Computer graphics processing and selective visual display system – Display peripheral interface input device – Cursor mark position control device
Reexamination Certificate
2000-08-31
2003-03-25
Nguyen, Chanh (Department: 2675)
Computer graphics processing and selective visual display system
Display peripheral interface input device
Cursor mark position control device
C345S167000
Reexamination Certificate
active
06538640
ABSTRACT:
FIELD OF THE INVENTION
The present invention generally relates to a computer pointing devices, and more specifically, to encoding a signal produced by a computer pointing device.
BACKGROUND OF THE INVENTION
Computer pointing devices are well known in the art, and have been quite popular with computer users since their introduction. A computer user manipulates a pointing device to change the position of a cursor or other object on a computer display, or to select objects on the display, including letters, one or more words, controls, etc. The physical movement of the pointing device results in a similar movement of the cursor and/or object on the display. One of the most widely used computer pointing devices, which is commonly referred to as a “mouse,” typically includes a ball rotatably disposed in a cavity formed on an undersurface of the mouse. The ball rotates as the mouse is moved across a flat surface, often over a pad, and an encoded signal is produced in response to the movement of the mouse.
A conventional ball-based pointing device includes a pair of encoder shafts in the housing that are frictionally driven by the ball as it rolls over a surface. The encoder shafts rotate about a pair of orthogonal axes, i.e., the “X” and “Y” axes, in response to components of the ball's rotation along those axes. At one end, each encoder shaft terminates in a serrated disc. As the encoder shafts rotate, these serrated discs periodically interrupt a light beam traveling from a light emitting diode (LED) source to a photo detector, generating a nominal square-wave signal at the output of the detector. The signals produced by the detector indicate the device's incremental motion along the orthogonal axes and are processed by a micro-controller or other logical processor in the pointing device, which produces a corresponding stream of digital values indicative of a position of the pointing device relative to the X and Y axes. The digital values indicative of the displacement of the pointing device relative to these two axes are passed to a driver program executing on the computer to which the pointing device is attached. The relative position data are processed by the computer, causing a cursor or other object displayed on a monitor to move in response to a user's movement of the mouse. Thus, the X and/or Y movement of the cursor on the display screen is a function of the rotational motion of the ball within the pointing device relative to the X and/or Y axes.
The frequency of each square wave is a measure of the X and Y components of the velocity of the mouse, but this frequency only indicates the speed of movement along one of the axes and does not indicate the direction (left/right, up/down) of the mouse motion. To determine directional information, a second photo detector is disposed near each disk such that its waveform is displaced 90 degrees in phase with respect to the first detector's waveform. By sampling the two X waveforms, a microprocessor can infer both the speed and the direction of the X component. The speed and direction of the Y component are determined in the same manner.
The waveform for one axis represents a series of transitional states. It is the sequence of these transitional states that indicates the direction of travel. For every state change, a new movement command is issued by the microprocessor. Normally, when a user manipulates a computer pointing device at a modest speed, the frequency at which the microprocessor samples or “looks at” the waveforms is sufficient to detect every transitional state of the waveform. However, as the mouse moves faster, the frequency of the square waves increases, effectively reducing the number of samples taken by the microprocessor. If the mouse speed is sufficiently great that a transitional state is not sampled, then the processor sees an illegal transition between non-adjacent states. Traditionally, when an illegal transition has occurred in prior art encoders, no movement command is issued for the illegal transition. Lack of movement commands for such illegal transitions can cause the mouse-controlled motion of a cursor on a computer screen to appear to falter. The traditional rationale for ignoring illegal transitions is that such transitions are directionally ambiguous, i.e., a two-state jump could mean a movement in either of the two possible directions (left/right, or up/down).
However, a conventional mouse encoder actually misses more states than can be accounted for by an excessive rate of mouse movement. The theoretical symmetrical rectangular idealized waveforms are rarely ever attained from the optical sensors. There are several causes for this problem. First, the vanes on the serrated disks used in the encoder are not zero-size objects. Although the vanes on the serrated disks are sharply defined, their finite size causes the light reaching the photo detectors, and hence their response, to increase and decrease gradually, rather than instantaneously. Also, the response speed of the LED detector pairs is not instantaneous, and that response speed is a function of the sensitivity or gain of the LED detector. A third reason that the actual waveforms vary from a theoretical ideal relates to the one/zero decision threshold of the processor in response to the signal input from the encoder, which is seldom one-half of the peak signal magnitude. A final cause is that physical constraints mean the ideal 90 degree placement of the photo detector pairs for each encoder disk is in reality not exactly 90 degrees, being subject to achievable mechanical tolerances.
Due to the first and second causes described above, the rise and fall times of the detector waveforms are finite. Because of the third cause noted above, the waveforms will be sampled at other than one-half of their peak values, which means that the one/zero decision threshold for the processor depends on both the sensitivity of the LED/photo detector combination and the actual threshold voltage of the processor input. Both of these factors can vary considerably over a production batch. The result is that the duty cycle of the waveforms will vary from the ideal value of 50%. Finally, the imperfect mechanical positioning of the pairs of photo detector pairs for each encoder disk means that the dual waveforms that they produce will not actually exhibit the expected ideal 90 degree phase relation.
These effects combine to cause the different transitional states to have different widths (that is, different time durations), which causes the missed-state/illegal transition effect described above to begin occurring at slower mouse speeds than might otherwise be expected. Clearly, states whose durations are shorter than the sampling interval will not always be seen. For instance, a state whose duration is equal to one half the sampling interval will not be detected about half the time (assuming the state period and the sample period are incommensurate, as is normally the case).
It would be desirable to extract useful information from illegal state transitions, and to use that information to improve the performance of a computer pointing device, rather than completely ignoring these illegal state transitions as is taught by the prior art. Contrary to the current view, the directional ambiguity of a skipped state should not be a concern, because when a pointing device is being moved at a sufficiently high speed to cause illegal state transitions, it is very unlikely that the direction of movement of the pointing device can be reversed rapidly enough to cause an error in determining the direction of movement. To underscore the improbability of such rapid reversal in direction of mouse movement, note that a typical sampling interval is 200 microseconds, and skipped states might start occurring at speeds over 5 inches/second. To reverse the direction of a mouse in a time period of less than 400 microseconds (i.e., 2 sample intervals) would require the mouse to decelerate and then accelerate in the opposite direction at over 25,000 inches/second
2
. Realistic
Ostrum Brett T.
Rensberger Gary
Microsoft Corporation
Microsoft Corporation
Nguyen Chanh
LandOfFree
Skipped-state method for mouse encoding does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Skipped-state method for mouse encoding, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Skipped-state method for mouse encoding will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3012793