Distance and Space for DIA in a mix

[meathman]

hi,

I was watching a lower budget horror film yesterday, and the dialogue editing sounded fine, it was clean and very distinctive, but also very dry.

The problem I noticed was that the 'closeness' and 'dryness' of the dialogue always stayed the same. For example: there was a scene where there would be a close up of a conversation which would sound fine, then, the camera would cut to another more distant angle, a wide shot for example, but the dia stayed totally dry and upclose...it just sounded funny.... another thing that happened was a character would spin around in circles, but the line was always totally in front and dry.... Im guessing most of the recording was done using lav mics...

This made me think about something ive never really considered.. Do dia editors usually always split for camera perspectives? also, how do the mixers handle this kind of situation....

[danijel]

i recently did a short where we decided that there will be no perspectives in dialogue/sfx during the story-board phase. characters speak at great distances from camera in certain scenes, so i advised director to shoot everything with lavs, as it would be strange to switch sound style from scene to scene.

there's a good scene in 'la haine', where main characters walk at great distance from camera, and you hear them (close-up) cursing policemen which are actually standing in front of the camera, out of focus, inaudibly speaking.

can you tell if there was a reason for this approach in the film you've seen yesterday?

[meathman]

honestly, i think this approach was used because the mixer was not very good, it was obviously a low budget film and i think he just didnt know how to handle the dia..

with that said, if lavs were used for the entire film, and you did want a more natural 'distance' for pers. cuts, how would you handle it? Would you put all the pers. on the same seperate track, and simply have them a few db lower?

And considering most of the cuts were outside, would you apple reverb, or simple roll off some low end with eq?

[ggegan]

I mix FX, not Dia, but I can tell you that every mixer has different preferences, so it's a good idea and just common courtesy to chat with them before the mix so that you can get an idea about how they like things laid out. After all, one of your your goal as an editor is to make the the mixer's job easier and in order to do that you need to get a feel for their work methods.

I personally hate it when editors split FX and BG tracks for every perspective because it really destroys my flow and causes me more problems than it solves. It also has a tendency to multiply the number of tracks and get confusing. I want to spend as much time looking at the screen and as little as possible looking at cue sheets or computer monitors trying to figure out what's going on with the tracks. If I really need a split when there isn't one it's isn't that much trouble to just do it myself.

The other thing about splits is that that they are very abrupt, which can be ugly in the middle of a word or effect. To my ears a quick fade sounds better than a hard cut, but that can be done with a split with a 3 or 4 frame crossfade between tracks. I even like those kind of transitions in BGs for scene changes because the audience still hears it as a cut, but it is less jarring and doesn't call as much attention to the mechanics of the filmmaking. That's just my personal mixing style, I like round edges and fluidity unless the dramatic content demands a jarring impact or I need to underline a change.

In general I think it's a mistake to think too literally when it comes to perspectives. Often overplaying them is distracting and makes the audience more aware of both the picture editing and the mix, which is the last thing you want to do. IMO the mechanics should be invisible. If you think about the dramatic intent of the scenes, that is the best guide to when you need to split for perspective. Obviously you need to split for matching mic angles and futzes, but perspectives are a different issue.

[JoeMilner]

Well said, Gary

[danijel]

the way it is taught in my school is that for every actor you need to have 3 channels (close, mid, far) of production dialogue, and 3 for ADR also. this always pissed me off, because it's the 'old school' leftover, from the time there was no automation. you'd make a setting for each channel (EQ and reverb send), and preferably not touch that for the duration of the scene.

as for outside perspective, look for a lengthy thread not so long ago. it contains every possible method of getting the distance. try searching 'outside', 'distance' or 'exterior'.....

[charles maynes]

I appreciate Gary's comments too, but I am going to be the devil's advocate from editorial-

We editors have NO idea sometimes what the director is going to want to do- especially on those sorts of perspective changes-

If we move from a tight two shot, to a more distant wide shot, the sound elements might require different panning or even different sound altogether-

A good case in point would be a Seashore or boardwalk type scenario- you will need the sound of the water, but depending on the camera angles It could be very much more difficult to mix without the split. I suppose it comes to discretion really- but to say that one way or the other is "Better" denies the unique quality of the program we are working to make real with sound...

