trackpoint drifts off

[jaynyls]

Hey all,

My trackpoint seems to drift off occassionally while I'm using it. That is, the mouse cursor will drift on the screen for a little bit. Is there some sort of calibration that I'm supposed to do?

thanks.

[Goobie]

Did you by any chance installed SP2?

Ever since I installed SP2, my regular optical mouse pointer will start to drift, very annoying! I have no solution for it yet.

[Humpa]

The IBM trackpoint has a recalibration program that runs under certain conditions. If your finger is resting on the trackpoint (even slightly, such that you dont even think you are touching it) for a certain number of seconds, then this self-recalibrating program will run (which is the "drifting" you see).

I used to see the "drift" frequently, but once I learned how to keep my finger off the trackpoint except to move it, I stopped seeing any drift. For example, when you move the cursor over to a certain part of the page, take your finger completely off the trackpoint - do not let your finger linger there or it will start the calibration.

[plucky duck]

Is this a characteristic/flaw of trackpoint devices? Cause I've worked with several different laptops, not just IBM, that do the same thing.

[ZPrime]

It's not a "program" though.

The way the trackpoint works is similar to an old analog joystick. The stick periodically "recalibrates" itself internally, in case it has drifted off-center.

You'll get weird drift if you hold it in one direction for a long time. The stick will attempt to compensate/recenter itself "against" your direction -- if you are pulling down a lot (scrolling, perhaps?) the stick will try to center itself by adjusting in the opposite direction. if you take your finger off the stick, you'll see it pull upwards for a bit. However, after the lack of pressure it will again recenter itself and stop moving.

Humpa's solution is right though - keep your finger off it, or use it in "bursts." for extended scrolling, use the "traditional" middle button technique (click middle, pull it down slightly, and then LEAVE IT ALONE to auto scroll). IBM's "universal" middle-button scrolling will cause the trackpoint to suck more frequently.

[jaynyls]

thanks a lot for the input. i wasn't aware that you're supposed to treat trackpoints differently than mice.

[Humpa]

Is this a characteristic/flaw of trackpoint devices? Cause I've worked with several different laptops, not just IBM, that do the same thing.

I have read that this is a characteristic of any trackpoint device.
As with joystick-like equipment, the trackpoint needs to self-calibrate in order to remain accurate. So it takes the position the trackpoint has been in (for the last 10-30sec?), and uses this as it's center position.

According to IBM, these 4 conditions cause drift:
# When the computer is turned on.

# When resuming normal operation.

# When the TrackPoint III is pressed for a long period of time.

# When the environmental temperature changes.

I also just found this document:
http://trackpoint.almaden.ibm.com/files/ykt3eext.pdf
Which says this:

2.2.3 DRIFT CORRECTION
Due to the significant temperature sensitivity of the force sensors and circuit components, it is necessary to
recalibrate the zero force origin of each axis on a periodic basis. In order to properly recalibrate, it is necessary to
determine when a ‘hands off’ condition exists (i.e. when no force is being applied). This is done by monitoring the
X and Y axis low order (l count) force samples, and when there is minimal change in the readings (no more than ±1 l
count) for a specified length of time (0.56 seconds by default) it is assumed that the hands off condition exists. At
this time the X, Y, and Z axis origins are reset to the smoothed current value of the X, Y, and Z axis force samples
respectively. It should be noted that noise introduced by any source other than applied sensor forces will severely
impact the effectiveness of this technique. Parameter controlled smoothing can partially compensate for moderate
noise.
The Z axis, due to its higher drift sensitivity, also uses two additional drift correction mechanisms. Since Z axis
forces can only be applied in one direction, if the Z axis value ever exceeds the Z axis origin in the incorrect
direction, the Z axis origin will track the present value of the Z axis. The remaining drift correction mechanism is:
when the Z axis running average approaches the Z axis origin (within 3 counts by default) the Z axis origin will be
reset to the present Z axis value. This last drift correction is limited in frequency, by default to once per second.