Is stripping obligatory before assemble edit to DigiBeta?
Hi, I'm audio oriented and at the company where I'm working my job is audio post production for video programme.
Recently my colleague (he is good at video editting, but new to DigiBeta delivery) and I run into a problem. We'll deliver our programme to a broadcaster by DigiBeta Tape. And time code is required.
We'll rent a Sony DVW2000 for tape recording in a few days. That'll be the first time for us to do assemble edit to DigiBeta. I have searched the net for info. All said that even a blank tape should be striped with timecode and black frames first. And all infos are about how to stripe, none is about why should do like that. It's the behind-scene theory that I can't understand clearly.
Since the tape recorder can record the video, audio , timecode (DVW2000 can record timecode while syncing to external time code) simultaneously, why can't we start assemble edit to tape without stripping first? If for the reason of pre-roll or enough time for the recorder to catch up, we can just add extra black frames at the beginning and choose an earlier time.
If say without stripping, FCP does not know exactly at what time and where on the tape to start a assemble edit. But look it from the other way, while stripping the whole tape with time code ( someone even prefers to stripe the whole tape, right?) , why not record the video and audio at the same time? The recorder can sync to external time code even while stripping, if I don't make it wrong according to the operation manual.
BTW, we have Blackmagic interface and FCP Pro. And there's no need to do any edit to tape operation later.
Anyone can solve our puzzle? Thanks in advance.
Assemble editing records video, audio, control track and timecode all at once, so if you are assemble editing there's no point in striping the tape in it's entirety. Striping the tape allows you to insert edit one or more tracks at a time rather than everything at once.
However, whenever you are editing onto tape, the tape deck has to lock up it's playback before dropping into record (i.e., be at the correct speed and sychronous with the reference signal), so you do actually need a few seconds of recording at the front of the tape in order to perform an assemble edit. So, switch the tc generator to preset and set a time a few seconds before you want to start your recording and record until you're past that point, then stop the tape, switch the tc generator to regen (or whatever it's called on that particular deck) and away you go.
TBH, it's quicker to do it than it is to type it here..
Thanks for the reply.
However,if a few seconds of recording at the front of the tape are necessary for the tape deck to lock up it's playback before dropping into record, but after that there's no control track data or time code on the tape, the deck has no reference to adjust its motor speed, it can record all kinds of tracks on the fly, right? If so, why the stripping is needed?
Seems complicated. Or where can we find info about how those DigiBeta decks work?
When the playback is synchronised to a reference signal the motors in the deck are being driven to run the tape at precisely the same speed as it was running when the initial recording was made and timing it so that at any instant a defined part to the tape is passing the record/play heads.
So the precise speed of the tape going through the mechanism makes no difference on the first recording, as long as it's within the system boundaries. An assemble edit is basically a first recording that continues a first recording that is already there.
When you're assemble editing, the thing that is vitally important is that at the instant that the deck starts to record, the deck is moving the tape normally (not speeding up from a standstill) and the timecode on the tape is at a number which matches the editing system output (so that the timecode on the tape is continuous and at the numbers you are required to have). And the deck can only do that if it has something recorded on the tape already.
You are right that after the initial bit of recording there isn't control track or timecode on the tape, but the deck is creating those as it goes along. You only need them to be there already if you want to do an insert edit, i.e., adding individual tracks or combinations of tracks to an existing recording, in order to match a new bit of recording into an existing one.
Older guys (including me, I'm in my 40s) will remember switching between "auto" and "tach" and adjusting the deck to make sure that the track recorded during an edit matched the track that was there before, but you don't need to think about that kind of thing on modern decks.
I don't know what to suggest as learning material. My main reference is my BBC training notes (which aren't published). I've always found Sony's documentation ok if you're already well informed and just need specific information for this piece of equipment, but a bit heavy going and confusing if your coming from nowhere. Focal Press books are mostly very good but I don't have a particular one that I could recommend. It's probably worth asking in the Decks and VTRs forum and maybe the Broadcasting one on the Cow here for recommendations.
Striping (not stripping-there is no pole or dollar bills involved) lays down continuous control track (CT) and timecode(TC). This procedure is also called blacking. This is almost always done with a minute of bars, followed by switching to black until the tape runs out.
This is so you can do insert editing. Insert edits are either A or V and the CT and TC are not altered. Insert edits are clean on both ends, and can lay any combination of A1/A2//V tracks.
When you assemble, the deck has to preroll (back up) onto some existing CT and TC. If none exists, you have to crash record some onto the tape. Then it rolls forward and syncs to that existing material, and continues to the edit point and starts laying down fresh CT, TC plus A and V. It is the same as a crash recording, but the in-point is clean and controlled, but the out point is not and is dirty. You can not select what tracks to use on an assemble, it's all, or nothing. So you can not do a selective edit in the middle of exiting materiel using an assemble edit, since it is an over-record that breaks the control track and TC on the out. Color frame accuracy is also an issue on assemble edits, and it is shear luck if you try to do a match frame edit using assemble, and the edit doesn't shift.
