fixed bug that caused excessive heat

[iothing]

I did some measurements with a usb current meter [1] and found out, that we really should only set gpio_ms1 pin 2 to 1 when we want to change the position of the filter. How this position is changed is in fact controlled by the gpio_aud writes, maybe some latching is going on here I guess.

In daylight without an rtsp stream current is 0.29-0.32A at 4.95V.
In nightvision mode the current is 0.43-0.46A at 4.95V.

[1] http://www.banggood.com/en/KEWEISI-3V-9V-0-3A-USB-Charger-Power-Detector-Battery-Capacity-Tester-Voltage-Current-Meter-p-984613.html

[claell]

@samtap Unfortunately there is currently no release with this fix. As this seems rather important to me, would you mind to create a new release in the near future?

[samtap]

I have lots of fixes and improvements but creating a release is a quite inconvenient at the moment. So I'm trying to implement some kind of auto-update mechanism (pulling updates from github instead of flashing a new release to sd). I'll do a new release when that's done.

[claell]

Thanks!

[Sagonips]

This fix makes the image have a pink hue to it when the IR is on compared to the standard black and white. It appears to leave the camera in colour rather than switching it to black and white like it should. Did you experience the same?

[Sagonips]

I realised it was the second echo 0x40 > /proc/isp/filter/saturation in the IR on set of rules that needs to be deleted.