More tangled webs of cameras

This link seems to have good information on video for Linux and the simple program works well, there is some confusion however, and I wonder if the issue is version related. I am going to take the plunge to the new version 9.04, I thnk and see how that works with v4l. Certainly it is operational and it requires some effort to use the appropriate interfaces. It seems that I could write an article that was informative and separated the methods that might be used to operate the devices and the depth of technical operation which extends from the interface to libraries to kernel device driver interface methods through USB.

The implementation of a user interface could be daunting as it involves ioctl and device reads and numerous APIs, but a good example can go a long way to de-mistify the complexity. I think I grasp the overall structure and I can see how it is a WIP, I think I will just keep track and follow the progress of the interface for a while to see where it is headed. For a supported device using the test program, the delay to open the device is essentially zero and the quality and usability is certainly as good as with Wind Dohs! without all the associated BS.

I am still wondering about the name issue and the length of it limited to 31 characters, but if the spec is 31, then the video device is violating the spec, which I would expect to be more likely than the v4l people being wrong as them seem to be very detail oriented.


Automated Intelligence

Automated Intelligence
Auftrag der unendlichen LOL katzen