Sound design request

R

ronin99

Active member
I'm nearing picture lock on a film and I received an interesting request from the individual that will handle the post sound.

The person wants me to break the film down into 15 minutes reels and send the sound that way. Their reasoning is
"the track count can get rather high and I won’t be able to do as much sound design if the film is in one giant chunk".

Have you ever experienced this? Maybe I'm crazy and this is the norm in some circles but whenever I have transferred files to a sound person in the past, I'll send them the entire film and it's never been an issue.

The only thing I can think of is perhaps they are using an older system that can't handle it?

My main aversion to this is that it needlessly complicates matters and creates much more work for me.
Your thoughts?

Thank you,

Joe
 
This is the standard, or was, for 35mm films. Actually 35mm is usually edited in 10 min reels. That translates to about 1000 feet of film and is what a "standard" reel holds. For mixing you would "join" pairs of eel and make "A/B" reels that were around 15 - 20 min long. 2,000 foot reels are what is used by theatres that don't have platter systems.

In post sound you wanted to mix to picture so that was 35mm, but even if you could do it in one big stream you didn't want to because you have to create a bit of overlap so the theatre can "change over" from one reel to the next.

I'm going to take a wild guess and say this film is not destined for a 35mm release. So there is only one remaining reason to break it up into reels. And that is because of file size. OMF has a 2 gig file size limit and ProTools wants an encapsulated OMF so if you have much in the way of tracks you can get into trouble if you try to do it all in one.

But a simpler solution is to break it up "vertically". Send full length files but only select a few tracks at a time so you stay under the limit. It's easier for you because you just disable tracks before making the OMF's and it's easier for them because they don't have all these parts of a film. They can import the tracks from the OMF's and end up with everything you had and be able to send you a full length mix. Which is a LOT easier for you to bring back into your session for final rendering.

The good news is that it looks like you got someone who is used to working on 35mm films and not some high-school student with their cracked version of ProTools.

For the sound editor it is also a bit mentally easier to work in reels because you have these hunks to finish and it doesn't feel so HUGE when you are working on it. But they will get over it. And they don't have the PITA of walking you through reassembling the film. In 35mm you don't reassemble since you are going to print in 2,000 ft reels.

The track count reasoning ,actually doesn't apply. The track count IS going to get high if they do a full job but splitting it into reels isn't going to help that.

If they have a bunch of editors and want to split the film up by "reels" they should just give everyone the full film and assign them a section. Or go the other route and have editors work on elements for the whole film.

traditionally you either assign a reel to an editor or a "food group". So either I give you reel 1,2,3 or I give you backgrounds for the full film. I actually like the latter mode better since it tends to have better continuity.

If they balk at doing it this way you can have them talk to me if you want. We probably come from a similar place and I can probably convince them they will save a ton of time going the "new reeless" way.

If for some reason you are going to make 35mm prints then you do need to split up the film.
 
I have always tended to work with large projects split into reels or in the case of TV drama parts.

This makes sense even in the digital domain as it allows things to be split up for better file management and builds in a certain amount of discipline and safety to the project.

Even if this is done in the virtual domain using bins at the edit stage it can also build in easier time management to a project where for example a reel is worked on each day for whatever track laying or editing is required. It can then also help in time scheduling post processes and can give production feedback of how a project is progressing in the post process.
 
The indie features that I did were always busted out into reels even though they never saw a foot of film. I found it easier for me as a sound designer because it broke the movie up into chunks and it was easier to manage, and I knew that "oh, the scene with the dream sequence is in reel 3", rather than having to drop in lots of markers into my Pro Tools session or search for it.

I never requested that the movie be broken into reels, though. It was always the re-recording mixers. I really didn't care either way. I just went with the flow. For me having a timecode burn in the picture window and a 2-pop was a must.
 
IF you go to reels, especially out of FCP you have to do a lot of prep work on the picture end. You need to put leaders and two pops AND tail pops on all the reels. You have to do this IN the timeline and not the auto functions that FCP has.

