1 .. include:: <isonum.txt>
3 =========================
4 N-Trig touchscreen Driver
5 =========================
7 :Copyright: |copy| 2008-2010 Rafi Rubin <rafi@seas.upenn.edu>
8 :Copyright: |copy| 2009-2010 Stephane Chatty
10 This driver provides support for N-Trig pen and multi-touch sensors. Single
11 and multi-touch events are translated to the appropriate protocols for
12 the hid and input systems. Pen events are sufficiently hid compliant and
13 are left to the hid core. The driver also provides additional filtering
14 and utility functions accessible with sysfs and module parameters.
16 This driver has been reported to work properly with multiple N-Trig devices
23 Note: values set at load time are global and will apply to all applicable
24 devices. Adjusting parameters with sysfs will override the load time values,
25 but only for that one device.
27 The following parameters are used to configure filters to reduce noise:
29 +-----------------------+-----------------------------------------------------+
30 |activate_slack |number of fingers to ignore before processing events |
31 +-----------------------+-----------------------------------------------------+
32 |activation_height, |size threshold to activate immediately |
34 +-----------------------+-----------------------------------------------------+
35 |min_height, |size threshold bellow which fingers are ignored |
36 |min_width |both to decide activation and during activity |
37 +-----------------------+-----------------------------------------------------+
38 |deactivate_slack |the number of "no contact" frames to ignore before |
39 | |propagating the end of activity events |
40 +-----------------------+-----------------------------------------------------+
42 When the last finger is removed from the device, it sends a number of empty
43 frames. By holding off on deactivation for a few frames we can tolerate false
44 erroneous disconnects, where the sensor may mistakenly not detect a finger that
45 is still present. Thus deactivate_slack addresses problems where a users might
46 see breaks in lines during drawing, or drop an object during a long drag.
49 Additional sysfs items
50 ----------------------
52 These nodes just provide easy access to the ranges reported by the device.
54 +-----------------------+-----------------------------------------------------+
55 |sensor_logical_height, | the range for positions reported during activity |
56 |sensor_logical_width | |
57 +-----------------------+-----------------------------------------------------+
58 |sensor_physical_height,| internal ranges not used for normal events but |
59 |sensor_physical_width | useful for tuning |
60 +-----------------------+-----------------------------------------------------+
62 All N-Trig devices with product id of 1 report events in the ranges of
67 However not all of these devices have the same physical dimensions. Most
68 seem to be 12" sensors (Dell Latitude XT and XT2 and the HP TX2), and
69 at least one model (Dell Studio 17) has a 17" sensor. The ratio of physical
70 to logical sizes is used to adjust the size based filter parameters.
76 With the release of the early multi-touch firmwares it became increasingly
77 obvious that these sensors were prone to erroneous events. Users reported
78 seeing both inappropriately dropped contact and ghosts, contacts reported
79 where no finger was actually touching the screen.
81 Deactivation slack helps prevent dropped contact for single touch use, but does
82 not address the problem of dropping one of more contacts while other contacts
83 are still active. Drops in the multi-touch context require additional
84 processing and should be handled in tandem with tacking.
86 As observed ghost contacts are similar to actual use of the sensor, but they
87 seem to have different profiles. Ghost activity typically shows up as small
88 short lived touches. As such, I assume that the longer the continuous stream
89 of events the more likely those events are from a real contact, and that the
90 larger the size of each contact the more likely it is real. Balancing the
91 goals of preventing ghosts and accepting real events quickly (to minimize
92 user observable latency), the filter accumulates confidence for incoming
93 events until it hits thresholds and begins propagating. In the interest in
94 minimizing stored state as well as the cost of operations to make a decision,
95 I've kept that decision simple.
97 Time is measured in terms of the number of fingers reported, not frames since
98 the probability of multiple simultaneous ghosts is expected to drop off
99 dramatically with increasing numbers. Rather than accumulate weight as a
100 function of size, I just use it as a binary threshold. A sufficiently large
101 contact immediately overrides the waiting period and leads to activation.
103 Setting the activation size thresholds to large values will result in deciding
104 primarily on activation slack. If you see longer lived ghosts, turning up the
105 activation slack while reducing the size thresholds may suffice to eliminate
106 the ghosts while keeping the screen quite responsive to firm taps.
108 Contacts continue to be filtered with min_height and min_width even after
109 the initial activation filter is satisfied. The intent is to provide
110 a mechanism for filtering out ghosts in the form of an extra finger while
111 you actually are using the screen. In practice this sort of ghost has
112 been far less problematic or relatively rare and I've left the defaults
113 set to 0 for both parameters, effectively turning off that filter.
115 I don't know what the optimal values are for these filters. If the defaults
116 don't work for you, please play with the parameters. If you do find other
117 values more comfortable, I would appreciate feedback.
119 The calibration of these devices does drift over time. If ghosts or contact
120 dropping worsen and interfere with the normal usage of your device, try
127 The N-Trig windows tools provide calibration and testing routines. Also an
128 unofficial unsupported set of user space tools including a calibrator is
130 http://code.launchpad.net/~rafi-seas/+junk/ntrig_calib
136 As of yet, all tested N-Trig firmwares do not track fingers. When multiple
137 contacts are active they seem to be sorted primarily by Y position.