I had dithering turned on to “very high” in SGP and then even set it to “extreme” later on and even though the dither commands were being sent to PHD2 and the dither lines showed up in the PHD2 graph, it didn’t seem like there was anything happening. I looked in my PHD2 configuration and the dither factor was set to 2.5x (2.5x the normal dither amount that PHD2 recieves).
Here is a stacked image of the Rosette with only some small stacking artifacts around the edges: https://puu.sh/zoL8V.png
I had performed a meridian flip and platesolve in the middle, so that’s probably why there is a small black boarder around the image. But in general it doesn’t seem like there was actually any dithering going on.
Something very interesting I noticed: At the end of each frame, there was a large spike in the PHD2 graph like the mount stopped tracking while the image was downloading. I’m not sure if this is related.
Looks to me that there aren’t any hot pixels in the image, though it is hard to tell looking at the image. The large spike that you are seeing at the end of an image down load in phd is the actual dithering process taking place. Bottom left of screen will show the movement briefly.
When you run thru a stack of images, do they appear to move as you click thru them? Everything that you describe seems to indicate that all is well.
The large spike was not from the dithering though; I was dithering every 3 frames but it was happening after every single frame. When PHD2 does a dither, there is a white dotted line that shows up on the PHD2 graph. When the white dotted line showed up in PHD2, there were no spikes in the guide graph at all. The spikes I was seeing happened exactly at the end of every frame before the image download even finished and SGP hadn’t issued the dither command yet. Additionally, I cranked the dither to “extreme” and set the dither multiplier to 2.5x which means there should have been a significant amount of dithering going on each time. Yet it seems there’s none (because the black border that currently exists in the image is from when I did a meridian flip and re-solved the frame).
The PixInsight log of the registration process will tell you exactly how much offset each image has from the reference image. Referring to that you will have a very accurate measure of the presence or lack of dithering. This will be obscured to some degree based on the accuracy of your guiding. If your guiding is really good, the differences will be (almost) entirely due to dithering. If guiding is poor, then some of the difference will be due to guide errors. In either case, you will know if you are getting any offset between images.
I always get some degree of guiding spike at the image transitions. I have always attributed it to vibration caused by the mechanical filter change.
