ergoCubEmotions: avoid reading videos at each update of the module #214
Conversation
|
I modified the code to open only the transitions at runtime, while the "static" video are opened in the configuration and then read in the updateModule when the correspondent command is set. In this way, only 4 videos are saved in memory, instead of 16 (videos + transitions). This is the result tested on my laptop: Screencast.from.01-23-2024.03.14.16.PM.webmI'm going to test it on ergoCub. |
|
I did the test on the ergocub-head on bench and this is what I obtaned: ergocubemotions.mp4@GiulioRomualdi, @S-Dafarra, @traversaro could you please review? |
| #[expression_3] | ||
| #name shy | ||
| #type image | ||
| #file expressions/images/exp_img_3.png |
There was a problem hiding this comment.
All these commented lines are not used? can we get rid of them?
There was a problem hiding this comment.
I kept them to have an example of how to handle images instead of videos. So far, we haven't used them, but if we need, we will just uncomment them. If you think that they're not necessary, I'll remove those lines
This should fix #213. I tried to launch the module after these changes on my laptop and the cpu consumption significantly decreased.
To be tested on
ergocub-head.