[starcrash13]

Originally Posted by meathman

hi,
This made me think about something ive never really considered.. Do dia editors usually always split for camera perspectives? also, how do the mixers handle this kind of situation....

For a film dialogue premix, editors typically split out every shot of a scene. This would cover different perspectives because each camera angle is a different shot.

Originally Posted by ggegan

I personally hate it when editors split FX and BG tracks for every perspective because it really destroys my flow and causes me more problems than it solves. It also has a tendency to multiply the number of tracks and get confusing. I want to spend as much time looking at the screen and as little as possible looking at cue sheets or computer monitors trying to figure out what's going on with the tracks. If I really need a split when there isn't one it's isn't that much trouble to just do it myself.

Foley and FX is a little bit of a different story and can sometimes require a little more thought and discretion. Sure, splitting for every perspective is overkill, but most FX mixers I've worked with expect splits when necessary. Whether or not they process the split tracks any differently is up to them, of course. Sometimes the split even acts as a bit of a "red flag" for the mixer so he/ she gets a heads up that something needs to be panned or reverbed differently.

Whether to split or not to split is difficult to definitively describe in a forum post and requires some experience and a bit of logic. The most basic reasons are for EQ matching different shots/ mic setups (in the case of dialogue), major perspective changes, and panning (for FX and Foley).

When prepping tracks for the predub, it's important to try and "think like a mixer". I mostly cut Foley and FX and I'll split for major perspective changes if I think that the mixer might want to process things differently, like a wide establishing shot in an echoey alleyway that cuts to a close-up. I generally split for panning less frequently when there is dialogue happening. If a character is talking, I almost never split their Foley. It sounds weird when the dialogue is anchored in the center and the Foley is panning across the screen. These are just a few criteria off the top of my head.

[ggegan]

Originally Posted by charles maynes

I appreciate Gary's comments too, but I am going to be the devil's advocate from editorial-

We editors have NO idea sometimes what the director is going to want to do- especially on those sorts of perspective changes-

If we move from a tight two shot, to a more distant wide shot, the sound elements might require different panning or even different sound altogether-

A good case in point would be a Seashore or boardwalk type scenario- you will need the sound of the water, but depending on the camera angles It could be very much more difficult to mix without the split. I suppose it comes to discretion really- but to say that one way or the other is "Better" denies the unique quality of the program we are working to make real with sound...

I'm not saying one way or another is better. I'm saying that different mixers have different sensibilities. I know from talking to my colleagues that there are many who do not share my sensibilities, but that is exactly of my point. Mixers in general do not intermingle with each other as much as editors or other crafts, so there is not always a lot of common ground between them and they can become somewhat idiosyncratic in their work habits. If they tend to work with certain directors, editors and supervising sound editors more often than others, which is very common, then they adopt ways of approaching a mix that are different from mixers who work primarily with other clients (and these days I consider sound editors to be clients rather than support staff, as was the case when I first started out). And, from my experience, editors also can be very individual in their approaches so there are a lot of issues to be considered. My comment is basically that communication can iron out a lot of those issues before they surface in the mix when the client is present and the stakes are high. And the communication goes both ways. I am just as willing to adapt to the needs of the editor as to ask the editor to adapt to me. The point is that I don't want to have to find out what's going on in the middle of a mix with the director, producer and picture editor watching us try to work things out.

My experience has been that, on any given project, the supervising sound editor has been dealing with the director, picture editor and producers a lot more intimately than I have and so I rely on them to inform me about not just the aesthetics of the film, but also the politics of the project, which are essential to understand in order to achieve a successful mix. If the supervising sound editor is out of the loop about expectations then that's not a good situation for any of us to be in.

