TV Mix Stems

[Dallas Taylor]

I just started working at a network where I am mixing the following mix stems simultaneously:

Domestic---
INT 1 - Program Left
INT 2 - Program Right
INT 3 - Full Mono Mix
INT 4 - Mono M&E

Splits (for live tv packages/international mixes)---
INT 5 - Dia
INT 6 - Mono FX
INT 7 - Music Left
INT 8 - Music Right

The turn around times are very quick (3-4 hrs for a 30min network show) and I need to mix these simultaneously. I export these stems to a finishing editor. I am trying to create the most simplified template I can. Does anyone have any suggestions?

Here is what I am thinking so far:

Bussing all Dialogue, SFX, Music, ect to their respective stereo busses (Dialogue buss, SFX buss, and Music Buss). Bussing those to an Audio Track representing all of the stems. (INT 1, 2, ect..) Change the output depending on what I want to monitor. Put the tracks into record. record and export. My biggest problem is continually changing the output depending on what I want to monitor.

Does anyone have any more simplified ideas?

Dallas

[mr.armadillo]

You could send the dialogue, FX and music busses via multiple buss sends to the INT 1, INT2 busses ... Simply hit "solo" on the INT buss you want to hear.

[TVPostSound]

With a 3-4 hour turnaround, you shouldnt have to monitor anything but the mix stems, all the others are derived from those.

[tomcat]

I second that. When I do TV with these kinds of delivery reqs. i don't monitor them. When I built the session template at the beginning of a tv series, I monitored the stems a bit to make sure all was well, but since then, I print and deliver the stems without listening to them.

About the session setup, I double-bus all my submasters and reverbs to my 2 mix record track and to each respective stem record track.

[TVPostSound]

if i need to combine stems later,

You have a point there, but in regards to netowrk broadcast mixes:

Most likely the mono stems will go to the promo people (adding narration etc) , and any stereo stems would be used for language localization elsewhere, so most of this would be moot.

[dr sound]

[ and any stereo stems would be used for language localization elsewhere, so most of this would be moot.



[/QUOTE]
Or would that be mute?

[tomcat]

hey tomcat, i do the same thing. then it is all organized and fast like -- i have a template setup for this. ... but i have a question that is a bit basic to this, which applies to a stereo or 5.1 mix. do you use a limiter (or other processor) on your main bus? i do. you know, sometimes that last bit of glue (or sometimes a bit more) makes it just right. but if you do that, your respective stem record tracks will not be exactly represented because they don't go through the final limiter. also, you can't send them individually (an inefficient way to create stems) because the electronic summing will not be the same.

my only solution to this, is to make note of the final limiting settings. that way, if i need to combine stems later, then i can do so with a limiter on the final main bus again. or communicate this to a (re)mixer down the line. is there a more clever way to do this?

This is a problem that I have pondered a bit. Basically, I don't worry too much about the sum of the stems sounds EXACTLY like the the 2 mix, because I don't think they will be used that way. I don't like being sloppy, but it doesn't seem to be a problem, so far.

I limit every stem individually when doing a broadcast mix, because all elements should be brick-walled at -10dbfs.
To do this, I make a set of auxs for each stem and for the 2 mix. I call these "containers" and they each get an L2 set to -10.
There is a summing "2 mix container" that limits the entire mix, then there is a "Dx Stem Container", an "Mx Stem Container", etc... that the respective subs are bussed to. the outputs of the containers go to the respective record tracks, not to the 2 mix summing bus.

When I mix, I try to keep things off the limiters as much as possible, so they have less affect on the mix itself. They are really there just to get me by network QC, so I don't get that nasty phone call in the middle of my next mix saying some levels are illegal.

I don't know if this is the best way to do it, but it is what I arrived at over the years and it works for me, so far. Comments welcome. Marti?

[Rick Sanchez]

I also leave the individual stems unprocessed and only limit or compress as needed for any given delivery requirements. This way, if I have to do a foreign dub, I don't have to deal with pre-compressed or limited stems and will have more overall flexibility later.

If someone else is going to remix at another facility, they have a bit more flexibility and may not have my needs or tastes as far as compression or limiting goes. Granted, the overall mix may not be exactly the same, but the fact that it's being dubbed in another language or for some other purpose negates that point anyway.

[dr sound]

Tom,
If you have ever heard a mix that you've done in another language then you might be surprised. Some countries mix their stuff quite differntly such as lowering the Music and FX by at least 4 or 5 db. I would say whatever works for you then keep it doing it. The thing about Limiting or Compressing your stems individually is that only the FX for example are being Limited or Compressed and not the Dialog or Music etc.
On a side note how does your movie "The Baxter"" sound in the theaters?

[tomcat]

The thing about Limiting or Compressing your stems individually is that only the FX for example are being Limited or Compressed and not the Dialog or Music etc.

Not quite sure what you mean. The 2 mix (mind you, I am only talking about broadcast in this thread) gets limited to keep it legal, but the stems could have an occasional peak over -10, so I put a limiter on them individually too.

I have not had the opportunity to see The Baxter since it was released. It sounded good at the Tribeca Film Fest last year, but I will catch it before it goes away. Why do you ask? Did you see it?