Hi,
I'm a longtime zm user, but had been stuck on an old version (1.28?) for a while. I recently upgraded and started fiddling with the source settings.
I've historically used only JPEG storage, but I would like to move to using the video writer only, to reduce I/O load. However, the "auto" setting doesn't seem to work. I get an error from vjs when trying to view an event. Passthrough seems to work fine for recording and viewing, including viewing while recording (which the wiki suggests isn't supported) but no analysis seems to be done when in modect mode. Also, when I go to configure zones, I get an error and no image shows up.
Is this something that should work?
I'm using Foscam FI9900EP and Foscam FI9853EP cameras.
FFMpeg with RTSP, no hardware decode
Analysis doesn't work with passthrough?
Re: Analysis doesn't work with passthrough?
Yeah auto is no bueno.
Just select h264 and libx264 as codec and in the options set crf=36
Then start reading up on ffmpeg encoding options.
Or use passthrough.
Just select h264 and libx264 as codec and in the options set crf=36
Then start reading up on ffmpeg encoding options.
Or use passthrough.
-
- Posts: 11
- Joined: Thu Apr 12, 2018 10:28 pm
Re: Analysis doesn't work with passthrough?
Many thanks. This did the trick. I wasn't sure if I would be able to disable JPEG capture and still see individual frames, but this does indeed work perfectly.
I'm now debating whether I want to run my CPU higher and do the encoding or run the hard drive harder and save all the images (although it's not necessarily clear to me that JPEG capture is higher I/O than video storage).
I have plenty of storage, so that's not an issue. I'm running several other web apps on this machine, though. Seems like the biggest benefit of video storage is that it's easier to download... or at least on the previous ZM version I was using, going from JPEG storage to downloading a video file was a hassle.
Any other tradeoffs to consider in image vs video storage?
I'm now debating whether I want to run my CPU higher and do the encoding or run the hard drive harder and save all the images (although it's not necessarily clear to me that JPEG capture is higher I/O than video storage).
I have plenty of storage, so that's not an issue. I'm running several other web apps on this machine, though. Seems like the biggest benefit of video storage is that it's easier to download... or at least on the previous ZM version I was using, going from JPEG storage to downloading a video file was a hassle.
Any other tradeoffs to consider in image vs video storage?