×

INDI Library v2.0.6 is Released (02 Feb 2024)

Bi-monthly release with minor bug fixes and improvements

v4l2 ccd_exposure property

  • Posts: 210
  • Thank you received: 104
I first indicate that in the topic "V4L2 CCD PHD2 stacking not working" but this probably need it's own topic.

I make some testing of phd2 with a old webcam and I remark it no more work.
The problem is when I rewrite the phd2 indi client in 2015 the v4l2 driver do not reported a CCD_EXPOSURE property for the webcam that not support absolute exposure time. I use this property miss to know the client must use streaming instead to get an image.

Now the v4l2_ccd driver send a defNumberVector for CCD_EXPOSURE even this is not supported by the camera.
So phd2 try to set this ccd_exposure and this fail.

The attached log show a warning as exposure is not supported but it still send a defNumberVector for CCD_EXPOSURE.
The last two lines is the error we get after trying to start a 1 second exposure.

Is it possible to not report this property in this case? or is there another way to know we must use streaming instead of standard exposure?

File Attachment:

File Name: indi_v4l2_...2-15.log
File Size:8 KB
6 years 1 month ago #23337
Attachments:

Please Log in or Create an account to join the conversation.

  • Posts: 1029
  • Thank you received: 301

Replied by Eric on topic Re:v4l2 ccd_exposure property

That's a good point, although I don't know the side effects that could arise from not defining the property. In any case, you could probably try to expose, and if it fails, fall back to streaming?

-Eric
6 years 1 month ago #23346

Please Log in or Create an account to join the conversation.

  • Posts: 210
  • Thank you received: 104
I not really like to rely on exposure error to activate streaming because it can be difficult to be sure this is not another kind of error.
But you are right that removing a standard property is also not a good solution.

I also not like the idea to add one more config option in phd2, it will annoy much more people than it help.
Maybe define a separate driver interface class for video, different than ccd? but this is maybe overshot.

For me this is not a big problem because there is probably very few people that try to guide with a old webcam or a video device now there is so good and cheap small cmos camera, at least this is the first time I get an issue for years this problem is probably there.
6 years 1 month ago #23353

Please Log in or Create an account to join the conversation.

  • Posts: 210
  • Thank you received: 104
I make a test with switching to streaming if the first exposure result in an error.
This is not ideal but it look not too bad and it work fine with both my DMK21 that support exposure and my ToUcam that not support it.

The only drawback with the webcam is it wait for the error timeout after the first exposure before to continue with streaming. But this timeout can be reduced to 5 sec.

github.com/OpenPHDGuiding/phd2/commit/8e...c203243953f11290b6f5
6 years 1 month ago #23384

Please Log in or Create an account to join the conversation.

  • Posts: 1029
  • Thank you received: 301

Replied by Eric on topic Re:v4l2 ccd_exposure property

The situation could probably be improved by resetting min/max exposure to zero for cameras not providing exposure control. That may be an acceptable compromise. I think that right now if there's no V4L2 control available for exposure, the property remains restricted at [0.001,3600] erroneously.

-Eric
6 years 1 month ago #23450

Please Log in or Create an account to join the conversation.

  • Posts: 210
  • Thank you received: 104
Yes, this can be a good solution to set CCD_EXPOSURE_VALUE min and max to zero.
At the moment I add a checkbox to ask the user but it is good if we can be sure of the capabilities of the camera.
6 years 1 month ago #23463

Please Log in or Create an account to join the conversation.

Time to create page: 1.134 seconds