro separately or mix it with the song... wouldn't take long to try them both and see what you think works best.

 

[Farview]

What I do is create two (more) sub-groups/group channels/busses What ever they are called in your DAW and route everything that is fading out to one of them and everything that is fading in to the other and just automate them.

The main trick is to keep the acoustic part quieter than you think you want to. Essentially keep it out of the mastering compression, so when the big loud part comes in, there is still a jump in volume and density even once it hits the limiters.

 

[Greg_L]

I think it's best to scrap the acoustic part because they're wimpy and go balls out all the time.

 

[legionserial]

I think it's best to scrap the acoustic part because they're wimpy and go balls out all the time.

Balls out is ballsier when it's constrasted against something wimpy. Or something like that anyway.

 

[karumba]

Yeah, I know what stems are. That's why I put the word 'stems' in inverted commas like I did with 'mastering'. I'm neither using true stems or doing any true mastering.

ok, maybe i missunderstood your question, but what i wanted to describe is, that stems are normally not separated in time (like parts of the track e.g. intro / rest of the track) - of course you could do so, but i don't know if it makes sense. if only your intro has a specific problem, it could make sense. but normaly, stems should be grouped in exactly the way the ME tells you, so i normally request a mp3 of the normal mix (masterbus). then i decide which problems it has & which grouping makes sense - then i request the mix with these specific stems. e.g. if only the level of the vox is too low, but the rest of the mix is fine, then i will only request:
- stem of vox
- stem of the rest
from my client.