Alot of people are oblivious to the importance of timecode on large scale productions. Editing software now has become pretty powerful, and so editors often work from the highres video sequence, which hides a few timecode/referencing steps from them. Animated films complicate this process due to the fact that each shot is created one by one with their own start frames (very much like editing footage shot on a DSLR), which generally means you can have upward of 1200 movie files/clips/image sequences or ‘tapes’ all with their own timecodes, etc.
Timecode is effectively a piece of Metadata stored in the header of your frame sequence or video file. It is used all the time by editing and grading software to match or sync video files. It is also specific to weather or not you are working at 24, 25, 29.97 or 30fps, etc. It is broken into HH:MM:SS:FF where FF is frame number. So as FF reachs your fps it will roll over to SS. So at 25fps, 00:00:00:24, next frame would be 00:00:01:00.
Editors may vary, some edit from 01:00:00:00 vs the UK TV editors who often work from 10:00:00:00. Some in feature films edit Reel 1 from 01:00:00:00, Reel 2 from 02:00:00:00, etc etc. If you are an avid user the videos imported are most likely converted to an internal format (often DNxHD) and the start frame is set to 01:00:00:00 (unless you have specified differently). Default movs created from FFMPeg are more than likely going to be 00:00:00:00, which means if you import that into Avid your edited file and your orginal source file would now be 1hr out. This discprency in the timecode, that you may not have noticed has changed, could have knock on effects futher down the pipeline when you plan to re-link to the highres footage or if you intend to switch your footage out with a .dpx sequence with its own timecode.
Right at the beginning of production the chances are that your editor is working & editing with animation captures, which generally do not have any timecode (except for any pushed into the file for you by your editing software). You should at least have a frame counter burnt into the frame in some place. This will work fine for the early parts of production, until the point you start editing with files based on final rendered dpx files. Now is the time to set up your project correctly, choose your starting timecode and lock that for the rest of production. Nuke users can place the ‘AddTimeCode’ node into thier B-line, and also add ‘[timecode]’ into the timecode slot of your Write Node. The timecode is now written into the DPX header.
The next step is to make sure the timecode remains correct between DPX image sequences and the video file (.mov, etc). So if your start timecode in the DPX is 01:00:00:00 your mov should match that, regardless of frame number. This can either be done by adding the timecode flag into FFMPEG or making sure your edit software ingests the mov and adds the matching timecode.
Software like djv can be setup to show timecode and often appends the HH:MM:SS:FF with @25, or @24 to show the frame rate.
You Forgot about Timecode?
Not all is lost if you have completely forgotten about timecode and have nearly finished your film. Lets say your DPX sequence has no timecode or the incorrect timecode, your best plan of action would be to find out from your editor what timecode he has been editing with and then using software like Davinci Resolve, push the correct time code into the header of each DPX sequence. Now when you begin your conform in Resolve, Speedgrade or the like, the EDL/XML timecodes from the edit should accurately match that of your sequences and thus you should have a painless conform.
Without timecode your conform could take days, spent slipping each shot carefully to sync up to the reference video. If shots where animated and rendered with handles those extra frames could become the horror of many a conform or assistant editor.