When it's all over you have to reassemble the film and the video editor will need to join the reels together (including sound). It is really kind of silly to do all of that for video. I learned on reels and did a few low budget films where I made them break it up, but I got talked into trying it the otherway and haven't had a film (that wasn't going to 35mm) split in reels in years. It is easier for editors (providing you have assistants to assemble it all) bu,t it really doesn't take that much discipline to work on the whole film. If you are working in A/B reels you are talking about 5 (FIVE) markers you would need to drop.

IMHO it's silly to go to a work flow that was forced on the industry because of the physical limitations of projectors when there is no actual benefit and a whole lot of work involved. It's an extremely wasteful workflow. It requires a lot of tracking and adding of material just to remove it later and a bunch of places where you can go out of sync and have problems.

On a big film with a big staff it makes some sense because the reels can evolve separately and so much is changing as you work that there is a counter to the waste because not everyone needs to conform every time the picture changes. In a no budget workflow where sound gets a locked picture and you often have only one or two people working on the sound there is no savings and a bunch of extra work.

It's especially problematic if not everyone is familiar with the reels workflow.

Don't get me wrong there is a nice nostalgic feeling working in reels. And if I'm only thinking of me as an editor then I prefer it.

But if I'm also mixing then I will have to join the reels at some point, and I have a bunch of duplicated files as I switch sessions to find out what the BG's were in the apartment in R1 and make new copies in R6. Just the splitting and joining will probably add a day or two, and that is a day or two you were not actually working on the film, just doing housekeeping.

For a video project it's anachronistic to work in reels, unless you have a big staff and/or you are getting constant changes and having to conform.
 
Gary, I think your signature is off. I assume it should read "Over 15 YEARS in broadcast...", or are you being funny?

I just noticed.
 
Thanks for the replies. I see how this can be beneficial. However, Scott's last post summarizes how I feel about this.

Let's say I did go this route of splitting the film in 15 minute segments.

Since I currently have the entire film in one sequence, would the correct workflow be:

1. Break the film out into six new separate sequences of 15 minutes each.

2. Make sure the 1st frame of pic starts on the hour for each of these sequences.

3A. Place the audio 2-pop at start and end of each of these sequences.

3B. Add time code to the video of each sequence, should it be a TC reader or generator?
I'm also unsure if the timecode for each sequence should read 01:00:00:00 - 01:15:00:00, or should each sequence read 01:00:00:00 - 01:15:00:00, 01:15:00:01 - 1:30:00:00, 1:30:00:01 - 1:45:00:00, etc...

4. Export the files in the desired format.

5. Once work is completed, get the mixed tracks back from post sound.

6. Sync up tracks to each sequence.

7. Create a final new sequence, and assemble the reels to get the full 90 minute film.


I also assume I would want to try and find a natural breaking point for each of these reels. However, what if you would like to have audio overlapping scenes that occur between reel breaks, not necessarily dialogue, but it could be music, or background ambience.
 
Noiz2 said:
Gary, I think your signature is off. I assume it should read "Over 15 YEARS in broadcast...", or are you being funny?

I just noticed.
Click to expand...

Ha Ha 10/10 for observation and spotting my little tribute to Andy Warhol. :laugh:

Over 33 years now so I must be doing something right ???
 
Let's say I did go this route of splitting the film in 15 minute segments.

Since I currently have the entire film in one sequence, would the correct workflow be:

1. Break the film out into six new separate sequences of 15 minutes each.
Click to expand...

yes

2. Make sure the 1st frame of pic starts on the hour for each of these sequences.
Click to expand...

Each reel has FFOA on the hour of the reel #. So Reel 2 starts at 02:00:00:00

3A. Place the audio 2-pop at start and end of each of these sequences.
Click to expand...

Two pop is two seconds before FFOA, tail pop is two seconds after LFOA (except for the final reel which has no tail pop).

3B. Add time code to the video of each sequence, should it be a TC reader or generator?
I'm also unsure if the timecode for each sequence should read 01:00:00:00 - 01:15:00:00, or should each sequence read 01:00:00:00 - 01:15:00:00, 01:15:00:01 - 1:30:00:00, 1:30:00:01 - 1:45:00:00, etc...
Click to expand...

Traditionally since this is a workflow for 35mm film the TC would be both LTC and F+F as a window burn and as a TC track (on the tape). Files don't sync via TC so you only need the window burn.

