G-Sensor Sensitiveness

Gabolinche

Active Member
Joined
Aug 14, 2013
Messages
203
Reaction score
91
Location
Monterrey
Country
Mexico
Dash Cam
Viofo A119 V3, Mobius A-Lens, Powerucc Panorama II (RIP)
After upgrading to the newest firmware (v1.12.02), I finally remembered to activate the g-sensor this morning. I set it on low and it was triggered 5 times in my 6.5km commute. I must say, in my perception, it definitely is less sensitive than in previous firmwares, but it's still not a usable sensitiveness.

Maybe (as a suggestion to @Sungmoon Jang and possibly also @jokiin ) you can add an option that lets us users manually select the triggering values for the g-sensor. Since road and vehicle conditions vary so much around the world, users may need to personalize g-sensor triggering values more than just the usual 3 values: low, medium and high. Maybe add a fourth option like: low, medium, high and advanced. Once we select "advanced" then we can manually enter the triggering value or values.

I studied my recordings from this morning trying to find the values to offer an exact suggestion (with actual numbers), but what I found confused me. This is a graph from the values I'm seeing in Registrator Viewer:
GsensorPNG.PNG
I'm looking at g-sensor value (G), total acceleration vector (V), x-axis acceleration (X), y-axis acceleration (Y) and z-axis acceleration (Z). I found that the event I highlighted with a green oval did not trigger the g-sensor but the event in the red oval did trigger it. Why is that? If I'm understanding the graph correctly, the values in the green oval are higher (if positive) or lower (if negative) than the values in the red oval (with just one exception: the z-axis).

Anyway, without further understanding how a g-sensor event is triggered, I cannot offer an exact recommendation for triggering values.
 
Last edited by a moderator:
as far as the g sensor is concerned i think it would be good if the high settings were kept the same medium settings use the current low setting and low use lower settings than at present.
the other option is do what the gs9000 does and have 8 diffrent sensitivity settings from very high to very low and allow the user to set to there requirements
 
I think the issue is the resolution of the data. I'm not sure if it's just Registrator Viewer is trimming the acceleration data, or it's the PII which is only reporting every Nth data point.

But lets take a step back and have a look at your original question. Comparing the two sections you've higlighted I think the red activated because there was a bigger change in the acceleration (slope of the graph) in the Z-axis:
Gabo event.png
You can just barely see that the slope is steeper in the red, than it is for the green.

Now what I was getting at, is that RV doesn't report everything that's going on. It seems to only capture one point every 0.5s or so. Have a listen to me testing out my unit. The clicks you hear are me flicking the camera with my finger when I'm holding it in my other hand.
Here's the acceleration plots from RV to accompany (sorry I couldn't get Fraps to record the full screen so I could show the data with the video). I've highlighted the beeping "event" in the green circle. At 9 seconds I'm violently shaking the camera. Now if this was a true acceleration plot there would be hundreds of spikes everywhere. But either the camera is only taking a snapshot every 0.5 seconds (which would explain why each data point lines up above/below) or it's averaging out the data.

My event.png
 
Back
Top