What about the "Film Sound"?
 

There's so much (read TOO much) hype going on about achieving a film look using DV. But it seems like there's not that much concern about mimicing the film SOUND.

If you take a set of headphones and listen only to the audio tracks of a commercial film and a DV "film" side by side, you can usually pick out which one is the film and which one isn't. You know they say that 70% of your moviegoing experience is auditory. I've myself have always struggled to get that perfect sound.

So I'm wondering if anyone has any tips on getting a DV film to sound more like a real film? Specifically;

1. How to best "cut a hole" for dialog through a particularly intense piece of music and sound effects. In films, even when all hell is breaking loose and the 150 piece orchestra is booming, you can still hear the dialog pretty clearly. I've tried various forms of compression, riding gain, etc., with little luck.

2. How to get a good "punch" from an explosion on the subwoofers. Is there a good source of sub data for things like this or is it just a matter of taking the explosion sfx and boosting the sub range? To me, that always just seems to muddy up the sound.

Thanks

Re: What about the "Film Sound"?
 

<<<-- Originally posted by Rob Henegar : There's so much (read TOO much) hype going on about achieving a film look using DV. But it seems like there's not that much concern about mimicing the film SOUND.

If you take a set of headphones and listen only to the audio tracks of a commercial film and a DV "film" side by side, you can usually pick out which one is the film and which one isn't. You know they say that 70% of your moviegoing experience is auditory. I've myself have always struggled to get that perfect sound.

So I'm wondering if anyone has any tips on getting a DV film to sound more like a real film? Specifically;

1. How to best "cut a hole" for dialog through a particularly intense piece of music and sound effects. In films, even when all hell is breaking loose and the 150 piece orchestra is booming, you can still hear the dialog pretty clearly. I've tried various forms of compression, riding gain, etc., with little luck.

------------------------------
One way is to pan the background sounds to the left and right. Another is to use an equalizer to leave a frequency hole right in the middle of the speach frequencies. Douglas discussed this during his Acid seminar and Jay Rose discusses this in his excellent book on sound for DV.

Both or a combination of these techniques work very well.
-------------------------------

2. How to get a good "punch" from an explosion on the subwoofers. Is there a good source of sub data for things like this or is it just a matter of taking the explosion sfx and boosting the sub range? To me, that always just seems to muddy up the sound.
----------
Dunno
---------

Thanks -->>>

I guess it would start with the best mic for the particular job. Cutting through the noise and isolating the dialog. Read Jay Rose's article on mic patterns (click here)

Also read what Schoeps say about the rejection of off axis soundclick here

It has to start with clear dialog, sound effects or ambience and that's only goung to happen with a a good mic that was engineered for that particular situation.

For an example of axis resonse on the variois mics carefully listen to some of the clip here. Use good headphones through a decent sound card. Listen to the ambience vs the dialog. A mic like the Schoeps MK41 or Sennheiser MKH416 will cut right through the noise .

I assume mastering has a lot to do with it as well just like any commercial music album is mastered after final mixdown.

i am not sure but you had to remember that most big movies have the sound on the film (optical or magnetical sound ).
It means that probably the bandwith is very poor (under 8-10 kHz) when video with the digital 48Khz sampling bandwith (can)give perfectly clear sound. that is probably what make the difference. just try to cut over 10kHz and try again...

Thanks Bryan,

Of course having a thousand dollar plus mic would be great, but I don't have that kind of $$$.

I wasn't really talking about cutting through noise to get the dialog, actually. Our dialog is recorded clean, crisp and clear on a quiet set. What I was talking about is during the mix, when you're trying to add in music and sfx in a particularly intense scene.

Mike put you in the right direction. Learn to cut unwanted frequencies in the music, amiance or foley track when dialog is going on.

Also, when you mix, mix to your dialog (like all rules this is subjective and not cast in stone - it's more of a guidline really). What happens is we mix the audio, say a musical score and we mix it loud, becuase, hey it sounds better loud. But then we add dialog on top of it and there is only so much headroom to play with. Then you lower the music and it doesn't sound good - but it does. If you played the mix louder as a whole you'll start to hear how it really sounds. I don't know if I was too clear on that....

Just as important, learn to place sounds in your mix using paning, reverb, delay, eq, etc. If you can sit infront of your monitors (speakers) and hear where every instrument is in your orchestra then they are well placed. Often times, when writing orchestral scores in a sequencer we have to adjust all the sounds to get them to sit right in the mix. Even when using a single audio file is this important. You might for example duplicate the track process them differently and pan them left and right with the dialog dead center (where you've also made a frequency hole).

What program are you using to mix in? (or is it all hardware based).

