Audio Post Production Tutorials

Assisting on larger films I was constantly using QuicKeys, Excel, Word and Filemaker ,with some AppleScrpt tossed in.

I had a whole flow worked out where I could take the EDL and generate pull lists sorted by roll# and TC. I could then re sort the list for scene for the editors. I needed to do it for something else (I was the first not the Dialog Assistant) and showed the dialog assistants and they kept coming back since I could do it in 10-15 min and it would take one of them hours to do in manually.

In a lot of ways those non sound tools are really important to know. I used WORD almost exclusively to manipulate paperwork coming out of the Avid that somehow never seemed to come to us in the same format.

Other than low budget films that were usually done in my and others small studios most of my work was with the film post houses (almost all now RIP). Some of the ads we did got mixed in SF though.
 
gumby1220 - just saw your tutorial number 4.. Couple questions. Why 2 frame Equal Gain fades? and why 1 frame for perspective changes for BGs? Perhaps you read Purcell's book about fades and took it the wrong way? I believe he was referring to use inverse Equal Power 1 Frame fades only when there are identical IN PHASE audio regions checkerboarding. You don't have identical in phase material since you're getting backgrounds that are completely unrelated. Dialogue editors, however, sometimes have this perspective cut/issue where they have to use the inverse of the equal power fade to null out the 3dB gain increase. I know Purcell mentions this in his book. Remember, if you have 2 sources each spitting out, say 82SPL each, then the sum of both is 3dB more, which is 85 SPL total for both speakers. The reason why he used his particular fade is because you can hear a 3dB gain increase without the inverse equal power fade. Typically, for very straight forward films with no visual fade ins/outs, I only use equal gain fades that are one frame (however, if one scene crossfades into the next, then I'll have a equal gain fades that are equivalent to the visual fade out/in). I've used this technique with a few supervising sound editors and they were taught the same way.

For dialogue editing, I use all TYPES of fades - enough to hide all the crappy noise in the background. However, I do follow Purcell's book when he says to use Equal Power fades mainly. This has worked very well for me and the transitions between regions are seamless - well with a little help of some plugins they are. But regardless of all this, do whatever makes the edits sound clean and smooth.

If I'm working on BG edit session, I typically go into PT's preferences and set my default fade to Equal Gain. If I'm working on a DX edit session, I change the default to Equal Power. For crossfades, I always leave it at Equal Power and I haven't changed it since.

Are you looking into get into SFX editing professionally? If so, I recommend getting Soundminer or AudioFinder or some kind of asset management tool. The spot to timeline function saves a lot of time and has taken away the need of my "edit" or "work" tracks. The only time I use those "work" tracks is when I'm cutting dialogue (to find roomtone or something else) or when I'm sound designing something. For BGs, all of my edits are done within the track. I highly recommend getting QuicKeys as well for reasons explained in the above posts. I think it's an invaluable tool for many editors. Heck I use it to generate ADR scripts with PT, FileMaker Pro, Excel, and Preview.
 
I would agree with a lot of what Ryan says. I was trained to do 1/4 frame head and tail fades (scrapes), but that has a certain amount to do with the mix process, which doesn't nec. apply to no budget post. My general pref is to use equal power all the time n fades but that's me. I generally don't do the transitions with fades. I automate levels, ie mix. The fades are just to keep things from popping on the cut. On a hard cut the fade is all that is usually needed. There isn't really a "rule". I start out with the least I can get away with and adjust till it works. Sometimes you need a clean cut sometimes you need a lot of overlap.

I do still use a mono and Stereo "work track". I generally bring sounds in (with Soundminer) and edit in PT. I bring things in on a work track so I don't run the risk of overwriting regions already laid down. Also I record some things from the sampler so I record into a Work Track and then move it where I want it to live.

There isn't really a "right way" and a "wrong way" to do it. As long as it works for you and your fast enough with it go for it.

I tend to bring in a bunch of prospects and "chose in place" so I'm not bringing in "the" sound but a bunch that might work.
 
gumby1220 said:
I am starting to make a tutorial series on how to get a video setup properly to be imported into pro tools
Click to expand...
OMG, people still use Pro Tools?
<joke, joke>
I think the main thing with fades, regardless of which software you prefer, is you need four point cross. And here, reluctantly, I do slightly prefer Pyramix to Sequoia as far as features. Any platform with four point cross is fine. But maybe I'm delusional from listening to too many fades. I have to accept that it may be possible.
 
