Dialoge Mix Guidelines

Let's stop and consider for a moment.

Just how low budget is this film? Pocket change? A few grand? A million dollars? If it is a million listen to Scott! I was kind of thinking it was pocket change. I was also thinking "steam punk".

A lab in a trailer would be kind of echoish. The guy in the tranporter, would I think sound right if he sounded like we heard him through an old am radio, assuming an intercom. I agree with Scott there should be some kind of "something unexpected is about to happen" warning in the sound. A big generator puttering outside in the background that gets loud when the door opens would be kind of nifty (Figuring that they are using a trailer come lab out in the middle of nowhere. Why else use a trailer?)

Getting into director's territory: a long shot showing that the guy is hauling off the trailer/lab would be cool.

See, we can come up with lots o'things to change. Are they appropriate? Only you, and the director, know.
 
To respond to a number of posts.

The typing etc may be too loud. But I actually think that it's not too loud as an absolute thing but that it is consistently too loud. As in it's volume doesn't reflect the way people listen to things. So it might start at that level and then drift down.

The editor/ designer's job is to do what the brain would do IF the audience were actually there. So important things stand out, dangerous things stand out, any thing new or unexpected stand out AND anything constant, non-dangerous, expected, etc. are lower.

BG's for instance often start out louder than they end up. That great cricket sound wants to be noticeable as you step outside but it also wants to drift down like your brain would filter it out in real life.

Watching a film your brain doesn't do the filtering because it knows you are watching a film, it doesn't have the environmental clues to help and you are not in danger. But if you act like the brain and do the filtering the audience will have a much more "real" experience.
---

Forget going to 12, you are capped at 10. A classic mistake is making everything loud and then you hit the big climax and try to push it to "12" and it just has no impact. One reason is that loudness is exponential so to have a big bump it's not 12 but more like 20 you would need to aim for and that would let you recreate that scene in Back to the Future with speaker cones flying at your audience.

So for an opening sequence I would STRONGLY recommend loosing the music in the middle because it gives you someplace to go rolling into the opening credits.

Muffled voice. My problem wasn't so much the muffledness but that it didn't sound like his space looked. It looks like he is in some small metal chamber, all hard. He would be muffled going through the glass but the outside room is also lots of hard surfaces and the muffle sounds "soft" not hard. It might work as it is if there is enough equipment noise etc. in the room since that will eat up some of the "softness". But greywolf had a great idea with the raspy squalk box. You can probably get something useful out of a free plugin from iZotope called VINYL . It's a good solution because it fits well into the "world" of the visuals, it just makes sense that there would be some form of talk back AND it solves the muffle issue.

I would argue that budget is irrelevant to sound track quality. Nobody watches something and says it's not good but they didn't have a budget so it's great.

Budget can be a factor in how close to "great" you can get but it doesn't change OK or bad to great just because you didn't have the budget.

I ma right now working on a short from a director I know that has zero budget and I just turned down one from a different director I know that I am sure looks fantastic and did have a budget. The first really wants to have a great sound track and the second wants a short turnaround (but still great sound...?). I have worked on everything from very well paid BIG budget features to ads to no-budget shorts. I don't often do work that won't pay the rent but I won't do work I would not be willing to have my name on, unless you have a lot of $$$$ to shower on me. I mean there are some things I wouldn't do, but I have paid the rend with some junk over the years. And of course after the fact sometimes you realize you were wrong and you wonder why you thought it was going to be a great project. But going in if a director doesn't want to take the time or isn't interested in doing something really professional with the sound track they better have a nice budget for sound, because I'm not doing it unless it pays well enough to let me give a break to someone doing something interesting that does want a really good track.

Where you can "cheat" on a low budget show is with things like using PFX if they work and knowing where the music is going to be big and you will only need a few poke through SFX. A small crew, as in you on your own, can work far more efficiently than a big budget crew. You know everything and the film is (or should be) locked so you are not dealing with conforming. The actors are usually local so you can do some ADR with out flying off to London or? Also you can do the ADR in an environment that is similar to the scene so matching can be much easier. You don't have completion bonds so time is cheap. One requirement I put on no or very low budget shows is that I will take paying work if it comes in and they have to be willing to allow that extra time.
 
LOL, the RV we bought for $500 and it only ran for about 8 miles and died. We had to pay another $275 to tow it to the location, so moving it was not much of an option, but yes a "drive away" would have been better. The whole budget was about $40K, and we shot it in 12 days, 12 long, very long days. Now I'm not making excuses here, that's the first thing I learned, was the viewer doesn't care about MY problems, he/she should only care about the hero's problems, but we did have a lot of limitations.