EDIT: For creating more of a punch in your sub you need to process the sound, not just increase the level. Experiment with some envelope filters and compression. You probably want a bigger attack and lesser sustain and decay. Just remember, your level can't go past 0.0 (cumulative) otherwise you're toast.

Poor quality audio will always be poor quality audio regardless of how much you treat it.

You don't need $1500 mics just decent ones used with care.

Matthew pretty much hit it on the head..

another thing to note is that most of the dialogue is run the HArmonic/Vocal Exciters to thicken the talents sound. Its also run through various compressions depending on the situation.

as mentioned "cutting a hole" can work, with aps like vegas, its quite easy to create a flowing moving envelope of moving frequencies to allow teh subtle fluctations of the frequencies, another way to do it if you dont have vgas or pro tools is with Hardware. Basically run a unit like a Yamaha O1v Digital desk connected thru midi and into a software sequencer. The sequencer will control the fluctuatng frequencies via midi to the desk. this of course is based on a routed audio source from your pc, into the mixer, back out into the PC

with sub bass, Dolby recommend a cutoff frequency for the low pass filter to be no lower than 120hz, which is thw film "standard".

you will notice that your film sound is alot tighter, alot more controlled and very precise as opposed to your dv film, simply because its been processed and has had its various sound sources compressed and remastered/remixed.
Most of the time, its then encoded to Dolby Digital which adds its own compression algorythm to the mix.

commercial movie studios have sound stages, effect units and dedicated engineers who manage and process this. Theyre using equipment worth millions of dollars compared to what your average DV Filmaker has. It all adds up..

to look at it in a different way, its like comparing a dvx100 to a panavision monster.. sure youll get good imaging results, but its just not going to be the same..

Can you guys in the know clarify a bit on the frequency issue... So when mixing your dialog you should roll off everything below 120hz? And then kill everything above 10khz?

Normally I find success in pulling down some of the midrange frequencies in dialog - seems to give dialog a more dramatic feel. Is that what's meant by a frequency hole?

I must say I've attempted to read Jay Rose's book before, but I found it bored me to tears when all I was looking for were nuggets. I don't really want to be a sound designer - I just want to be able to get by on the best I can do within a reasonable amount of time with a reasonable budget. I've found sweetening in Vegas to be a lifesaver for myself. And if I want that hollywood sound, there are projects that have money for that in which case I hire someone much smarter than myself in audio concerns.

Re: Re: What about the "Film Sound"?
 

<<<-- Originally posted by Mike Rehmus : One way is to pan the background sounds to the left and right. -->>> -->>>

What does this statement actually mean? It sounds too simple and I don't understand how it can help isolate the dialog track.
Thanks
-Cjh

<<<-- Originally posted by Mike Rehmus : One way is to pan the background sounds to the left and right. -->>> -->>>

I agree, that sounds too simple. How does one get a sense of the left/right separation ... sounding as if there's a "hole" in the middle? I've tried a slight delay between the left and right tracks, and that sounds great, until you listen to it on a mono television. Then you can hear the delay quite easily.

Is there a better way to get that "dead space" between the hard left and hard right channels?

What Mike is saying is to learn how to make space in your final Mix. (I'M EDITING THIS BECAUSE I ORIGINALLY ANSWERED THE WRONG QUESTIONS BUT JUST NOW REALIZED WHAT YOU WERE ASKING).

Yes it's simple and no it won't "work wonders", but getting good audio isn't about doing "something that fixes it all". It's about making a lot of small subtle changes that work together to achieve the sounds you want. I'm just trying to get you to think differently with regards to sound.

Let's assume we have a stereo mix. You have an orchestral track playing under a dialog scene. Let's say in your dialog you are panning the dialog slightly to mach the position of your characters (very, very slightly). You can offset each track so that you can apply processing to them separately so that the opposite side is still full(er) and the side where most of the dialog is occurring (for that character) is thinner. Bring the volume down on one side slightly while keeping the other the same. You can eq them differently. Let's say you have an explosion or an orchestral hit. played full volume, it's too much. so you dupe your tracks. On one channel you lower the explosion or hit (almost like adjusting opacity) on the other channel you use an envelope filter to give a big attack but then lower the sustain and decay settings. Will this create movie magic? No. But it will give you a little more space in the mix, a slightly different sound. Producing audio is like producing video, you are creating an illusion, except for audio it's an illusion for the ears. We tend to think that the best audio is the most realistic, well that's no more true for audio than it is for video. Foley sounds are a great example of that.

I would recomend anyone here getting a book called "The Art of Mixing: A Visual Guide to Recording, Engineering and Production" by David Gibson. While there are many great books on the subject, and some more technical and better than this one, its visual approach is very unique and I think would communicate the concepts to visual thinkers better than most others. It shows you through graphical illustrations, how to create space in a mix and how to place sounds in different parts of the stereo field. Check it out.