Believe me, I realize that in the end we all have to work it out on the day. I'm just saying that the more advance coordination between the supervising sound editor and the mixers, the better. It is impossible to anticipate every eventuality, but at least we can come up with a plan ahead of time to try to give us the best shot at being able to gracefully adapt. The goal is to support each other and to both be seen as heros at the end of the mix.

[ggegan]

Originally Posted by starcrash13

For a film dialogue premix, editors typically split out every shot of a scene. This would cover different perspectives because each camera angle is a different shot.


Foley and FX is a little bit of a different story and can sometimes require a little more thought and discretion. Sure, splitting for every perspective is overkill, but most FX mixers I've worked with expect splits when necessary. Whether or not they process the split tracks any differently is up to them, of course. Sometimes the split even acts as a bit of a "red flag" for the mixer so he/ she gets a heads up that something needs to be panned or reverbed differently.

Whether to split or not to split is difficult to definitively describe in a forum post and requires some experience and a bit of logic. The most basic reasons are for EQ matching different shots/ mic setups (in the case of dialogue), major perspective changes, and panning (for FX and Foley).

When prepping tracks for the predub, it's important to try and "think like a mixer". I mostly cut Foley and FX and I'll split for major perspective changes if I think that the mixer might want to process things differently, like a wide establishing shot in an echoey alleyway that cuts to a close-up. I generally split for panning less frequently when there is dialogue happening. If a character is talking, I almost never split their Foley. It sounds weird when the dialogue is anchored in the center and the Foley is panning across the screen. These are just a few criteria off the top of my head.

I agree with what you're saying. I would just add that the type of console the mixer is working on changes the equation. If I'm working on a Neve DFC or a Harrison MPC, then I need splits more often than when working on an ICON. The workflow and available tools are very different if the mixer uses the consoles to their maximum potential.

[starcrash13]

Originally Posted by ggegan

I would just add that the type of console the mixer is working on changes the equation. If I'm working on a Neve DFC or a Harrison MPC, then I need splits more often than when working on an ICON. The workflow and available tools are very different if the mixer uses the consoles to their maximum potential.

Agreed. I'm mostly working with with mixers on traditional consoles, so my perspective may be a bit different.

Gary - for ITB mixes, since you prefer to not split things, are you doing quick snapshot automation changes on the same track? Even for plugins? Do you ever have problems with wiggy plugin automation not changing fast enough?

p.s. Great work on Being John Malkovich! That's one of my fave tracks! It's great to have experienced mixers like yourself here on the GS forum. Kudos!

[ggegan]

Originally Posted by starcrash13

Agreed. I'm mostly working with with mixers on traditional consoles, so my perspective may be a bit different.

Gary - for ITB mixes, since you prefer to not split things, are you doing quick snapshot automation changes on the same track? Even for plugins? Do you ever have problems with wiggy plugin automation not changing fast enough?

p.s. Great work on Being John Malkovich! That's one of my fave tracks! It's great to have experienced mixers like yourself here on the GS forum. Kudos!

Thanks for the compliment, it was a fun show to mix.

Just to be clear, I'm not opposed to splitting tracks, I just think that sometimes there is a tendency to overdo it. It probably comes from the frame of mind that it's better to split just in case because the mixer can always match the original tracks if he/she doesn't want a change, but that eats up time and energy, especially if the editor "helps out" by automating a level or panning change in the split track.

Obviously, doing things like splitting BGs for int and ext are necessary, but sometimes I get splits on every camera angle change as though I'm supposed to re-pan everything for every shot. I've even gotten splits during CU car bys because there is a quick picture cut to a reverse angle POV as it passes. The editor thought I might want to put the car in the opposite speaker for a fraction of a second and then pop it back to the other speaker when the POV shot was over because that is logically where the sound would be coming from. IMO, that's over-thinking the geography of the scene. That's why it's a good idea to talk to the mixer first, especially if you don't work together often. You might be able to save both yourself and the mixer some unnecessary work.

