[ros-users] call for testing: camera1394 in C-turtle

Jack O'Quin jack.oquin at gmail.com
Wed Jun 16 16:02:41 UTC 2010


2010/6/16 von Wichert, Georg <Georg.Wichert at siemens.com>:

> now I tried two other cameras of the same type AVT f145c2 , but ~2 years
> younger, and now the node detects the camera, but reproducably says "Could
> not reset camera" in dc1394_camera_reset(), control.c line 500.
>
> Coriander also has this problem. Pushing the reset button in the camera tab
> leads to the same error message. the camera does work nevertheless. Is the
> initial reset really neccessary?

I am not sure whether it is necessary or not. I suspect the answer
depends on the exact camera make and model. As far as I can tell, it
*does* seem to be customary for libdc1394 programs to reset the camera
when opening it. Perhaps others with more experience will correct me
if this is wrong.

Just to be completely clear: when you say the camera "does work" , do
you mean with coriander, with camera1394, or with both?

Looking at the code, it appears that camera1394 will fail to open the
device if the reset fails. That looks unnecessary and probably
counter-productive. If coriander works and camera1394 does not, let's
open a ticket for the open failure, at least.
-- 
 joq



More information about the ros-users mailing list