I recently downloaded CCD-I and managed to get an hour or so of clear sky last night to try it out and learn. As dusk fell I was able to get more and more stars on the image.
During the 10+ second exposures it was recording ~50 stars and the error message about too few stars over the plane of the image occurred always. I did note a possible bug: sometimes the new downloaded image was ignored - so I had to close the program and re-start it.
Next problem was that the reported collimation error was far from consistent - causing me to stop doing any adjustments. Is this to be expected when there are "too few stars"?
I need to have consistency from image to image befgore I can try adjusting anything.

Hi Lawrence,

One thing that may help is to average say 4 images to produce your numbers for adjustment. This will usually generate a composite that uses 1000s of stars. I try and shoot an open cluster when I want to check my collimation and average 4 images. This gives me very consistent results. HTH ...

OK - easy to try! BTW, surely the notes say *avoid* clusters?

Lawrence,

Avoid GLOBULAR clusters. Non-globular star clusters are ideal for measuring collimation.

Globular clusters have a large concentration of stars near the center of the image and very few anywhere else. Because the stars are too close to each other at the center, their shapes and sizes cannot be properly measured. And because there are too few stars everywhere else in the image, CCDI can't properly measure star shape distribution.

Regards,

-Paul