As to your question, I often go into preview mode, highlight the area of the regions I want to change, loop it, adjust the settings and then do a "write to all enabled" command. If the change in automation is too radical and abrupt it can sometimes cause digital snaps in the monitors, especially if there is a lot of low end content, in which case I'll ramp the changes in over a few frames or use a "glide to all enabled" command over a few frames. I like round edges anyway, so this works well for me. Sometimes it's faster to just split the regions to another track myself if the editor didn't do it for me and use 4 or so frame crossfades, especially if I have another track that already has the automation settings I'm looking for.

The only time plug in automation latency is a problem is if I apply heavy EQ or some other treatment to a region and neglect to lead the start of the region by a frame or so. The latency causes the EQ setting to kick in a few milliseconds after the start of the sound (assuming it starts at the very beginning of the region) which can result in a nasty glitch. Since I generally try to do quick ramps in and out of changes, this isn't a problem with "virtual" splits. I don't think most people can detect a sync problem from plugin latency anyway, especially considering that when one is sitting 30' to 40' from the screen the time it takes the sound to reach your ears puts everything out of sync by about one film frame (30 to 40 msecs), and nobody ever seems to notice it. In fact when I mention it to clients they usually think I'm pulling their leg. If you are sitting more than halfway towards the back of a large commercial theater, the sound is going to be 2 or more frames out of sync, so a few msecs extra isn't going to matter. Even with TV I'm often seeing really bad sync problems with digital broadcasts, so I'm not going to lose sleep over latency caused sync inaccuracies of a few msecs.

[Lake]

Originally Posted by ggegan

Thanks for the compliment, it was a fun show to mix.

Obviously, doing things like splitting BGs for int and ext are necessary, but sometimes I get splits on every camera angle change as though I'm supposed to re-pan everything for every shot. I've even gotten splits during CU car bys because there is a quick picture cut to a reverse angle POV as it passes. The editor thought I might want to put the car in the opposite speaker for a fraction of a second and then pop it back to the other speaker when the POV shot was over because that is logically where the sound would be coming from. IMO, that's over-thinking the geography of the scene. That's why it's a good idea to talk to the mixer first, especially if you don't work together often. You might be able to save both yourself and the mixer some unnecessary work.

I think that you hit on a good example of one of the roles of a Supervising Sound Editor/Designer during tracklay. "Quick cutting to match changes of shot" or "Taking a smoother approach" are creative concepts that should have been worked out before the mix and during the tracklay. As is often the case in our large mix theatres, they turn very quickly into the world's most expensive edit rooms if things like this are not already ironed out.

Regarding 'dialogue matching picture perspective', IMO it really comes down to what serves the story and the movie best. Dialogue mixing is a surgical procedure and usually benefits from a subtle, steady hand. Maybe why I mainly mix FX! I also frequently find that panned dialogue in movies is a distraction rather than an effective treatment, although there are exceptions (the opening scenes of "The Diving Bell and the Butterfly" and the finale of "The Orphanage" are a couple of recent uses that I thought worked).

[danijel]

Originally Posted by starcrash13

p.s. Great work on Being John Malkovich! That's one of my fave tracks! It's great to have experienced mixers like yourself here on the GS forum. Kudos!

this place just gets better and better!

lake, are you gilbert lake of "bridge to terabithia"? one of my favorite movies in the last couple of years! great sound! kudos to you and tim prebble (who is already on the forum), and welcome!

[ggegan]

Originally Posted by Lake

Dialogue mixing is a surgical procedure and usually benefits from a subtle, steady hand. Maybe why I mainly mix FX!

Yeah, I think of mixing dialogue as reconstructive surgery, or just damage control. But then, the guy in the dia chair gets to also mix the music, and generally speaking gaff the mix.

I love mixing fx though. To me it often feels like composing music, even more so than rerecording music does. You start off with a clean slate and create something new from fresh and (hopefully) pristine elements. I never have to ask myself if I can save a track or will I have to make due with a poor quality ADR performance.

[soundslikejoe]

I'd like to hear more about the "DRY" aspects of the original post.... especially the opinion of the experienced members on usage of reverb for narrator VO.

I did a doc last year that's in several festivals. I went to a screening and was totally disappointed in how the reverb settings translated to the bigger room. It sounded much better on smaller systems and home systems.

