They can often reduce video processing tasks quite a bit.Ī further improvement can be found in purpose built hardware. GPUs, on the other hand, are designed for doing simple operations very quickly. They are designed for doing a wide variety of general purpose operations, but aren't super efficient at basic operations. General purpose CPUs aren't really ideal for many audio/video processing tasks. There are two main things that can improve the performance. There are some things that may need to be tracked over time, but for the most part, it is a stream operation with data going out as fast as it comes in, so there isn't much accumulation of data. Similarly, RAM wouldn't be expected to cap on encoding as it is a stream operation. If this is the case, there isn't an easy way to break it down across multiple threads and that would make it impossible to hit 100% CPU usage on a multi-core computer.
The problem is most likely that some of your effects require the result of the previous frame to begin processing on the next frame. By default, Premiere has always done as much parallel processing as possible for me and I frequently see it hit 99% CPU usage on my hyper-threaded quad core desktop.
While the video encoder itself is able to do multi-threaded processing, the image processing you are doing may not be able to. Based on your description of the problem, it sounds like the processing is the slow part.