[Shawn Hyper] "the tape, the deck has no reference to adjust its motor speed,"
The record (or playback) deck is not free running as it may appear.
The simplest explanation is:
There are electro-mechanical circuits called the drum tach and the capstan tach in tape machines. They are part of a servo-feedback loop that monitors the video head drum rotation speed and tape path speed in comparison to the genlock or house master black reference that is input to the recording machine. This (genlock) is also the same signal that keeps the H and SC phase in sync with other machines, cameras, the TC generator and the switcher so you can do wipes and dissolves etc. There is also the sync pulse of the recording signal itself. They (the tach signals) lock to the control track of the previous recording, and then provide many fast kicks to the drum and capstan motors so that they continue run at a speed that will accurately lay down the recording signal. There are also some machines that can keep the drum and tape path speed in spec without any external reference besides the input recording signal because they have on board high quality time base references, but usually only on a crash recording, or recording from something stable like a camera. Even then, it is better to have a master external reference, and almost mandatory when doing edits on tape with any degree of accuracy. Start/stop edits generated by camera are usually low quality, and you have to roll off some tape to get up to 'speed' before shooting anything important.
Despite all the work that goes into this, it is never perfect due to tape stretch, sticktion (where the moving tape actually sticks to part of the tape path for a micro second) and mechanical wear on parts, so there is a device on the output of tape machines called the time base corrector or TBC. The TBC is essentially a type of buffer that cleans up any minor mechanical generated timing issues, and outputting a signal that is clean and stable like what comes out of something entirely electronic like the Chyron, or camera.
"If you think it's expensive to hire a professional to do the job, wait until you hire an amateur." ---Red Adair
There is also a semi-superstition that "re-packing" the tape by fast forward/rewinding it from end to end helps with pre-conditioning the cassette for mastering, but modern manufacturing methods make this a dubious practice. You can actually induce bridges and skewed winding if the machine itself isn't perfectly aligned. Not saying a casual rental machine is a bad choice... I'm pretty sure nobody has dropped it lately...
Assemble editing used to be a nightmare for keeping the recorded tracks phased and synchronized properly, but today's digital decks are more like data recorders than old style flying-head analog VTRs.
The operating system remains identical, though, and there still is a need for continuous Control Track, Time Code, and a well-aligned mechanical tape path so there will be no issues with machine interchange. The easiest way to achieve that was to lay down a continuous recording, the "pre-stripe" to essentially plow the road for the actual program material to be inserted later. It really does lay all the ground work for revision inserts, technical corrections, audio laybacks, and all the rest of the insertion process.... and really helps guarantee that the tape will play back elsewhere.
However, laying down a minute of black that establishes the starting time code and then assembling on the program is not significantly different anymore from inserting the material on a continuously pre-recorded tape.
Saying an "out" on an assemble edit is "dirty" doesn't completely cover what is really going on. During assemble editing, new control track and time code is being generated -- and even if there is existing control track and TC on the tape, it will be erased as the tape progresses. This makes "stripping" the tape prior to an "assemble" even more redundant.
So be extremely careful when selecting "Assemble/Insert" with your edit controller, because coming "out" of an assemble leaves a gap with no CT or TC. It isn't fixable. But everybody makes this error at least once, or you can't call yourself a seasoned professional tape operator.
"I always pass on free advice -- its never of any use to me" Oscar Wilde.
Thanks to all of you for patient explanation.
But excuse me for my verbosity。
[Andrew Rendell ] However, whenever you are editing onto tape, the tape deck has to lock up it's playback before dropping into record (i.e., be at the correct speed and sychronous with the reference signal), so you do actually need a few seconds of recording at the front of the tape in order to perform an assemble edit.
we can add extra black frames before whatever shoud be recorded. This will give the deck enough time to lock up. Can it work?
[Scott Sheriff ]This procedure is also called blacking....This is so you can do insert editing.
If we're sure there's no need for insert editing, can we just crach record? And the TC, CT , audio and video will be in sync with each other?
And if we crach record on a blank tape from the beginning to the end, can we say that we have blacked the whole tape in a special way? since there's continous TC and CT on the tape.
[Joseph Owens]The easiest way to achieve that was to lay down a continuous recording, the "pre-stripe" to essentially plow the road for the actual program material to be inserted later. It really does lay all the ground work for revision inserts, technical corrections, audio laybacks, and all the rest of the insertion process.
Why not record the audio and video while plow the road? Yes, I see the need for the "pre-stripe" , it's necessary for later insert edits, etc. But for assembling edit, I can't see it. If say, assembling edit can't be started from nowhere (no reference), the "pre-stripe" is from nowhere. Why not record AV while striping? The deck can't do like that or something else?
Sorry for I'm entangled here. Thanks in advance for any help.