Hi, What is the policy for handling _speed == 0.0 on frames incoming to consumers from producers? Is that anyhow different than receiving frame with _speed == 1.0 from the consumer’s perspective?
I think I’m doing it wrong, because my consumer uses same amount of CPU time whether _speed == 0 or not, and I noticed in „melt + sdl consumer” setup there is a noticeable decrease in CPU usage when I pause. In some apps, like Shotcut or kdenlive which use qglsl consumer (I think) CPU usage drops down to 0% on pause! The thing is, I do nothing special when „_speed == 0.0”. So, when _speed of incoming frame is 0.0, i just do get_image() and get_audio() on it, so it goes through usual resize+convert pipeline as if I was fetching a frame in play state (_speed == 1.0). So the CPU usage of my consumer is the same whether it is playing or in pause state. Obviously this is wrong. I was trying to detect in code what is being done, but it seems pause isn’t treated like special case in neither sdl consumer nor qglsl consumer when it comes to fetching image/audio from received frames. So there is something smart going on there, just under my nose, but I can’t detect what :( Best, Paweł ------------------------------------------------------------------------------ What NetFlow Analyzer can do for you? Monitors network bandwidth and traffic patterns at an interface-level. Reveals which users, apps, and protocols are consuming the most bandwidth. Provides multi-vendor support for NetFlow, J-Flow, sFlow and other flows. Make informed decisions using capacity planning reports.http://sdm.link/zohodev2dev _______________________________________________ Mlt-devel mailing list Mlt-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/mlt-devel