4. Export the files in the desired format.
Click to expand...

OMF and video for each. I also would want an XML to help conforming those changes you aren't going to make ;~)


5. Once work is completed, get the mixed tracks back from post sound.

6. Sync up tracks to each sequence.

7. Create a final new sequence, and assemble the reels to get the full 90 minute film.
Click to expand...

yep Don't forget you have to cut off the added pops and leader etc.


I also assume I would want to try and find a natural breaking point for each of these reels. However, what if you would like to have audio overlapping scenes that occur between reel breaks, not necessarily dialogue, but it could be music, or background ambience.
Click to expand...

Yes. Part of the trick to reels is you need to find a scene break where either everything changes or where you can have some neutral BG that will cross over seamlessly. A break is easier. You can't have music cross over, or just about anything sound wise that could create a jump if the reels are not switched exactly right (for film playback).

It also means that for mixing you really have to be set up and calibrated because you won't be able to hear from one reel to the next till you reassemble the film.

On a big feature (back on mag) you would duplicate a bit of the head of reel 2 printmaster and splice it on to the tail of reel 1 so that there was a bit of overlap for changeover. These were called pull ups (or downs I can't remember).
In your case IF they add some head and/or tail sound and you stack the mixed tracks you can do a little crossfade in the dead change over area and get smooth transitions.

Eazy Peazy.

Well for the sound editors anyway. That is until you do this with a video editor who has no clue about how this all works and you have spend days walking them through it. Or something gets nudged and you have a reel out of sync, etc.

Basically your doing a bunch of shorts instead of one feature so many more chances for some mistake to gum up the works.

It is a better flow for features with a post crew since changes to reel 3 don't affect reel 2 or 4, so the folks working on the reels that didn't change don't have to stop to conform. But it also mean someone has to keep track of all the various versions, since different reels are at different version and depending on who has and has not conformed different departments may be working on different versions of the same reel.

I have Excel templates and FMP data bases for all this, someplace, if you want to get into the real "big budget feature" flow.

But there is a reason why the #s you seen on the sound crew are what they are, it takes a lot of coordination to keep that all going.
 
One other small, or not so, thing that can happen is that since you are never seeing the film as a whole in sound post, it's easy to get an impression about how something plays that is wrong.

The one that really stands out is the freezing survivors scene in Titanic. That scene was the start of an A/B reel.

So when mixing everyone gets settled and ready and we start with the deathly calm, and it was heart wrenching. I mean REALLY hard to watch. In fact the picture editor wouldn't anymore. So I know this is going to just rip the audience up.

The first time I saw the film as a whole was in a theatre, and the scene plays OK but no where near the impact. Why?, because the ship JUST sank, it's loud and dramatic and it takes the audience three or four minutes to settle down after it. By that time we have passed all the quiet death and are back into dialog. They wee screaming (studios) to cut as much time out as possible, but another minute to the head of that reel or some quiet at the end of the one before would have made a Huge impact in the effectiveness of the scene.

Now I'm sure it was screened as a whole before it was released but it's really too late to make changes unless you have more clout than god. IF we had been working on the whole film at once it would have been obvious early on and maybe something would have been changed. Or maybe nobody would have noticed how powerful that scent could have been? Hard to say.

But the only reason films are done in reels is because of the historical physical limitations of 35mm film projection. A side benefit is the compartmentalizing of changes. There weren't many if any changes before magnetic tape made sound editing flexible enough to accommodate changes and they still were not a huge deal till Dolby made magnetic recordings quiet enough that you could mix a lot of tracks and not end up with noise soup. And all of that is long after the reel workflow was fully entrenched.


Hey here is an old timer test. What is "popping a core", and what is required of someone who "pops a core"?

First correct answer gets a prize.

I couldn't find a reference with Google so this will take some real knowledge or a good search ;~)
 
GaryNattrass said:
Ha Ha 10/10 for observation and spotting my little tribute to Andy Warhol. :laugh:

Over 33 years now so I must be doing something right ???
Click to expand...
Funny, I always thought it meant you had tributed 15 minutes into movies. I figured if they use 1-second cuts then 15 minutes would still amount to quite a bit of work done..
 