Last edited:
Noiz2 said:
There isn't really a "right way" and a "wrong way" to do it. As long as it works for you and your fast enough with it go for it.
Click to expand...

Yea there's no standard at all. The last few shows I did, the wanted 1 frame equal gain fades since a lot of mixing was done with PT10 and they want to use clip-based gain. Which made a whole lot of sense.

For SFX and sound design editing sessions, I 'import session data...' a template that I have which is a set of mono audio tracks, stereo audio tracks, verb returns, and master aux's, which is an addition to the AFX, BFX, CFX tracks. This gives me a good set of tracks to work with if there are more sound design-ey stuff.

No right or wrong way. Just make it sound good.
 
DrDave said:
I think the main thing with fades, regardless of which software you prefer, is you need four point cross. And here, reluctantly, I do slightly prefer Pyramix to Sequoia as far as features. Any platform with four point cross is fine. But maybe I'm delusional from listening to too many fades. I have to accept that it may be possible.
Click to expand...

That may be more of a music thing. In film post 90% of fades are just to smooth out the ending of a region. Almost all the rest are Xfading looped BG's. Transitions and other mixing chores are done in, well, the mix.

Now the lower the budget the smaller the crew and the more the editor is "mixing" as they build, the more you will end up using fades as more of a mix function. But I rarely use fades much in that capacity. Also most of the time you have many layers of BG going on so it often doesn't matter much if the Xfade isn't perfect since you won't hear it anyway. Again the more your trying to do it all in one the bigger the fade issue is.

Music editing is where I end up using fades in a more critical way. And if you were doing a full dialog edit the RT XFades would be pretty critical.

I'm not advocating crappy fades but what is acceptable in say a city BG XFade VS what is acceptable in say a string loop are significantly different.


BTW what is a "four point fade"? I have never heard of the term. I'm guessing separate Ins & Outs. You can kind of do that in a XFade with different curves but in PT the only way to actually do it would be with checker-boarded tracks. Generally, for post anyway, you would want to checkerboard any tracks that you would need different in's and outs on so... But it would be a nice addition to the tool kit, assuming I'm talking about what you are ;~).
 
ryanjf said:
gumby1220 - just saw your tutorial number 4.. Couple questions. Why 2 frame Equal Gain fades? and why 1 frame for perspective changes for BGs? Perhaps you read Purcell's book about fades and took it the wrong way? I believe he was referring to use inverse Equal Power 1 Frame fades only when there are identical IN PHASE audio regions checkerboarding. You don't have identical in phase material since you're getting backgrounds that are completely unrelated. Dialogue editors, however, sometimes have this perspective cut/issue where they have to use the inverse of the equal power fade to null out the 3dB gain increase. I know Purcell mentions this in his book. Remember, if you have 2 sources each spitting out, say 82SPL each, then the sum of both is 3dB more, which is 85 SPL total for both speakers. The reason why he used his particular fade is because you can hear a 3dB gain increase without the inverse equal power fade. Typically, for very straight forward films with no visual fade ins/outs, I only use equal gain fades that are one frame (however, if one scene crossfades into the next, then I'll have a equal gain fades that are equivalent to the visual fade out/in). I've used this technique with a few supervising sound editors and they were taught the same way.

For dialogue editing, I use all TYPES of fades - enough to hide all the crappy noise in the background. However, I do follow Purcell's book when he says to use Equal Power fades mainly. This has worked very well for me and the transitions between regions are seamless - well with a little help of some plugins they are. But regardless of all this, do whatever makes the edits sound clean and smooth.

If I'm working on BG edit session, I typically go into PT's preferences and set my default fade to Equal Gain. If I'm working on a DX edit session, I change the default to Equal Power. For crossfades, I always leave it at Equal Power and I haven't changed it since.

