End HTML template completely

Hi all,
Regarding the ‘finished state’ of HTML templates.

Is it enough to send this command?

Or should we also clean the layer completely?

I’ve noticed that even with the window.remove(), the HTML template stays “active” in a way. Here’s the diag window with a couple of layers that have been ‘closed’ a while ago. They just stay like that unless you clear the layer / channel.

Do we know if this is something that could affect performance or have unwanted results?


I would not remove the layer, if you use it again later as it affects performance more, if you always create a new layer and load chromium again.

I usually use only 2 to 3 layers and just reuse them. One for a bug and one for all lower thirds, for instance.

1 Like

I see.

I had the idea that cleaning the layer was a good idea from the flash era.
We encountered several unwanted behaviours and choppy playbacks with long .png sequences, and we managed to fix it by clearing the layer before playing the next flash template.

I’m glad to see it’s not the case with HTML, but the fact that you want to do the exact opposite is a bit mind blowing :rofl:


Did you call removeTemplate after the out animation? That should clean up stuff.

In HTML it‘s good practice to do the same by calling window.remove after everything has animating out.

We had to check old templates, but yeah we used the removeTemplate.
I think the issue was probably the png sequences were sometimes massive, so I suspect some kind of bottleneck when loading a new batch without clearing the layer completely.

Do you load full screen PNG sequences? I usually crop them and use only the part where something is visible. You then put it into a MovieClip…

In HTML for webM videos I use the same trick whenever possible.

Yeah at some point we realized we needed to crop the sequences to improve performance, but even then, having IN + Loop + OUT , sometimes it was finicky…

For webM I feel the performance constrain is way less, so we haven’t tried to crop the assets yet, but it’s good to know it’s also a thing. Only doubt is that repositioning in HTML feels more tricky than it was with flash…
Do you use a creation software with a GUI or do you just reposition them in the HTML code?

I am currently developing a software called “BluePrint” that is able to generate templates from AfterEffects exports coming from the BodyMovin plugin. Hope to bring it out in the next few months.

To position a non full screen WebM BluePrint uses a div that is positioned absolute. That is quite simple. The coordinates are in pixels. But yes, currently the positions need to be keyed in, so no GUI for that,