I have Excel templates and FMP data bases for all this, someplace, if you want to get into the real "big budget feature" flow.
Click to expand...


Hey Noiz,

I was reading this thread and realised that this is what we should have been doing all along =/ there's always a next time I guess...

I'd be very interested in that templates and workflow files that you have lying around if you're willing to share them =)

Always learn something new, that's what I say!
 
I'll have a look, it's been years since I had to deal with reels (at least as a first assistant).

I had/ have templates for tracking LFOA's, really critical when working in changing reels.

I also had a workflow using Word, Excel, & Filemaker Pro to extract a pull list from an EDL. In theory it should have been simpler but the EDL's were a moving target so it always took a bunch of interaction to get them to parse reliable. Once parsed into a delimited file you suck them into FMP and it would create a pull list by sound roll and TC and eliminated duplicates. This used to take a dialog assistant the better part of a day, I could do it in a half hour or so.

The other thing that is pretty much a necessity is a large white board with reels vertically and departments or food groups horizontally. The first column is whatever the current version is for the reel. At each intersecting box you have the version they are working with.
 
OK so the really polished stuff is on a zip disk in a box someplace. There is just a ton of stuff I still have packed in boxes from the move. The good news is probably most of those files are openable and I do have a zip drive that can hook up to the computer so when I find the disks...

In the mean time I did stick some earlier files up.

There are some FMP databases for cataloging SFX - these were the days of DAT and optical disks so you would probably want to change some of the fields.

A FMP for loging and tracking ADR

A couple of Excel LFOA lists. THey are for film so there is the dupe-date and LFOA in F+F plus they calculate the F+F for the tail pop.

While the convention is for TC to start at FFOA F+F starts at the start mark so FFOA is at 12feet which is why the FFOA is always 12+00. Also while TC starts on the hour of the reel # F+F always starts at 00+00 on the start mark.

There is also a Cut List.

One of the "fun" assistant jobs was to scroll through the entire film and log the in and out locations of each scene.

There is also a Q-Sheet check list. When your making up Q-Sheets for every reel you need to keep track of what is done and preped etc.

"Binky" BTW is what they called the Q-Sheets that were created from the premix. The mixers would mark up the Q-Sheets with where they recorded everything and we got to plow through all that and create new Q-Sheets that reflected where the sound now was. So for instance a gun shot might be on track 6 of the AFX for reel 2 and it now is on the AFX stem for reel 1A/B. Remember editing on 1,000 ft reels and mixing on 2,000 ft reels. So your going from 10 edit reels to 5 A/B reels in the premix.

The one Titanic FMP file I have handy is more for the picture department - and then handed to the sound department. This one I didn't create but it's quite nice so I have used it a number of times.

It's a code book. The place you enter every foot of film and sound created in production and all the FX shots etc.

If picture does something like this Dialog has a much easier time finding alt takes and places where you might be able to steal some RT.

I'll add to the folder when I find other stuff.

The LINK
 
OK here is another semi useful file SOUNDSEARCHER.

This is a spruced up version of a cataloging db I used a lot till I plunked down $$$ for Soundminer.

It is a FileMaker Pro db so you need FMP to use it. Log in as Guest.
To make it actually useful you have to have Applescript installed, so it's only going to work on a Mac.

What it can do is catalog a directory of files. You can add more directories as needed.

Once in the db you can play the sound by clicking on the name. You can go directly to the file in the finder, you can search the db, and narrow or expend the search. Clicking on the box on the right of an entry will sort the db to show only the files in the same path. The check box to the left "flags" the entry. You can copy flagged files to a new folder. So you can search and find stuff you want and then make copies to a folder that you can drop into your project.

It works decently for what it is. I was / am porting it to something that can be cross platform and doesn't require Apple script or Quicktime to work. But that is a project that could take a LONG time.

I originally created it for inhouse work because there wasn't anything decent around. now there are a bunch of applications around. It's possible that there might be a market if it were cheap enough but? I may finish the port as a way to learn Xojo but it would need a number of new features to make it attractive. Some I did in the past, like notes - descriptions, that broke on the move from fmp6 to 11, and some like spotting into timelines are trickier. You used to be able to do it with ProTools and Applescript but Digi broke that awhile ago. If I was enough of a programer to wade into the PT API I could probably do it (Soundminer does) but that may be above my pay grade.