I have reworked the sound today and have another pass. There are some more things I want to do, but I think this is definitely getting better...
https://vimeo.com/user8773489/review/192952339/b70ed364e0

I very much like Scott's explanation of how the brain filters sound. Again, I am not looking for numbers or plug-in specifics, but rather an approach to the larger issues. In this pass I used the "beep" sounds to move us back and forth from the close up of the guy in the transporter to the control room. It surprised me how effective it is.
 
Better. Actually I think a lot better. One thing I am still really missing is something for the lights on the machine. They are going crazy then the slow then he pulls the lever they are going crazy and suddenly shift to a different mode. This seems significant like something just happened and???? It's not a completely true "rule" but generally if the audience sees something change they expect to hear something change. So here relay clicks for the lights would drive us all crazy, plus they have a bit of continuity error so... But there might be something like the machine "building energy" and then have that "release" when the lights switch after our victim gets sent wherever.

Personally I would still like to see it with no music in the center, but this mix is a lot better.

Something I didn't mention in the last post RE: loudness. The way to make something sound loud is to get quiet just before the loud sound. So in this case if the hit is a key event (seems like it) then if you do the energy build up and then release everything will quiet down a bit so our new "normal" level is quieter. Then swoosh/ whack and we are back in the action.

But already it feels like it's all in the same world and not SFX slapped on top of a production track.

I have a theory, kind of backed up by some observations Walter Murch has made. My theory is that you have to get enough layers (that fit the world) so that the brain can't actually pay attention to all of them and has to do a little work. At that point everything suddenly feels much more "real". Kind of like in music one is one, two is two, three is a bunch. You hit three notes and you don't have three notes you have a chord. The other "rule" is that humans can't really focus on sound and visuals at the same time. For instance if an editor has a jump cut they just can't get to work, if you slap a sharp sound ON the cut nobody will notice the jump. And you just made a friend! Door closes are the classic but any sharp surprising sound will make everyone focus on sound for a fraction of a second and if that happens on the visual glitch nobody will "see" it. Sounds silly but it's amazing how well it works.
 
It is clear to me now that due to my background in commercial work I was working too hard in my recording technique to get what we, in that world, consider to be good. i.e.. neutral, audio. In commercial. we are constantly fighting noisy environments and searching for neutral (close mic'd) flat sound. In narrative the goal appears to be to sound "real" or "of that place", and a bit of room acoustics, background noise, and a little of what my crew would consider "distractions" to keep it real are desirable. I have learned a lot here and I thank you all for the guidance and your patience. Fortunately I have tracks with boom and lav on almost every take, so I can mix them, or even just use the more acoustic track to inform my flat track if I need to. Thank you for your time in leading me through this, and will continue to post some scenes here on occasion for feedback and advice. This is a great forum.
 
Ric, I think that background noise (wild Foley?) would be recorded on a separate track, so you can mix it in as needed.

As I understand it, everything goes on separate tracks. Watching really well done movies some of the sound track is almost subliminal. There but just barely. Faded up when you want to bring it to the audiences attention. What I am saying here, is that you "do" want to record your dialog as pristine as you can.
 
graywolf said:
Ric, I think that background noise (wild Foley?) would be recorded on a separate track, so you can mix it in as needed.

What I am saying here, is that you "do" want to record your dialog as pristine as you can.
Click to expand...

I am not suggesting that dialogue not be pristine, rather, that it need not be recorded as "dead" as I was doing with close mic'd lav's. A bit of the "sound of the room" is desirable in narrative, whereas we tend to go flat, and studio-like in commercial work.
 
ric,

As a rule I think the way you recorded is the way to go. On this particular scene since the set is actually what it is in the film you probably could be less up close and personal with the recording? Boom is almost always the preferred track for film work. The big BUT though is that it is rare for a film set to actually be what it is in the world of the film. Unless your film takes place on a low budget filmset the "set" probably doesn't sound like the "location" in the world of the film.

Here is where there is kind of a split with low VS big budget films. The big film is going to work hard for that very clean and neutral sound in the edit because everything else is going to be added and it makes ADR easier to fit. All of those room characteristics on the dialog will be added in post. It gives you flexibility and fits the workflow better. On a low budget you are probably inherently going to pick up more of the space because you don't have the $$$ and people to get rid of it all. It's a mixed blessing because it's harder to match ADR and often you end up with background jumps on cuts etc. But IF it works for the location in the film world you can save a bunch of dialog processing. On the other hand if you do need to do processing on the dialog it's going to affect the BG and anything else you can't move off the dialog track. Very mixed bag, so If you can get clean neutral dialog it's probably good to do so.

greywolf,

There is something that (at least in northern California post circles) is sometimes called wild Foley (WIFO), but it is sounds that would normally be done in Foley but instead were recorded in the field as FX. WIFO is still FX and not Foley nut it's a way to identify sounds that you can use and take off the Foley list since Foley is often not allowed much time even on a big film.