I didn't even want to use it but the director requested it based on the advice of others (old teacher).

What's wrong with a dry narration track?

[danijel]

Originally Posted by joenovice

I'd like to hear more about the "DRY" aspects of the original post.... especially the opinion of the experienced members on usage of reverb for narrator VO.

I did a doc last year that's in several festivals. I went to a screening and was totally disappointed in how the reverb settings translated to the bigger room. It sounded much better on smaller systems and home systems.

I didn't even want to use it but the director requested it based on the advice of others (old teacher).

What's wrong with a dry narration track?

joe, this is a topic about which i heard a lot of differing opinions (and is therefore interesting). would you consider editing your post a little and making a new thread out of it?

[soundslikejoe]

Usage of Reverb for Dialog (narrator)

new thread

[Lipflap]

Rules of thumb are like any other thumbs: they tend to get broken. Still, if you keep a few priorities in mind, it's not too hard to keep perspective on perspective (I couldn't help myself!).

Remember that you're going to split off all production sound clips (regions, segments, whatever you want to call them) by shot (which happens to be camera - or mic- angle). This initial split is intended for better control throughout the editing and mixing process and its main purpose is to put like things on the same track, within a scene. This way you achieve a smoother, more controllable scene.

Since you've already split by shot, certain "natural" perspectives take care of themselves. If, for example, picture and sound cut from a 2-shot to a LS, you automatically have a "freebie" perspective cut, since you've previously placed the two shots on separate tracks. But this is not the kind of perspective cutting you're talking about.

There's already a consensus that nothing beats talking with the supervising sound editor or mixer as you plan perspective cuts. Most uses of perspective are logical and based on “standard” film language, however, there are always grey areas or places where the supervisor or the director wants to do something you'd never have predicted. So ask...

Having said that, there are a few pretty safe "rules":

When the sound continues from one perspective but the picture changes to a markedly different view, split. (If your CU sound continues over the "fake" LS), it's a split. The mixer is not forced to make something of your split, but if it's not there, you've hogtied him/her.

Normally, a conversation between two people maintains the same perspective, unless the audience's point of view changes. Two people talking usually means some form of connection, so whether we see a CU or a MS or a 2-shot, you'll likely maintain the same sound. Doing otherwise calls attention to the cuts, which is usually not want you want in the middle of a conversation.

When the above conversation is heard/seen from another person's POV, always split.

If you want to impose a bit of psychological or emotional distance between two characters in a conversation, you may try splitting the off-camera pieces of dialogue. You can use this split to create very minor changes in level and EQ, which the audience may perceive as a lack of connection.

Phone split. Everyone hates it but it still exists. That’s usually four points-of-view.

Just because we cut to a WS doesn't mean that the dialogue has to become distant and muffled, but it's irresponsible not to prepare the split. Only the gods know how the wide shot will be handled at the end of the day, so be prepared.

Within a film, it's useful (but not necessarily mandatory) to be consistent about "why" you're splitting. You don't want to be a slave to your own rules, but a bit of a system makes it appear that you had a plan and weren't just reacting to each situation in a vacuum. It's very useful to be consistent as to "how" you organize your splits (do you put the split track on a higher or lower track?, etc.) Ask the boss about these things.

Finally, it's usually better to make the scene a smooth as possible BEFORE you begin making it "bumpy" through perspective cuts. I don't mean that you should do any processing to the tracks before attending to perspective, but rather that you first take care of the normal smoothing issues (remove transient noises, fix articulations, ease shot transitions, etc.). THEN you are better equipped to make perspective cuts that don't stick out too much.

[Sean L]

Great thread learning lots thanks lads thumbsup

[Lake]

Originally Posted by danijel

this place just gets better and better!

lake, are you gilbert lake of "bridge to terabithia"? one of my favorite movies in the last couple of years! great sound! kudos to you and tim prebble (who is already on the forum), and welcome!

Thanks! Working with Tim at the moment...or trying to do some work, in between games of fussball.