Matt

One other thing. Film scoring is an amazingly complex art form. All of us who compose love to score to picture because it's fun and challenging, but few do it really, really well. But there are more film scoring students today than ever before (when it wasn't even taught in school, because it was considered the illegitimate stepchild of composition).

Just like others in this thread have reccomended good mics on the GIGO (garbage in, garbage out) principle, a good score works the same way. The composer will write and orchestrate in a way that naturally leaves space for dialog. Of course this means that the composer has to have a finished (or relatively finished) film to score to.

Just something to think about.

Matt

<<<-- Originally posted by Giroud Francois : i am not sure but you had to remember that most big movies have the sound on the film (optical or magnetical sound ).
It means that probably the bandwith is very poor (under 8-10 kHz) when video with the digital 48Khz sampling bandwith (can)give perfectly clear sound. that is probably what make the difference. just try to cut over 10kHz and try again... -->>>

I'm not certain that the big movies have on-film sound anymore. I know the magnetic sound is almost gone as a good friend of mine produced the magnetic heads used in most of the movie projections systems around the world. Even got a technical Oscar for it. Their company is gone now because the demand for heads fell down a cliff.

The on-film sound was pretty bad even at its best. The current sound systems with their Dolby or THX stamps of approval are way beyond the wildest dreams of on-film sound.

Re: Re: Re: What about the "Film Sound"?
 

<<<-- Originally posted by Cliff Hepburn : <<<-- Originally posted by Mike Rehmus : One way is to pan the background sounds to the left and right. -->>> -->>>

What does this statement actually mean? It sounds too simple and I don't understand how it can help isolate the dialog track.
Thanks
-Cjh -->>>

I guess you'll have to try it and listen to the results. Obviously you cannot try spatial tricks in mono but it does work in stereo. Douglas Spotted Eagle taught me that trick.

>>>I would recomend anyone here getting a book called "The Art of Mixing: A Visual Guide to Recording, Engineering and Production" by David Gibson. <<<<

Indeed a GREAT book, with awesome illustrative examples.

Film sound is a combination of making frequency holes, dead space, compression, panning, fades, and a LOT more. Good monitoring tools are the first critical point. Good room is next CRITICAL point. However, every time I bring up good room practices in here, seems to be met with some sort of remark, so I've given up on that particular discussion.

I'll go for the room issue, cause it's critical. Maybe this will help, maybe not, just some ramblings from hobbyist/fool....

I like to sail. While I have many of the electronic goodies that make it easy, I get the most enjoyment from navigating traditionally with chart and compass. The compass had two flaws when using it variation and deviation. Variation is the difference between true north and magnetic north. This is easy to account for given where you live on the globe and applying some simple arithmetic to compensate. We also adjust our compasses so that the variation is as little as possible - like calibrating a monitor. Deviation is different. Deviation is the effect the boat (or metal/magnetism on the boat) has on the compass. There isn't a whole heck of a lot you can do about it. You can relocate the compass to a more suitable position (usually not much of an option), you can change the structure of the boat to minimize its effects (often difficult as well) or you can do a lot of testing and record your findings in a deviation table which allows you to again apply some simple arithmetic to compensate (easiest and most used method).

Monitoring audio is very similar. The best solution is to have a dedicated room tuned to monitoring audio. Bad (ok not BAD but lesser) equipment in a good room will often produce better results than good equipment in a bad room. Unfortunately for most of us, this solution is impractical. I know how to make a good mixing room and if I can't build the perfect room I know how to modify an existing room to work very well (ok, to be honest I have a rudimentary understanding of it, but I do know the right people to bring in to do it). The problem is I don't have a room I can dedicate to mixing and the cost of living in my area prevents me from getting one right now. I think most of us who are hobbyist, or students are in the same position. So we need to think more in terms of deviation as opposed to variation (variation being setting up the room). Yes we should make our monitoring and mixing environment as good as we can within our given limitations, but once that has been accomplished our job is still not done. We now have to account for the deviation that exists because we are mixing in a less than perfect environment.

How? Well first we do lots of testing. Import recordings you like into your mixing environment and look at what setting you need to make it sound correct. Make several mixes of different kinds and play them everywhere you can. I listen to mixes on my boombox, my computer, my home stereo, friends stereos, my car, friends cars, clubs and, most importantly, studios with professional mixing environments. I write down settings, impressions, changes. All of this provides me with a "deviation table" of sorts. No, it's not a problem that can be solved with simple arithmetic, this is subjective after all, but it gives me a good starting point for know what changes I'll generally have to make. For example, if I'm mixing a pop song, I know I have to mix the vocal track lower than what sounds right in my room. I also have to use less bass than what sounds right in my room.