Foley , wild or not, would never be "background noise". Though you may have been referring to the hard FX like the on screen typing? Typing is something that would normally be done in Foley, but could also, especially in a low budget film, be cut in as an FX. And if you recorded a bunch of typing with the idea of cutting it to visible typing (as opposed to off screen typing) then it would be WIFO.
 
I will get and read both. After 30 years of producing commercial work, making a very good living, and learning as I went along, I now have the time to only do what I want to do and do it right. Thanks to this film project and the help of the people on this forum I am thrilled and excited to be doing something (not entirely) new but delightfully creative. Again I thank you all, and stay tuned, I will post more problems and successes as I progress!
 
Noiz2 said:
ric,

As a rule I think the way you recorded is the way to go. .
Click to expand...

Good to know. The DP was often grouchy with me as I (re)mic'd the actors when it just did't sound right and I was cross with one actor that kept ripping the wires out of my Countryman B6's (4 mics in 12 days!), mainly because he did his scenes completely different every take and drove us ALL crazy. As the "new guy" I assumed I was being the pain in the ass. As I read more I realize the sound guy is almost ALWAYS a PITA, mainly because he's the last guy in the chain. When they run overtime on the lighting and blocking it always seems to be sound that get's pressured to make up the difference, but that's a different thread!
 
ric8138 said:
I will get and read both. After 30 years of producing commercial work, making a very good living, and learning as I went along, I now have the time to only do what I want to do and do it right. Thanks to this film project and the help of the people on this forum I am thrilled and excited to be doing something (not entirely) new but delightfully creative. Again I thank you all, and stay tuned, I will post more problems and successes as I progress!
Click to expand...

Cool. I hear you about learning as we go along. And so many of the newer books aimed at our world are fairly shallow. But Purcell's and Rose's books are a step above. And while some of what they say will be old hat to you, perhaps they'll fill in a few gaps for you (as they did for me) and give a good sense of workflow for longer projects.

Looking forward to reading and hearing more about your project and progress.
 
I am reviving this thread after reading the "Editing Diqalogue" book. Thanks for a great referral. I now have a good concept of work flow and what I should hope to achieve in my first pass at dialogue when I get back from winter holidays in January. Fortunately it would appear that I have the raw material in place to do a proper job of it. On a side note, I watched Resevoir Dogs last night on a big screen with a good sound system, here where we are staying, and was shocked at how bad the dialogue recording was in all the warehouse scenes. There were an awful lot of wide shots that sounded like a shotgun with outrageous slap back from the hard walls. At least 20% of the dialogue was completely unintelligible. I am listening much more carefully to my movies these days.
 
Jim Feeley said:
Looking forward to reading and hearing more about your project and progress.
Click to expand...

OK, I'm going out on a limb and posting my first pass on 7 minutes of the feature mentioned earlier in this thread. You can see it here:
https://vimeo.com/user8773489/review/201529691/b1eb68446b
PW is "TWT2016"

This is not finished, I still have to do SFX in the "lab" scenes (beeps and boops) etc, but this pass is to get the dialogue in shape for some test viewings. I would appreciate comments and suggestions.
 
I will do a more specific "hit list" but a few general notes. Most of the interior production sound is overly bassy. Very "in your face" proximity effect bassy. It kind of dulls it down. This was very noticeable in the house scenes.

You are either using volume graphing or gating to reduce the "air" when folks are not talking and the "breathing" of the BG is pretty noticeable. A better way is to use noise reduction to knock down some of the air, but really what you need to do is to add some Ambience so that it evens out. The "world" is largely missing. This may be on your list but you only mentioned hard FX for the lab. The house should also have backgrounds.

I know the logic with the car/ park cuts but you have to fudge the BG's so the switch is not so jarring. You probably can't cut back on her BG's so you need to add some to the car. Even though it makes no logical sense what is often done is to basically use the louder BG at a lower level under the quieter shots. It's more of a case of doing what the audience needs than what is "real".

But in a way it is the more "real" choice. Where is the audience? Are we jumping back and forth or are we listening to both at the same time? The later is almost always the case. WE the audience are not in the park OR in the car we are in both places the whole time so it "logically" makes sense that we can hear both all the time. Try it and I think you will find the cuts are much smoother and the whole thing will flow better.
 
Noiz2 said:
You are either using volume graphing or gating to reduce the "air" when folks are not talking and the "breathing" of the BG is pretty noticeable. A better way is to use noise reduction to knock down some of the air, but really what you need to do is to add some Ambience so that it evens out. The "world" is largely missing.
Click to expand...

Thank you for your input.

