I've never edited anything on film, so I'm trying to figure out the workflow between the lab processing of dailies and the final locked edit before negative cutting begins. So can someone help me out and fill a couple of gaps. I'll mark the parts I need help with in red. Or please correct me if I'm wrong anywhere else in this text.
This is what I've gathered so far:
1. Dailies assembly (in case the lab doesn't do the entire work):
a) A print comes from the lab (without an academy leader?) and also 35mm mag transfer of the sound.
the print is cut into scenes, and the blade comes down at the point of the flash frames, where the camera starts (right?). An x would be marked at the point where the clapper hits, and some scene information written along the couple of frames before the clapper hits. Also, the part of the 35mm mag where the clapper is heard is also marked with a pen.
NOTE: I sense that there is an alternative workflow here, where you just leave the reel as it was when it got from the lab and proceed directly to syncing.
During dis-assembly of the lab print, no synchronization is done, just the sync marks are placed on both stocks for later. Right?
c) The assembly of the daily reel is started by making a sync point on the academy leader "picture start" frame, by punching a hole in it and also a hole on an arbitrary point on a blank piece of 35mm mag stock (this I'm not sure about) .... then I guess you'd take a piece of 1 KHz alignment tone and cut it into the 35mm mag after the number 2 in the picture academy leader?
d) The scenes prepared earlier are then cut-in after the synced leaders, each of them synced using the previously made synch marks on the film and mag stock, and by cutting out the extra length of the mag stock that you don't need (or perhaps adding blank mag stock in case the soundtrack is shorter, but it shouldn't be because the tape machine would roll first on the shoot).
e) tail leaders are added.
f) The dailies are viewed and if the sync is ok, they are edge-coded.
a) the daily reels are once again dis-assembled into scenes, just like when the prints got from the lab.
the editor makes a new synced leader at the start of each reel he is about to assemble.
c) When the editor uses a shot, he cuts away the slate part, leaving himself with no sync point other than the edge-codes. So at each step in the cutting process he cuts picture and sound in parallel and syncs it only using the edge-codes.
When watching a print of an "outtake" in bonus materials on DVDs and Blu-rays, I've noticed that each take has a punch hole and a alignment 1kHz sound at some point before the clapper hits(in some productions, in others not). At which point in the process would each take get a punch hole and a alignment tone? This is something I've missed in the above workflow obviously. And also, which frame is chosen for this? It obviously isn't the first frame of the take, because the punch hole appears later than the first flash frame. Is it any arbitrary frame between the start of the shot and the frame where the clapper hits?
As I remember it, the picture roll is synced with the sound on 35mm mag stock but without chopping up the picture roll, then once you have a sound roll of equal length, you get both the work print and the sound roll inked with new edge code numbers (not sure if those inked numbers have anything to do with the film's edge code numbers) -- once both rolls are inked with matching numbers, then you disassemble it into strips or small rolls. This way, once you begin editing and cut off the slates, you can still line up the sound with the picture.
The very original sound film workflow is that you have your sound recordings processed as well as the pictures. The lab will provide you with positives struck from picture and sound negatives. The Academy head and tail leader come into play with prints that are run on projectors where the sound reproducing unit is equipped with a heavy flywheel.
Else you have picture rushes or dailies accompanied by rerecordings of your original magnetic recordings (made at a specialised sound house or at the lab). First comes synchronisation after clapper slate but you won’t necessarily mark the clapper closing picture and sound stock level. It can be the very first frame of the take, its last or any other convenient one. You might want to retain the onsets for future whatever. No holes need to be punched. Such marks have to do with lacing up printers in the dark. The synch mark is always two seconds before the first frame of content, so frame minus 48. There is the 1000 Hertz beep, dead on level. The sound advance is 21 frames with 35mm release prints, the picture-sound displacement 20 frames. That one frame difference compensates the difference between speed of light and speed of sound in about 45 foot distance from the screen, the speakers presumed about a foot behind the screen. The original model of the average sound movie theater is 100 feet long, 90 feet hall, 10 feet booth.
