I also added another ioctl constant and a method to obtain a "sync"
pulse back from the camera.
I would love to see a mechanism to get a signal when each new frame is
available. I used to use a Meteor frame grabber. The driver for that card
generated such a signal. That facilitated writing programs with a
foreground/background architecture where a signal handler processed each frame
with real-time low-level signal processing code that was interrupt driven and
the rest of the program did high-level stuff in a non-interupt-driven fashion.
I simulate this with v4l and the pwc driver on my own setting a SIGALRM to
occur at frame rate but it doesn't work as well as having a true signal from
the driver.
Jeff (http://www.ece.purdue.edu/~qobi)
_______________________________________________
pwc mailing list
pwc@lists.saillard.org
http://lists.saillard.org/mailman/listinfo/pwc