Actually I did a separate track of clean room tone, as described in the "Editing Dialogue" book, so my "world" is a distinct track that has been ducked during dialogue. I can just bring that track up to more closely match the voices. I am still trying to find the right balance there. In the phone scene there was quite a bit of "city roar" at that outdoor location, and I recorded clean ambient of that as well that I used at a lower level behind the actress. Think what I'm getting from you is that the ambience needs to come up in the mix some and I need to EQ some bass out of the voices in the house scene?
 
Last edited:
ric8138 said:
Thank you for your input.

Actually I did a separate track of clean room tone, as described in the "Editing Dialogue" book, so my "world" is a distinct track that has been ducked during dialogue. I can just bring that track up to more closely match the voices. I am still trying to find the right balance there. In the phone scene there was quite a bit of "city roar" at that outdoor location, and I recorded clean ambient of that as well that I used at a lower level behind the actress. Think what I'm getting from you is that the ambience needs to come up in the mix some and I need to EQ some bass out of the voices in the house scene?
Click to expand...

When I say "world" I'm not talking about Room Tone. I'm talking about the World of the story. What does that house want to sound like? Do they have a refrigerator? Is it windy? how about the boiler, cars going by out side, dogs barking, birds chirping, garbage trucks, low riders? I don't know because I don't know the story and haven't talked with the director but one of the big jobs with FX is to create up the "places" the scenes take place in. Everyone tunes out this stuff in their own lives but take a recorder and just set it up in a room of your home for a few hours and then play it back and list all the different sounds you can hear and I think you will be surprised. All of that is what I mean by noise.

What Purcell is talking about is Room Tone. And maybe he wasn't clear enough but the purpose is to make a seamless dialog track. Ideally you should not hear any difference between when they are talking and when they are not, in the dialog track. RT is part of Dialog, where the "world" is part of FX.

The phone scene:, the short version is that when you are cutting back and forth like that the "audience" is not in either location but in a special fourth wall location and so they get to hear both sides. So the ambience really wants to be much closer to being the same than being two wildly different locations.

Sometimes the point IS that they are in two wildly different locations and you would keep it like you have it with abrupt changes but this is not one of those times. The story here doesn't have to do with where they are as much as what they are saying.

So IMHO you want to balance the two ambiences so it's a subtle switch and not a big switch. You can do that by lowering her Ambience or by rising his or by mixing them a bit so neither goes all the way away. The last one is what I would probably try first.

BTW, IF you were going to go with sudden changes in the Ambience then you want to do it on the cut with only enough fade to keep it from popping. You are kind of fading it back and forth and it makes it even more distracting. I understand the logic it "smoothing" the transition but actually it makes them stick out more since it doesn't match the visual.

As to EQ. To me it feels like there is a lot of low mid in the mix. I listened on headphones that probably made that stand out more but it still feels a bit over full in the low mid range. This is especially true on more distant shots in the house where it sounds too close for the visual.

Lastly I would add that you asked for suggestions and especially over the web with out knowing the person it can come off as overly negative. For a first pass this is good work. The EQ stuff is a common one and may be your speakers or even personal taste? I would try scooping out some lower mid and see if it doesn't sound a lot cleaner.

The concept of the "world" of the story is not talked about much anywhere. Film schools pretty much never talk about sound and most books etc are on the mechanics. I happened to get tossed under a very creative bus and had the concept drilled into me by folks who either were or were trained by some of the best. After years of getting grilled about choices it kind of becomes second nature but it is something I don't think a lot of editors spend much time on. If you take a close listen and break down some scenes from masters like Alan Splet you will see what I mean. At first shot many see things like the sound design of Blue Velvet as a bunch of weird sounds but if you think about a bit more everything is there for a reason and they all make sense in the "world" of that film and the "world" stays true to itself always.
 
Noiz2 said:
When I say "world" I'm not talking about Room Tone. I'm talking about the World of the story. What does that house want to sound like? Do they have a refrigerator? Is it windy? how about the boiler, cars going by out side, dogs barking, birds chirping, garbage trucks, low riders? .
Click to expand...

OK got you, and I didn't see any of your notes as negative. As far as "The World", in this pass I am mainly dealing with the dialogue edit, picking the tracks ( I did boom & lavs on iso tracks) breaking out the PSX onto another track, some of which will stay and some of which will be replaced with Foley later on. This is all within Premiere at this point, so final balancing and EQ will be in Audition, later on. I did want a very intimate, in your face, menacing sound to the confrontation between the 2 Todds at the table, but I agree that there may not be enough difference between that stuff and the more distant shots. All good points, Scott, and I take your comments seriously and I'm thankful for your time to look listen and share. It's very valuable.
 
Story World Sounds? Suspension of disbelief sound? Sounds to pull them into the story instead of thinking "This is all BS"?

Hummm.....? Something to think about here.
 
You must log in or register to reply here.
Back
Top