Extra edge codes are also no necessity if the recording stocks were exposed heads first. Sometimes spooled-down film and short ends are exposed backwards which leads to factory edge numbers on the other side and subsequently not transferred to the dailies. Some stocks don’t bear foot numbers. Then one switches on the numbering machine.
Something else, perhaps perplexing: You don’t have to cut anything of the return material at all. It is perfectly possible to only mark both image and sound on the Moviola or flatbed editor and then pass everything on to an assistant who will do the physical breakdown. A well-rehearsed team can thus speed up work in the editing room. It’ll be a constant to and fro of material until it is in script order. You can even prepare a print order listing foot (edge) numbers under a given roll number. Section printing will be done at the lab, an effective procedure under certain circumstances, mostly with documentaries and for parts elaborated on the optical printer.
I’m sure you’ll see the difference between a 30-to-1 ratio wildlife documentary with a lot of unrelated sounds and a tightly produced industrial 15-minutes promo, both in 35mm. Polyester base film, by the way, can be ultrasonic welded for assembly once the editing is complete.
That's right. You don't cut picture rushes (what you call dailies) at all. There is, as you say, almost always more sound that picture so you cut out the extra sound on a multi-headed pic-sync.
Edge numbers are quite separate from the film edge code. They are of the format AB1234 so you can edge number 10,000' without repeating. You then log each roll of rushes by edge code so the neg cutter can find them when he conforms the neg to the workprint.
There's really no need to put leaders or tone on rushes if the sync marks are clear.
I think I made a mistake calling those codes "edge codes" because the name "edge code" is reserved for the codes printed on the film by the manufacturer. I just called them such because they are on the edge of the film. But I did mean those footage-count codes stamped with ink.
As for cutting the roll up into scenes before assembling it back again into a daily reel, here is where I got this idea.
In the book "Practical art of motion picture sound" by David Lewis Yewdall M.P.S.E., the author says the following:
Before you can actually sync the dailies, you must first break down all the footage,
which I now describe in detail. As you wind through the rolls of workprint, you
develop a rhythm for winding through the film very quickly. You watch the images
blur by, and every so often you notice several frames that “flash out”! These were
overexposed frames where the aperture of the camera was left open between takes
and washed out the image. These flash-out frames are at the heads of each take.
Pause when you come across these flash-out frames, pull slowly ahead, and with the
help of a magnifying loop, determine precisely which frame the slate sticks meet as
they are snapped together. That is where the clap is heard on the mag track.
Use a white grease pencil in marking the picture once you have identified the exact
frame where the slate marker has impacted. Mark an “X” on that frame of film where
the sticks meet and then write the scene number, angle, and take number (readable from
146 • Practical Art of Motion Picture Sound
the slate) across the frames just prior to the “X” frame. Find the scene and slate number
on the lab report form included with the roll of workprint, and mark a check next to it.
Roll the workprint quickly down to the next group of flash-out frames, where you
pause and repeat the process. After you have marked the slates on the entire roll of
workprint, take a flange (a hub with a stiff side to it, or, if you like, a reel with only one
side) and roll the workprint up. Watch for the flash-out frames, where you cut the workprint
clean with a straight-edge Rivas butt splicer. Place these rolls of picture on the back
rack of the film bench, taking time to organize the rolls in numerical order. You should
probably practice how to wind film onto a flange without using a core. For the beginner,
this is not as simple as you might think it is, but you will quickly master the technique.
After you have broken down all the picture, take a synchronizer with a magnetic
head and commence breaking down the sound transfer.....
You are now ready to start building your daily rolls. Professional picture editorial
departments always build the daily rolls starting with the smallest scene number and building
out to the largest. By doing this, the director and picture editor view an entire scene
and its coverage starting with the master angle; then they review the Apple (“A”) angle
takes, then the Baker (“B”) angle takes, then the Charley (“C”) angle takes, and so on. You
also review scenes in order; for instance, you would not screen 46 before you screen 29.
It helps to begin the process of continuity in theminds of the director and picture editor.
Obviously SOMEONE does it like that, regardless of the fact it seems the more common practice is NOT to brake the footage into scenes. Perhaps this is done when there are more cameras and the DP or the director wishes to see all the angles intercut from multiple rolls of film.
