How to stop your lower thirds and subtitles fighting for the same space
Ask Vizard Agent to assign each bottom-of-frame element its own band and its own time window. Shrinking the subtitles to clear a name strap works until the ticker appears, and then something else has to shrink — the durable fix is that two things never occupy the same strip at the same moment.
What is the short version?
A news-style package puts four things in the lower third of the picture: a name strap, a ticker, a source credit and the subtitles. Each looks fine alone. Together they overlap, and fixing one moves the problem. Vizard Agent plans the space instead.
- Tell Vizard Agent every element that lives at the bottom.
- Ask for each to get a band and a time window.
- Check it on rendered frames, not on the layout plan.
What do you need before you start?
A list of the elements. Subtitles, name strap, location label, ticker, source attribution, logo — whichever of those your package uses. The list is the whole design problem, and it is usually longer than people expect because the source credit and the logo get forgotten.
Say which ones can move and which cannot. A broadcaster's ticker position may be fixed by a style guide while the subtitles are free, and knowing that up front means Vizard Agent solves the problem in the direction that is actually available.
What do you type into Vizard Agent?
Describe the layout rather than the symptom you noticed. "The subtitles collide with the name strap" gets that one collision fixed and leaves the next one waiting; describing the whole bottom of the frame, and which elements are fixed, gets all of them solved at once.
The bottom of the frame has the ticker, the lower-third name strap and burnt-in subtitles. Keep them out of each other's way — the ticker position is fixed, the rest can move.
That is the version worth typing. The sessions behind this article went the other way, one collision at a time, and each fix created the next one.
What does Vizard Agent actually do?
It separates them in space and in time. A lower third that only appears during the presenter segments does not need to share the frame with anything for most of the runtime, and subtitles that sit in their own band above the ticker never reach it.
In those sessions the sequence of corrections ran:
- Shrank the subtitles so they stopped colliding with the lower third.
- Adjusted the subtitle scale again for the landscape frame, because the first reduction went too far.
- Moved the lower third so it no longer met the subtitles, and confirmed the two were separated at the opening.
- Restricted the lower third to appearing only during the presenter segments.
- Lowered and lightened the subtitles again so they cleared a portrait on screen.
- Fixed the ticker's safe area and checked the subtitles were not covering the news card behind it.
Six corrections, alternating between two elements. The lesson is in the fourth one: giving the lower third a time window rather than a smaller size is what actually ended the loop.
What does the result look like?
A bottom strip that reads as designed rather than as negotiated. Each element has room, nothing is smaller than it should be for legibility, and the subtitles are still readable at phone size because they never had to shrink to make space.
Vizard Agent checks this on rendered frames at the moments each element is on screen, not on the layout values. A strap and a subtitle line can be in different bands on paper and overlap in the render because the subtitle wrapped to two lines.
When does this not work well?
When there is genuinely not enough room. A vertical frame with a ticker, a strap and two-line subtitles has run out of bottom, and something has to give — fewer words per subtitle line, a shorter strap, or the ticker moved to the top. Vizard Agent will say which trade is cheapest rather than quietly shrinking type.
It also gets harder when an element is burnt into the source. A ticker already in the footage cannot be moved, only worked around, and that constrains everything else.
And a style guide that fixes two elements to the same band is a contradiction rather than a layout. That needs a decision from whoever owns the guide.
How do you fix a result that came back wrong?
Say which two elements collided and at what moment. "At the opening" and "during the interview" are different windows with different elements present in each, and Vizard Agent fixes the window you name rather than adjusting the layout everywhere.
If the subtitles have become too small to read, say so and let something else move. That is the failure mode of solving every collision by shrinking the same element, and it produces a package that technically has no overlaps and cannot be read on a phone.
If the strap appears where it should not, give it a time window instead of a position. Most lower thirds only need to be on screen for a few seconds per segment, and restricting them removes the conflict entirely.
How does Vizard Agent compare to doing it yourself?
In an editor you would see the collision immediately and nudge something, which is exactly what makes this loop so easy to fall into. Each nudge is correct and local, and after six of them the subtitles are half the size they should be.
The difference is treating the bottom of the frame as one layout with a time dimension. Vizard Agent checks every element at every moment it appears rather than at the frame you happen to be parked on, which is what catches the collision that only happens during the segment you were not looking at.
Common questions
How many elements is too many? Four at once is usually one too many in a vertical frame. Vizard Agent will say when.
Can the ticker go at the top? Yes, if your style allows it. That frees the most space of any single change.
Will the subtitles get smaller? Only as a last resort. Vizard Agent moves things before it shrinks them.
What about a two-line subtitle? It is planned for. Vizard Agent checks the wrapped height rather than one line.
Can the strap show only sometimes? Yes, and it usually should. Vizard Agent gives it a time window per segment.
Does this work in vertical? Yes, and it matters more there because the frame is narrower.
What if the ticker is burnt into my source? Vizard Agent works around it and will tell you what that costs.
Can I give exact coordinates? Yes. Vizard Agent uses them and reports any conflict rather than overriding you.
How does it check? On rendered frames at the moments each element is visible.
What about the platform's own interface? Say which platform. Vizard Agent keeps the strip clear of the app's controls.
Can it show me the layout first? Yes. Ask for a storyboard frame before the full render.
Does the source credit count? Yes, and it is the one most often forgotten.
Will the logo be in the way? Not if you list it. Vizard Agent treats it as another element.
What if I change the subtitle style later? Tell Vizard Agent and it re-checks the whole strip rather than only the subtitles, since a heavier face changes the height too.
Can the subtitles sit above the picture instead? In a padded vertical layout, yes. Vizard Agent can put them in the header band if there is one.