Hi Mitch!
When enable Background processing the files are first stored in the avi format. When the workload of the CPU is lower the ffmpeg process starts to convert avi to mp4 format. When that is done the avi file is deleted.
The idea for the Background processing was to reduce a high cpu load and let the avi to mp4 conversion take place when the CPU load is lower. However, what happens is that the cpu load is still high, but the background processing introduces a delay. Since NCS displays only the mp4 format the background process introduces a delay from the time the motion is detected and the file is stored in avi to when the file is converted to mp4 and can be displayed. I have a system with a cpu load of 90%. Enabling background processing introduced a delay of 10 minutes.
When you have a low CPU load and enabled background processing it sometimes increased the cpu load instead of lowering it and also introduced this delay. The simultaneous process of encoding from avi to mp4 is so efficient so the idea was not a hit ;). We have discussed to remove this option and we will do so. So my recommendation is to not use the Background processing since as you point out it increase the use of space on the HD and yes, that extra space used for avi files are not included in the Maximum Library size! (Thanks for pointing that out).
To reduce the cpu load with the same hardware is always of interest since one can increase the quality of the videos or connect more cameras or have cameras with a higher resolution or … There are two other possibilities using hardware acceleration available in many processors. One is the DXVA2 which can be enabled in Settings and the tab Decoding/Encoding. That is not as efficient as we hoped for so the other type Quick Synch is in the process to be ready as soon as possible. These HW accelerations are not available in all processors, but in a lot of them.
Thanks,
Henrik