Maybe I missed it, but I still don't seem to have a clear answer as to why so many takes in uncut "bloopers" and "outtakes" start with a 1kHz blip.
I've watched a lot of outtakes, anywhere I could find them in my old DVD collection, and a large number of them has a alignment tone, wherever there is audio. Some have a combination of a punched hole and an alignment tome. Some have neither.
Could someone explain at which point do SEPARATE takes (not the start of the reel) get an alignment tone cut into them? Or is it perhaps something that a sync camera sends out into the tape recorder after it starts?
P.S. I've even found one rare case (from 70s) where the tone is saw, and not sine, which was odd.
I think I managed to find the answer about the alignment tone. It isn't cut into the soundtrack. It's generated by the tape machine itself when you start it. The beep sound heard in all of these "outtakes", is most commonly from a Nagra tape machine.
Perhaps breakdown is a US procedure. That description seems to be what you'd do for a rough cut or assembly edit.
What I've described seems to me to be a much quicker way of syncing up entire camera rolls. You'd mark the first picture slate, line it up with the corresponding clap on the mag, then run down to the second slate, mark it, mark the second clap on the mag, put that second clap on a spare sound head on the pic-sync without removing the mag from the sprocket, mark across both pieces of mag on the left of the sprocket using the guide channels as a reference point, then cut and join both pieces of mag, removing the spare mag, thus maintaining sync. And so on for the whole roll, the principle being always to cut on the left of the sprocket. The takes would only be broken down from the roll for editing.
I think the bloop tone was actually generated by the camera along with a flash of light in the gate which could be used to sync up. I don't know how widely it was ever used- a slate is more informative. But you're quite right, it's not cut in, it's generated in camera.
As far as I can gather from reading that chapter, and this is only my understanding of it, is that the point of breaking down the roll is to get the scenes and takes in numerical order, especially if there are multiple cameras, then it seems this way of doing it gives you a daily in which all the angles of a single take are exhausted before the reel moves on to the next scene and take. But it does seem to be a lot of work compared to the simpler way of just sincing it and projecting it.
As for the beep tone. I digged around some more in order to find more about it. This is what I've been able to find mentioned in a couple of books: there is a button on the Nagra which generates this tone, and the recordist would, in some cases press it two times to signify the end of the take, so at the end there would be two beeps. And as for the start of the take, you are right, the camera actually triggers it. I've found a description of this in an old Arri 35BL manual. The lamp would flash a couple of frames of film and one of the pins of the sync cable (in case the syncing was done by a cable and not by radio) would give a triggering signal to the Nagra's reference tone oscillator, so the beep would be automatic and synced with the flash frames at the start of the take. Otherwise the cable would transmit the 50/60 Hz continuous pilot tone. But the same interface was used for the start mark also.
I always thought that the flash frames were only due to the fact that it takes some time for the camera to achieve the proper speed, thus overexposing a couple of frames. And while this does seem to be the case, the flash seems to have been augmented with the lamp.
Did Panaflex (the older ones, gold, platinum...) use the same trick with the lamp?
Depends on the camera. I recall the Arri 16BL had no interlock, so the film could come to a halt with the shutter open. But the sync lamp would only light when it was at speed. I've just checked the only rushes and trims I have on the Steenbeck from a CP-16, which came up to speed instantly, stopped with the shutter closed and flashed only one frame at cut, and an Arri ST which did the same. Our CP-16 didn't have a sync lamp, IIRC the Arri did, and it lit a frame, but we didn't use it.
Correct an Arri16BL would not stop with the shutter closed....just stopped where ever it stopped.....so you could have a flash frame.
electronic slate.....when you had constant speed motor.....you connected a sync cable between the camera and Nagra.
this fed the pilotone (camera speed signal) which was recorded on the tape. When you threw the switch to start the camera a voltage pulse
flashed about 6 frames of film in the camera (a small bulb is mounted below the gate and can be removed and changed when needed) and the
voltage also fires a pulse turning on the tone generator in the Nagra. putting the tone on the tape.
once the camera is up to speed , the tone stops.
if you converted the camera to a Xtal motor.....no cable was needed between the camera and recorder....just the Xtal controlled motor in the camera
and a Xtal module in the Nagra. After installing the Xtal motor, you lost the electronic slate.
just a note here on the nagra.....as you look at the pilotone head.....you'll see it has 2 head gaps.....each one just off center.
the first head at the left is the erase head, then the record, the pilot, and the playback head.
in recording , the tape is erased, the audio is recorded and finally the pilotone......and of course you can throw the switch and listen to the audio in your headphones from the playback head while recording.