Is this a perfect solution? No, but it does get me WAY further than if I didn't do it at all.

Just some random thoughts while I drink my coffee....

Matt

I am gearing up to do the sound on my short film (building the sound studio, etc.) What would you recommend as the top 3 books to get me started?

Also, when setting up your studio (in a spare room in my house), what are a few key things to make it sound better?

You might want to read my brief article on this subject. You can find it on the VASST site.
Click the login link below, and it will transport you to the tutorials pages. It's called "Get In Tune."

If I were going to the expense of building a mixing room (sound studio) I would bring in a pro (after I read Douglas' article). I'd like to know what Douglas thinks about that first though...

I'm not suggesting that you hire someone to build your studio for you, I mean more of a consultant. Sure you can read a lot and that will get you going in the right direction, but there's a reason those guys do all that studying...

You could probably get a cat from a local university that would do it. Even if you hired a regular firm what I'm suggesting wouldn't be that much. Get them to look at the room before you rebuild and ask what them how you should handle the build-out. Once that's complete, they can tune the room for you and add (or tell you what/where to add) and absorption, diffusion, reflection, etc. These guys will also recommend speakers, placement, etc. Yeah this will cost you more in initial set-up, but then you're done and you know it's right and that's one less thing you never haft to worry about. Compared to the price of gear, their time will be a small investment in a great sounding movie.

I have read many books on this subject and while I mostly "get it", I would still hire someone. But hey, the dice are yours and I don't have a bet on this particular pass line...

Matt

Keep in mind that for reasonably 'standard/simple' rooms, Auralex usually designs/recommends for free, especially in pre-build stages.

Douglas, I have a question. What about for those of us that have been able to commandeer one half of a bedroom, without our SO strangling us in our sleep. I can't keep the monitors away from the wall (or bookcases to their sides for that matter). I've thought of selling the dining room set and setting up there, but I'm too young to die...

What do you recommend in a setting like that?

By the way, I hope you don't mind my ramblings in this thread...

I'm often lonely....

:)

If you can't get the speaks away from the wall, then isolate them as much as possible. Put Mopads beneath them and dense blocking devices behind them. That will kill most resonance hitting the wall. If it's a hollow sheetrock wall, it will still conduct the lower frequencies in the wall, however.

Thanks for the reply. I checked the article--it was great! That really helps me!! I am working in conjunction with my university, so I am hoping to get some more experience in on this short. But I have to go ahead and get started alone in case no one wants to help. :) I still wish to start learning (as much as I can) about the art of sound--any recommendations on books?

http://www.amazon.com/exec/obidos/tg...books&n=507846

http://www.amazon.com/exec/obidos/tg...books&n=507846

http://www.amazon.com/exec/obidos/tg...glance&s=books

http://www.amazon.com/exec/obidos/AS...639590-4019259

These books all have quality audio information. The last is a shameless plug for my most recent book, which has 2 chapters dedicated to this subject.

Thanks Douglas. You gotta make a living you know!!!

I've been working on doing the audio for a film noir short which has orchestral and jazz music along with the DV audio track and I thought I would follow up here with some of what I've been doing. Maybe it will help, or maybe it's all just rubbish, I'll leave that up to you (or David).

The first thing is I probably spent three to four hours working on the dv audio track before doing anything else. I filtered the track to remove unwanted noise (hiss) and then captures plain ambient sound from the various scenes. Once I did that I spent a lot of time going through the wave form and adjusting the gain of each event. Not a lot, maybe 5 db here or there, usually taking a bit away. For sound that had too much of a reverb trail (opening and closing a crystal carafe for example) I adjusted the gain and faded the sound out. I removed some door squeaks and added in a minimal amount of foley (light switch). I try to adjust gain instead of "silencing" part of a track but if I do, I add in the ambient sounds as a floor.

Once I have the events in the proportional levels I would like, I go through it again to make any eq adjustments I might want at different spots. I then look at the dynamic range and compress if necessary. I tend to keep this track centered although if there are sounds I want somewhere else in the stereo field, I'll isolate them into their own track and pan and add reverb accordingly. If the panning is dynamic, I'll automate that function.

Then after I do this and I know where the dv audio will lay, I start writing the score. In this instance it's not to detailed but still I write the music and mix it knowing what I'll be writing around. Then I do all the stuff we already talked about to make the audio present in the mix. I mix the hole thing using automation until it sounds how I would like it to.

The first few hours spent processing the audio (and this was for five minutes of video) was the most tedious, but it is, to my mind, the most important step.

Hope this helps,

Matt