Skip to content

Recording constantly stops

edited March 2015 in SecuritySpy
It seems that no matter what settings I'm using, the recording will stop in the middle of motion resulting in hundreds of movies per day when really, there should only be maybe 10 or 20 events. I understand that motion detection is not perfect and may experience false positives but I can't understand why during obvious motion that the program would stop recording and then start again 1 second later. Anyone else experience this?

Comments

  • Hi - it sounds like you might be using a very short "post capture" time for the motion detection. Try increasing this, perhaps to 10 or 15 seconds. See the Motion capture section of the SecuritySpy user manual for more information. Does this resolve your problem?
  • I will try this. But isn't it supposed to record through the duration of sustained motion? If a car passes by, it may only be 2 seconds. I don't need the 13 seconds after the detected motion. And if an event lasts for 50 seconds, should I expect that the video be broken up into segments? I wonder if there are other factors that I'm not considering.
  • Movement of people and cars is often jerky: for example, a person might walk into the frame, stop for a few seconds, then start walking again. You will want this whole event to be captured into a single file, and this is achieved by the post-capture time. Storage space is cheap and the files that SecuritySpy creates are efficient (especially if you are using H.264 encoding), so the few seconds at the end of the motion-detected movie that aren't capturing anything interesting is inconsequential. A longer post-capture time will resolve your problem, so this is what we would recommend in your situation.
  • Ok. Excellent. I'll try that. Thanks very much!
  • I've never actually thought about this, and I'm sorry to say I don't know the answer but this might be relevant:

    If I have a camera with a 20 second pre-motion buffer, and a 1 second post-motion buffer, then if I create three motions in 60 seconds each separated by a 5-second interval of no motion will that be one file, or three?
  • Hi @jtodd - the number of files you get is dependent on the post-motion time and the gap between motion events. As long as the frequency of motion events is less than the post-capture time, you will get a single movie being created. If however the gap between motion events is longer than the post-capture time, SecuritySpy will already have stopped the previous capture before the next motion event occurs, and you will get additional files.

    Another consideration is that SecuritySpy uses the post-capture time in order to "catch up" with recording any video buffered during the pre-capture time. It doesn't capture the buffered video immediately upon trigger (this would put too much strain on the CPU and/or disk), so instead it captures this video at double real-time speed. So it's a good idea to have a post-capture time that's significantly longer than the pre-capture time (more than double), in order to allow SecuritySpy to do this effectively. If the post-capture time is too short, there will still be a lot of frames in the buffer when the period of motion ends, and SecuritySpy has to finish processing these frames before a new capture can be started, effectively rendering SecuritySpy blind for a few seconds.
Sign In or Register to comment.