Copy link to clipboard
Copied
Hi,
I'm working on a temporal filter plug-in for Premiere.
I noticed that, when I attach the Effect, the Render() function is called twice: once at half of the resolution and a second time at full resolution. As it's a temporal filter, some data structures are allocated at the first frame and then reused at the following frames.
What is the purpose of this mechanism?
Is there a way to disable that to avoid having to manage two different data structures?
Thx,
Xavier
>What is the purpose of this mechanism
For some reason, PPro wants both renditions, at that time; the effect must be prepared to handle such requests.
One simple case = during playback, the last frame might get requested at half-res (to preserve playback timing), and get requested at full-res at the end of playback, because the CTI is 'parked' on that frame.
For temporal filters, the time-based calculations are usually begun during PF_Cmd_SEQUENCE_SETUP; here's some simulation-related documentation
Copy link to clipboard
Copied
>What is the purpose of this mechanism
For some reason, PPro wants both renditions, at that time; the effect must be prepared to handle such requests.
One simple case = during playback, the last frame might get requested at half-res (to preserve playback timing), and get requested at full-res at the end of playback, because the CTI is 'parked' on that frame.
For temporal filters, the time-based calculations are usually begun during PF_Cmd_SEQUENCE_SETUP; here's some simulation-related documentation, from the AE SDK docs.