Are you looking into get into SFX editing professionally? If so, I recommend getting Soundminer or AudioFinder or some kind of asset management tool. The spot to timeline function saves a lot of time and has taken away the need of my "edit" or "work" tracks. The only time I use those "work" tracks is when I'm cutting dialogue (to find roomtone or something else) or when I'm sound designing something. For BGs, all of my edits are done within the track. I highly recommend getting QuicKeys as well for reasons explained in the above posts. I think it's an invaluable tool for many editors. Heck I use it to generate ADR scripts with PT, FileMaker Pro, Excel, and Preview.
Click to expand...

Although I am in the process of reading Purcell's book (awesome read so far), but the information did not come from there. The 2 frame equal gain fades are really just a starting point and the 1 frame perspective changes because normally the fades don't need to be as long as a full scene change. I thought I mentioned real quick in the video that the fades would need to be played with to get it to sound right. I may have forgot to mention that though. But yeah just really starting points and the way I was taught to get things going then fine tune later.

I would definitely like to do SFX editing professionally, but I've been having a hard time getting experience in outside of school and extremely bad indie films. I mean really bad shot to where I won't even show it as part of my portfolio. I've looked into Soundminer and it looks great as well as QuicKeys but right now to where I am still looking to break into the field professionally. It really wouldn't serve a purpose.
 
Last edited:
A four point fade allows you the best control over any fade. The fade is divided into four quadrants, or points, per track. You can adjust each panel independently. IMHO, four point fades are cost effective for two reasons: you will make your fades much faster and much better.
There is also four point mutitrack fades, but unless you do really big scores, like big film soundtracks, large orchestras, etc, regular four point is best.
Four point allows you extremely flexible adjustment over the final sound, and, in addition, it allows you to make seemingly impossible fades possible. There is some vagueness about the terminology, and you will hear even three point and so on, but basically you need two things: four independent visual panels that hold all edits in the project either locked or unlocked, and the ability to cross cut using these points.
Currently the four point Daws that are mainly in use are Pyramix, Sequoia and Sadie. Of course many of the refinements to the system were made by Sonic Solutions back in the day.
Without this kind of editing, I would find it very difficult to do tiny repairs in video, or to seamlessly transition from one take to another.
 
DrDave said:
A four point fade allows you the best control over any fade. The fade is divided into four quadrants, or points, per track. You can adjust each panel independently. IMHO, four point fades are cost effective for two reasons: you will make your fades much faster and much better.
There is also four point mutitrack fades, but unless you do really big scores, like big film soundtracks, large orchestras, etc, regular four point is best.
Four point allows you extremely flexible adjustment over the final sound, and, in addition, it allows you to make seemingly impossible fades possible. There is some vagueness about the terminology, and you will hear even three point and so on, but basically you need two things: four independent visual panels that hold all edits in the project either locked or unlocked, and the ability to cross cut using these points.
Currently the four point Daws that are mainly in use are Pyramix, Sequoia and Sadie. Of course many of the refinements to the system were made by Sonic Solutions back in the day.
Without this kind of editing, I would find it very difficult to do tiny repairs in video, or to seamlessly transition from one take to another.
Click to expand...

Hmm. I still don't know what a four point fade is. It sound fairly complicated and time consuming. From a post sound perspective none of the apps you mention are at all common in the US. I think they are also all PC. I don't care much but post sound these days is about 99% Mac. Of course than can change, it used to be much more even.

There are a few folks that loved sonic Solutions but most editors hated it. Mostly because you could make changes to things not on screen and you wouldn't find out till you were on the stage. Music folks liked it, bit that may be because it was 24 bit before everyone else and of course they weren't working on long timelines.

When I was doing big films music would often be in Sonic, Dialog and ADR (and sometimes Foley) were often on Waveframe, and FX were on Pro Tools. Pro Tools being the last man standing is the tool du jour now.

As to fades, my point was post sound as a rule isn't using fades to be seamless. We are almost never crossfading two sounds that are not layered over a bunch of other sound that are not crossfading at that point. If you have 30 or 40 tracks playing a lot of this stuff just gets lost in the mix.

People have a tendency to get obsessed with small points that in the big picture of post are often a waste of time.

Not that this is nec one of those but it's starting to sound a bit like the quest for the holy grail of fades. IF your post sound work is centered around editing production tracks into a smooth track then I can see spending a lot of time on it. But that is not sound post on narrative films and really on almost anything but maybe news shots and very low budget films. But I would argue that in the later (and even in the former) case they are not working smart by not layering up.

