Batch Preprocessing RANSAC crashing?

[OrionC8]

So, I've had pixinsight for sometime now, and ever since I got it I havent been able to get any good stacks out of it. Usually I use Batch Preprocessing to make my life easier when it comes to darks/bias/flats. Ive only been able to get a picture of M42 out of it once. Every batch of data I take always end up with pixinisight halting when it comes to RANSAC, like in the picture below. Even with the newest issue of pixinsight I have no idea what the issue is, and no idea what to do.

My most recent set of data was of M65 and M66. 
15x30 light frames
7x30 darks
10 bias

Obviously not much data, I just took some to test out to see if I fix this issue. It didnt work.

Can you guys help me out?

pic related



Thanks and happy Easter!

[Geoff]

 Can you post some examples? Difficult to diagnose without data.

[OrionC8]

I made a gallery over on imgur of my lights

http://imgur.com/a/3jG0A

[Geoff]

I had a look at some of your pics. They have some trailing and the focus is not very good. This is probably why you are having trouble.  There was a post somewhere by Juan on this very issue, but I can't find it--you'll have to do a search.  Star alignment does fail on these images (being jpgs probably doesn't help), but I was able to align the images using DynamicAlignment (Processes>ImageRegistration>DynamicAlignment).  This is more tedious than StarAlignment but it works.  See here http://www.harrysastroshed.com/lrgb.HTML for a video on how to use it.
Geoff

[Juan Conejero]

StarAlignment is detecting more than 18,000 "stars" in your images --- this definitely gives you a hint of what's happening: wrong star detection.

We cannot help if you post stretched JPEG images. We need two of your original images (probably in FITS format) where SA supposedly fails.

Going directly to a probable cause of the problem: Are these images in 32-bit floating point format? If yes, have these files been generated by another application?