BTW it's "locked" to keep people from breaking it. If you want get in and make some changes drop me a line, there is nothing super secret in the coding.
 
Hi Scott,

Thanks so much for your help. One more question, so on this project, there's one person handling all post sound (sound editing, sfx, mixing) and one composer. Since I'll be sending reels to the Sound designer, should I send reels to the composer or would you recommend sending the entire film as one long piece to the composer? I'm looking for the method with the least amount of problems, I can see some issues that could pop up like where the composer could compose music that would overlap reels, etc...
 
Personally I would be sending full feature to both. I would ask you post person how they are planning on mixing with reels?

But if he is mixing in reels then he won't be able to use any music that crosses over a reel.

I know he asked for it that way but I would sit down with him and your composer and go step by step through the post workflow.

As I said the reel based workflow was mandated by physical limitations connected to 35mm film. It's not a simple workflow and was done on films with budgets and staffs. You didn't see a lot of no-low-micro budgeted films before video, and most of those were on 16mm.

So your question is very on point but needs to be answered by you post crew since they need to be all on the same page.

I suspect you post guy has some experience as an assistant and maybe an editor on bigger films and wants to do it the way they work. It's not a bad thing but he is not going to be able to hand off his reels to the mix stage. I suspect he hasn't gone through the full workflow to final mix working this way on his own.

But I could be wrong and he know exactly how it needs to go.

So talk to him and bring up these points and step through the workflow all the way to your mastered DVD (or whatever your final delivery is) and make sure nobody sees any problems with the flow.

IF you have cut reels on dead spaces (scene changes etc) and he has really calibrated his system AND he is grabbing a hunk of the mixed reel before and tagging it on to the head of the next reel for reference, AND he makes sure that each reel starts and ends ON a full frame AND your composer doesn't have anything that drifts over the reel change, then it might turn out fine.

I will say that I personally think you need to stay with one layout. So IF you are working reels with FX you should be working reels with the composer. I would also recommend that IF you work in reels you work in A/B reels (about 20 min) which is what a 35mm films would be mixed in and some are edited in rather than the shorter "edit reels" (about 10 min).

Unless you are still making picture changes though I think working in reels is a lot more work than it's worth (for video) and has a lot more places where problems can creep in.
 
Finally remembered where this topic was!

Thanks a lot for the resources, Scott! I'll give them a thorough studying once I get back to my regular programming. There's been a lot of troubleshooting and conforming during post prod on this film we've been working which could have been prevented. Like I said, i should have read this post earlier! =b
 
I've completed a few projects since this initial thread and whenever I send reels, in the end, I will receive a single stereo or 5.1 track back for the entire film. Therefore, that process is what I've become accustomed to. However, this has been working with freelancers.

I'm working with a post house for the first time, and I delivered the reels. When I mentioned delivering a single file back for the entire film, it seemed like it caught them off guard, which was strange to me as I thought this was standard.

I mean if I get a file for each reel, how do we know that the audio will sound good across the reels? Am I really supposed to put the audio from the separate reels together, then adjust the levels between reels myself?

So post sound guys, chime in, what's the industry standard workflow when working with reels?

Thanks
 
The design/ edit happens on a reel basis, the mix is where you get the finished file. If you were going to a film release you would be mixing to 2,000 ft reels since that is what the theatres want but you started as a single piece and I'm assuming you are going to release in video so they should mix to a single file.

When they are doing the mix they will be balancing between reels etc. so you should end up with a single file you can just drop in.

I have said this in other places before but in film, and most things, you really want to start at the end and work your way back. You split the reels because they wanted you to, it should be on them to get you the mix you need which is either a series of files that will lay in and work or better a single file that will lay in and work. Assuming part of your deal with this place is that they are providing the mix you shouldn't have to do anything but assemble the sound tracks with the picture.

Honestly they made more work for them selves by splitting it this way. They could have just worked it by "virtually" splitting in to reels. Keep the video file as one big file and assign break points to the editors. That would have simplified the assembly into what you needed in the end.
 
You must log in or register to reply here.
Back
Top