I've worked the other way when I was working on 2,4,8 track tape, but with a DAW there is no good reason to not use lots of tracks. It make life easier and your work, faster and better. Checkerboard and use automation and you can do anything you want in a crossfade.
 
Try it and you will never go back :)
The four point system is less complicated and faster. You can make it more complicated it you want to, but you don't need to. I understand there are a lot of Pro Tools users, but when I'm in Europe I just don't don't see it. The only thing I would say, really, if you do 5,000 edits a year or more, then it is definitely worth looking into because of the time savings. I think there is a reason they can charge $3000 for a program and the reason is the cross fade editor. Why else buy it? OTOH, a lot of what we do is just zipping along with what we know. I hate learning new programs. I would rather just crank out tracks with what I know. I haven't tried the most recent version of PT, but I wouldn't use it for cross fades. The other thing I would mention, and this of course is subjective, and many debate it, is that some DAWS just seem to sound better at physical modeling than others.
 
WE are kind of sliding away from the thread but it's also a good point.
I'm just guessing Dr.Dave but I suspect you are in the music end of things. There is a much wider variety of platforms in the music biz and part of it is that in general projects stay in one place and are worked on by a few people. Film, especially big film, has hundreds of folks that all need to be working together. It was a huge pain in the but when there were a bunch of different DAWs on a show. If you have three DAWs on a show it means you need to have three separate systems all tied to the stage for final mix. It means you need staff that can support the different platforms/ OS's and it means editors get very locked into their little niche because to move to a different category means learning and getting up to speed on a new platform/ OS.

So there was a LOT of pressure to come to a standard. Digi happened to be doing enough of the right things at the right time and became the standard in the US. WaveFrame had been the semi standard first but in 1998 they were still building their systems around Win 3.1! They were slow (3.1 didn't support the newer processors) and it was a dedicated, locked in hardware system that started at around $30K. PT was cheaper and more flexible and worked with whatever the latest hardware from Apple was. Sonic Solutions just dropped the ball. I worked at a place that wanted to go all Sonic and we did a lot of testing for them. They were VERY slow to fix bugs, they didn't support many plugins and they were more expensive. The killer was they just didn't keep up with the changes going on or open there system up and get enough third party support.

The platforms your talking about and others like Nuendo came on the scene after PT owned the US post market so it's very hard for them to get a toe hold here. Now with AVID running things they may make enough bad decisions that others will have a better chance.

And Apple made a big (and successful) effort to position the Mac as the computer of choice for artists and filmmakers. So there is also a bias here to Macs and PT and Nuendo are really the only players then. Now Apple could make some bad decisions (like discontinuing tower computers) that will hurt them a lot in post circles so...

Europe on the other hand is a different ball of wax. First off Macs are MUCH more expensive there (for tax reasons I think?) so Pro Tools never took off (it wasMac only till fairly recently). So neither Macs or Pro Tools were ever big players out side the US.

Also most of the world makes films in a different work flow than the US. Much smaller crews, faster post, and less detailed post. Just a leaner meaner filmmaking machine. So some of the crew networking requirements here are less of an issue there.

The bottom line is today if you want to work professionally in film sound post in the US you need to know ProTools and Macs.

If you get a staff job someplace you need to know whatever that place uses but as a freelancer (all the post folks on major films with a VERY few exceptions are freelancers) then PT and the Mac are 95% of your job prospects.

IF Windows can get past using Win 3.1 design specs for the look and feel and Apple makes some bone head moves things could change relatively fast. PT went from low budget fringe to main player in about five years.


Cost is not much of an indicator. The latest PT10 HD software only upgrade for PT HD9 users is $1,000. An HD system will run you around $20,000.

And as I mentioned about 90% of the fades I do are by highlighting a mass of regions and having PT put auto fades on all of them. If PT had a zero crossing feature, I probably wouldn't even be doing that. Of the other 10% probably 9/10 of those are checkerborded tracks.

As cool as this four point fade may be I'm not going to change platforms for the remaining 1%.
 
RE Tutorial 4.

