Y aces center drift

[impulse200]

Hello.
Many times in many games as well as in trackir software I notice that when I tilting my head downwards or upwards Y axes doesn't return to it's center. I have to reset trackir position or shift my head up/down several times to set this axis to it's center. This is very inconvenient when dogfighting in aviasims.
Other axis seems to be just fine.
Is there any solution to this issue?

[MikeWitham]

impulse200,

Is there any way that you could post a video showing the problem? I have to re-center my TrackIR occasionally, but it's because I don't move back to the original position.

Regards,

[impulse200]

I'm uploading in-game test footage from il2 bos to http://youtu.be/rWi0MfS-JEE . I do not know (yet) how to capture video from desktop to show this effect directly in trackir control software.

My Y axis settings in trackir software are:

[impulse200]

And another one showing exactly what I experience while dogfighting.
http://youtu.be/Fk6jmPb7ZDQ

[MikeWitham]

impulse200,

After reviewing your profile settings and your videos, I suspect that the drift you're experiencing is due to the high scaling you have set on Y—in particular in the center of the graph. Even when you return to your "neutral" pose (where the graph should approach 0 on the X axis), the scaling remains 4x. This means that if you are "off" by as little as 5mm in reality, it will be scaled up to be treated as 2cm when streamed to the game. This kind of deviation (~5mm) is typical when returning to a neutral pose, so it would explain this kind of perceived shift in the tracking.

To minimize this effect, you could drop the center point in the graph all the way down to 0, as shown in the screenshot below. This will create a true deadzone, minimizing the likelihood that small deviations in your head position at neutral pose will be scaled up to appear as multiple centimeters.

Untitled picture.png (40.95 KiB) Viewed 3899 times

Regards,

[impulse200]

impulse200,
To minimize this effect, you could drop the center point in the graph all the way down to 0, as shown in the screenshot below. This will create a true deadzone, minimizing the likelihood that small deviations in your head position at neutral pose will be scaled up to appear as multiple centimeters.

Hello, Mike. Thanks for your suggestion. I've shifted central point down to 0,0. I guess it helps a bit. But my issue haven't gone away completely. There are still cases where my head stuck too high or too low off center. But gunsight is still visible though. It helps a bit