At a glance: Broadcast convention puts bars and tone at 00:58:30:00, the two pop at 00:59:58:00, and the first frame of programme at 01:00:00:00. The two pop is a single frame of 1 kHz tone, defined as 48 frames before first picture rather than two wall clock seconds, which is why it survives frame rate conversion. The film leader is a different convention: pop at 01:00:06:00, first frame of action at 01:00:08:00. Programmes start at hour one for tape era reasons that no longer apply and for one that does, which is that timecode cannot count backwards through midnight. Drop frame exists only at 29.97 and 59.94, where counting and clock time drift apart by 3.6 seconds an hour. None of these conventions is wrong; deliveries fail when two of them meet unannounced.
An AAF and its reference video land on the audio side, and the first thing anyone does is check that the two agree about where things are. That check depends entirely on conventions nobody puts in the file: where the programme starts, where the sync mark sits, and how the frames are being counted.
Each convention has a reason, most of the reasons are historical, and all of them are still enforced by delivery specs. This is the reference for what each one is, where it comes from, and what actually goes wrong when they get mixed.
Why Programmes Start at 01:00:00:00
Three reasons, from three eras, and knowing them is what lets you argue sensibly about the convention instead of just obeying it.
The reel number. In film post the hour digit was the reel: reel one started at 01:00:00:00, reel two at 02:00:00:00, and a timecode told you which reel you were holding. Production Expert's history of why timecode does not start at zero covers this, and notes that tape era television often started at 10:00:00:00 instead, precisely to look different from film. Checked August 2026.
Tape needed a run up. A broadcast tape carried a minute of bars and tone, a slate and a countdown before the programme, and machines needed seconds to reach speed. Starting the programme at hour one left room for all of it, with the pre roll living from 00:58:30:00.
Timecode cannot cross midnight. The reason that still applies. If the programme started at 00:00:00:00, everything before it, bars, slate, countdown, would sit in the 23:5x range, and a counter reading 23:58:30:00 followed by 00:00:00:00 has just gone backwards as far as any machine sorting by timecode is concerned. Larry Jordan's piece on the hour one convention makes the point that modern NLEs default to zero because servers need no calibration material, which is true, and it is also why the convention confuses people: your NLE does not care, but the delivery spec still does, because the leader still has to live somewhere before first frame. Checked August 2026.
So the working rule is unchanged: programme at 01:00:00:00, everything ceremonial before it, and if a spec says otherwise it will say so explicitly.
The Leader, Position by Position
Two conventions are in active professional use, and they disagree about where the pop goes. Both are correct in their own world, which is exactly why the delivery note has to say which one you used.
| Position | Broadcast leader | Film leader |
|---|---|---|
| Bars and tone begin | 00:58:30:00 | Leader begins 01:00:00:00 |
| Two pop | 00:59:58:00 | 01:00:06:00 |
| First frame of programme | 01:00:00:00 | 01:00:08:00 |
The broadcast positions are the ones CalArts' standard leader reference teaches and Wikipedia's 2-pop entry documents, both checked August 2026. The film variant comes from the SMPTE Universal Leader, where the countdown itself starts at 01:00:00:00 and first action lands eight seconds later, at twelve feet on a film footage counter. If you have only ever worked in broadcast or streaming, the second column is why a file from a features pipeline seems to have its pop in the wrong place. It does not; it is speaking the other convention.
FFOA, first frame of action, is the delivery spec term for where the programme actually begins, as distinct from where the leader begins. On a broadcast leader FFOA is 01:00:00:00. On a film leader it is 01:00:08:00. When a spec asks you to state FFOA, this is all it is asking.
The Two Pop Itself, and the 48 Frames Question
The pop is one frame of 1 kHz tone, paired with one frame of picture, the 2 of the countdown, two seconds before first frame. Henri Rapp's explainer gives the customary level as minus 20, standard though not required, checked August 2026. After that single frame the leader goes black and silent until the programme starts, which is the point: one unmistakable event, a known distance from first frame, on both picture and sound.
Two seconds or 48 frames is a real argument among working mixers, visible at length in a Steinberg forum thread on exactly this question, checked August 2026, and the defensible answer is 48 frames. At 24 fps they are the same thing. At 23.976 the 48 frames last fractionally longer than two wall clock seconds, and through a frame rate conversion a frame defined distance survives where a wall clock one drifts. Think in frames and the pop stays put.
What the audio side does with it: on import, the pop is the sync verification. One frame of tone against one frame of picture, at a known timecode, confirms the whole timeline landed where it claims. The tail pop, the same single frame placed two seconds after the last frame of programme, does the same job at the far end, and the pair together answer the question that matters on long form: not just did it start in sync, but did it stay there.