Generally that is not a good way to loop BG's. What tends to work better is to trim the head and tail a bit. Duplicate. Adjust the head of the dup till it loops pretty well then dupe the dup and adjust it's head. Once you have a loopable selection you pop that back over the first region and dup till you have enough and then auto fade the whole group. But it does depend on how loopable the sound is in the first place.

It's also not common at all to create a bunch of phantom regions as a spotting template. It's also not all that practical since your doing the work twice, kind of. I know people who do extensive spotting by dropping markers but even that is not that common.

On big budget films it varies by locale and supervisor preference. As a general thing in LA a supervisor will extensively spot a film and have assistants pull sounds and that will get handed to the editor. They may have a number of choices and can ask for more but the job is to cut what has been chosen. In the Bay Area (northern CA) supervisors generally spot more for what needs to get recorded. Editors and their assistants are generally pulling sounds them selves and editors are relied on more to find what they need or ask for new recordings. Partly this is because in BA places tended to have smaller tighter crews that tended to know the library and often brought their own library in to play, where in LA editors move around a lot more and don't know the libraries as well.

Not sure the NYC mode but I suspect it's closer to BA than LA.

Anyway my point is there isn't much reason to have fantom regions. If you see it you hear it pretty much takes care of the spotting. What you do generally want to spot are things you will need to record.

Also A and S respectively are head and tail trim keys. Using B and delete is the long way to trim.

And a helpful hint. If you do a find in the region list and leave it blank (but click the add new regions) it will clear out the region list and only the stuff you bring in will show up. Very handy when adding FX. You can always clear the find to get everything back.
 
RE: Tutorial 5

The E key is your friend for the kind of trimming you were doing.
Time Stamp and export region definitions??? That is baffling. First what use is time stamping them? I guess if your planning on using the same region at the exact same time in multiple videos... Still makes very little sense. Maybe for pops. That could be a use.

Exporting region definitions is also kind of problematic. Again it only makes sense IF you are going to use the exact same region someplace else. If that's the case you probably should have named the region so you would know what it was. Also since Digi abandoned SoundDesigner there is no way to edit region definitions, only add them. So every time you do that you add another region to the file.

Also when you are going to do something like head and tail fades on a bunch of regions you can highlight them all and do it in one step.

Sorry about this a bit rambling post but I'm flipping back and forth from the tutorial to this post so...

I know this is a mechanics series but one of the reasons for FX that phantom regions don't work well is that say with your splats. What is the critical sync element? The phantom region gives you the beginning and end but so what? Neither is really very important. What IS important is that the HIT of the sound line up with the HIT in the visual. The way you would spot that is with a marker. You put a sync point in your region at the sound HIT and set your cursor at the visual HIT and snap the sync point to the cursor. You can then trim head and tail as needed. The way you are doing it the hits are not even close to where they should be. It's really misleading implying that this is a good way to cut impact sounds because it's not at all.

Another speed tip. Anything you highlight will get deleted when you hit delete. So when trimming out a bunch of dead space just highlight the dead area and hit delete, no need to use the B key.

It's a personal pref. but very few editors lock everything. Some do late in the game when things are really pretty stable but many don't ever lock anything.

And one thing that really I should have mentioned before but every day you open a session you ought to do a save as with the current date. Most editors also do a save as with a date and a point x periodically in the day. Not having a backup of your session is a fireable offense on most films I've been on. It also saves your but when things go south.
 
RE: Tutorial 6

For ADR there are a few things that you will want to do. First you never want them hearing the original line on top of what they are recording. Usually what you want to do is have them hear the bit you want to dub and then repeat it. They will get much better timing that way. Some folks show the actor the video, but this often results in awkward phrasing as they try to catch up with the lips. If you are doing the listen repeat method you don't use beeps, you only use them with the visual method.

Loop record usually doesn't work well. Actors start feeling like they are on one of those Lucy assembly line skits and are struggling to stay up to speed. If you do loop make sure to leave enough of a breather before the loop starts. Some actors also like to do the line more than once in a row so you have to be careful about stopping too soon.

VO is more flexible, but again a lot of actors wont want to loop record.

For Foley the method you have demoed works pretty well. Though even there loop record can be problematic.
 
You must log in or register to reply here.
Back
Top