Vitamin D Agent quit unexpectedly

[classicshotokan]

I'm working on adding cameras. I've learned the if you turn the cameras on one by one you can easily identify the process in activity monitor.

I have an Axis 1054 HD cam that has 23 fps and a clear picture despite being in low light conditions. Yet, when I start it in V, it consumes 40% CPU. Is this because the cam is HD, or because it's in low light?

I'm trying to motivate to test all cameras at once, all looking at the same scene, all wired identically. Day and night. To identify the "cameras acceptable to VDV," is low CPU usage the only thing I need to be looking for?

[ryan]

The high cpu usage in this situation is due to a combination of the high resolution and the high framerate. The more pixels, the more processing is necessary, something also true for an increased number of frames. We really only need about 10 frames/second to do optimal tracking, so if you can change the camera settings to lower the frame rate a bit you should see lower CPU usage without loss of tracking accuracy. I'd probably say that not dipping to far under 10 fps is probably the most important thing to ensuring VDV works well. CPU usage is only going to affect how many cameras you can run on a system, it shouldn't affect the behavior of the app itself.

[classicshotokan]

I have resolved this situation, I think.

You were correct: there was a problem camera, an Axis 207. It was the first camera I bought. I didn't notice the problem at first because as the first cam the computer had sufficient excess resources to run everything OK despite the load. It was only after I bought an Axis 1054 HD cam that the problem showed up, so I blamed the 1054.

After buying another camera to replace the 207 the system is working great with even more cameras and room to add at least one more. More like what I'd expect.