Drop Frame, and When You Actually Need It
Drop frame exists because 29.97 fps is not 30. At 29.97, counting 30 frames per labelled second means the label falls behind the clock by exactly 3.6 seconds every hour, so an hour long programme reads 59 minutes and change at its end. Drop frame timecode fixes the label by skipping frame numbers 00 and 01 at the start of every minute except the tens, 108 skipped labels an hour, no frames of picture harmed. The notation is the tell: a semicolon before the frames means drop frame, a colon means non drop.
| Question | Drop frame | Non drop frame |
|---|---|---|
| Notation | 01:00:00;00 | 01:00:00:00 |
| What it preserves | Clock accurate running time | A continuous frame count |
| Where it is used | Broadcast and streaming delivery at 29.97 and 59.94 | Post production work, and everything at 23.976, 24, 25 and 50 |
The practical version for a handoff: drop frame is a delivery format question at 29.97, not a post production preference, and there is no drop frame at 23.976, 24 or 25 at all. What matters at the handoff is only that the AAF, the reference video and the session agree, because a mismatch shows up as a constant, maddening offset that looks like a sync error and is actually a counting error.
Where This Goes Wrong: The Mixed Batch
One file with a stated convention never causes trouble. The damage comes from batches. Commercial campaigns routinely arrive as ten to fifteen deliverables, and it is entirely normal for some to carry a two pop, some a slate, and some to start cold at first frame of action, inside one delivery. Each spot then has to land at a deliberate timeline position, and untangling which convention each one speaks is diagnostic work nobody budgeted for. The batch mechanics are covered in our guide to the multi AAF workflow, and the same discipline pays off again when the cut changes and everything has to be conformed against the new picture.
The fix costs one line in the delivery note: the start timecode, the convention, and whether there is a tail pop. Which convention you use matters far less than saying so, and the full delivery specification is four questions long. The same week's companion piece on audio handles covers the other setting that travels with it.
What the Import Side Does With All This
On the Pro Tools side, the conventions above stop being etiquette and become mechanics. Every clip in an AAF carries its source timecode, and the session either respects it or the whole delivery is offset. fPost preserves that layer through its import: clips are placed against the source timecode the AAF declares, and clips, fades, automation and timecode land intact through the reorganisation, with Import Session Data preserved. For the session whose programme start genuinely differs from the AAF, the classic case being a session built at 00:59:00:00 receiving an AAF that starts at 01:00:00:00, a manual timecode offset shifts the import as a whole rather than clip by clip.
What fPost does not do is decide the convention. Whether the pop is where the delivery note says it is, and whether the note says anything at all, stays a conversation between the picture editor and the mixer, and no tool substitutes for stating it.
Frequently Asked Questions
Why do sequences start at 01:00:00:00?
Historically because the hour digit was the film reel number and because tape machines needed a leader and a run up before the programme. The reason that still applies: timecode cannot count backwards through midnight, so starting at hour one leaves room for bars, slate and countdown before first frame without the counter wrapping. Modern NLEs default to zero, but delivery specs still expect hour one.
What is a two pop?
A single frame of 1 kHz tone, paired with a single frame of picture, placed a fixed distance before the first frame of programme. It exists so the audio side can verify sync at import: one unmistakable event at a known timecode on both picture and sound.
Where exactly does the two pop go?
Broadcast convention: 00:59:58:00, with the programme at 01:00:00:00. Film leader convention: 01:00:06:00, with first frame of action at 01:00:08:00. Both are in active use, which is why the delivery note should say which one the file speaks.
Is the two pop two seconds or 48 frames before the programme?
Define it as 48 frames. At 24 fps they are identical, and a frame defined distance survives frame rate conversion where a wall clock distance drifts. Working mixers argue this one both ways, and the frames answer is the one that keeps the pop in place across formats.
What is FFOA?
First frame of action: the timecode where the programme itself begins, as distinct from where the leader begins. It is 01:00:00:00 on a broadcast leader and 01:00:08:00 on a film leader, and it is the value delivery specs ask you to state.
Should I deliver drop frame or non drop frame?
Whatever the delivery spec says, and the question only exists at 29.97 and 59.94, where counting and clock time drift apart by 3.6 seconds an hour. There is no drop frame at 23.976, 24 or 25. At the handoff the only requirement is that the AAF, the reference video and the session agree.
What is a tail pop for?
The same single frame of tone placed two seconds after the last frame of programme. Together with the head pop it verifies the programme did not drift over its length, which matters on long form where a start that syncs proves nothing about the end.
fPost turns an AAF into an organised Pro Tools session with clip timecode, automation and session data preserved through the import. Details at forte-ai.com/fpost.