The pilotone is recorded over the audio.....one of the head gaps is connected out of phase with the other one.......any playback of the audio on a machine with a full track head will pick-up everything.....since the 2 tracks of the pilotone are out of phase......they cancel each other out......and all that you hear is the audio.
If you play the tape on a deck with 2 track stereo heads....or 1/4 track heads......you'll hear a buzz (the pilotone) in your audio.
I’d like to add a comment to that. Do not use grease pencils. You’ll mess up film, editor, and projector with little sticky specks.
Grease pencils or China markers as they’re sometimes called are out of date. Use modern felt markers and use them on the film base where you can wipe them off with a little isopropanol or white spirit.
Here, by the way, is a Carlos Rivas butt splicer for magnetic stock
We use to run one 1/4" reel to reel tape per camera roll. So each 1/4" reel had a camera roll number. When the film was processed, we'd throw white leader on the daily rolls and physically write the camera reel number on the outside for easy identification with a black magic marker. We'd then transfer the audio from 1/4" onto film mag stock and do the same procedure, throw white leader on it and label.
I always sync and edited on a bench editor. I hated moviola's, they weren't very smooth to operate and made a bunch of noise and when your trying to determine if you like a take or not, its important to hear the audio. So the steenbeck was where I always did my work.
Since the mag reel and camera reel were the same data as the roll of film, it was very easy to sync the two. On 16mm, we used grease pencils, on 35mm we used punches. The audio guy would always give us a tone at the head and tail of each take. This made it very easy to fast forward until you heard the tone and listen for the slate. You'd then lock the two in sync, make a mark, forward to the next take and simply cut the audio portion. On 16mm, we always ran one sided splices since we knew the reels were gong to be broken down anyway. On 35mm it's harder, we found it would jam in the rollers sometimes, so we tended to do double sided tape.
Once the reels were in sync, we'd watch them and take notes. We rarely used keycode numbers, we mostly used slate numbers and if they didn't work, we'd use a white piece of splice tape and make a note on it. We'd pull out the best takes only, label them with paper tape using scene number, shot, take and most importantly camera reel. We also marked on each camera reel log, which piece was cut out, so we knew where it came from on paper.
All the short clips were hung on a trim bin, all the longer cuts were wound onto independent cores, especially with 35mm. We did a paper cut of the film based on the script and organized those shots in the trim bin. Then we assembled those shots in order, most of which were master wide's or two shot's. Even though we hung the close up's, they usually hung until we were done with that scene and added them later once the flow was good. I always started at the head of the show and worked scene by scene on their own cores. Once the flow was good, we'd add in the pre-cut close up's and do some frame tweaking to get little things right. With film editing, you need to visualize everything your about to do, before you do it.
We then had a bunch of cores with individual scenes, mostly all with a piece of paper tape holding them together with a number on them, so we knew where they sat within the script. Once all the scenes were cut, we'd do what's considered a "master assembly" of the show. When working on TV, it was broken down into acts. So sometimes 12 min, sometimes 8 minutes, depends on the show and timed length between commercials. On features, it was always broken down into 20 minute reels.
Once master assembly is done, we'd project it for the producers. We had one of those steenbeck projectors, it was awesome and allowed more then a few people to watch the film before signing off. This process was the most tricky because sometimes we'd have to go back to the dailies and grab some missing material. Some of our dailies were very rough, lots of splices on top of splices due to fine tuning.
Once approved, the negative was sent to the negative cutter. Shots that required transitions were sent to the lab to do the photochemical work. The final negative was conformed and sent to the lab for color correction. Audio was also sent to the lab on 6 track magnetic stock and then the optical audio print was made. When done, we'd get a print with mated audio and ready for projection.