It would be interesting to see the PHD2 debug log for the same time period to see what’s going on over there - did PHD2 wait 7 seconds to send the settle done message or did SGP take 7 seconds to recognize it?
I will add my two cents here. I too sometimes run sequences with many short exposures. I agree completely with @spokeshave, and it seems Jared has understood the issue and come to the right conclusion. The issue is not just that SGP checks to make sure PHD2 is tracking accurately before starting an exposure - SGP actually waits through the specified settle time after a non-dithered exposure, even if the tracking is accurate within the specified number of pixels. You can watch the settle time counting down on the status line in this case, and occasionally the message “Waiting for Guider to Report It Is Done Settling” is seen after the count-down, even though dithering was not performed. This is completely unnecessary when PHD2 continues uninterrupted guiding after a non-dithered exposure - SGP should start the next frame (assuming the tracking accuracy is verified) as soon as possible, without waiting for the dither settle time, when no dither was performed.
It is true that there are situations other than dithering that can upset the guiding accuracy, but these situations can occur at any time, and the maximum pixel error abort feature will take care of them.
I don’t think anything needs to change from SGP, there’s already a settling tolerance level, if you want to increase the tolerance just raise the number.
@entilza, you miss the point. The tolerance has nothing to do with it. The problem is that SGP waits for the specified settle time after each frame, even if dithering was not executed. This happens regardless of the tracking accuracy. It is not waiting for the pixel error to be below the tolerance level - it goes right into the settling countdown after each frame, even when the tracking is perfect.
Yes but it’s an important feature just in case something has caused the mount to move. If you increase the tolerance it will still try to settle but much quicker.
I disagree. The situation where “something has caused the mount to move” can occur at any time. If it happens at the end of an un-dithered exposure, that is purely coincidental. If it happens at the end of un-dithered exposures frequently, then it is likely happening all during your imaging sessions, and you need to address that problem. If you don’t want to solve the underlying problem, you can use the maximum pixel error abort feature to deal with it.
If your guiding is under control, setting a higher tolerance will not help with this issue, because the wasted time is not due to the tolerance being too low, it is due to the settling time being implemented even though there was no dither performed. In my experience, at the end of un-dithered exposures, the guiding is within the tolerance limit, and SGP goes immediately into the settling count-down as soon as the previous frame finishes downloading. This represents time that a new image could be under way.
Having sometimes set up sequences to acquire many short exposures with dithering every X frames instead of every frame, I consider this to be a defect, not an important feature. I think that if you (or anyone) tried this, you would agree. I don’t believe this feature has ever saved a frame for me, but it causes dozens of frames to be lost in a typical night.
I think I know what causes this issue. Right now I’ve been observing PHD2 and SGP - what happens during this settling time, and it seems that PHD2 must do the full cycle of exposure before sending the “Go” command.
For example - I set 4 sec of exposure on guider. Frame in SGP will be downloaded during this 4 sec exposure, next - PHD2 waits for the next 4sec (summary is 2-3sec from previous cycle and full 4 sec of next exposure), and when it is ok (with tracking) program sends “go” command.
No @Pav1007, this is not what is happening. See my previous posts. SGP actually goes right into the settling time countdown after the previous file download completes. I typically use 0.5 to 1.0 seconds exposure in PHD2, and 10 seconds settling time in SGP. Watching the status line in SGP after the file download completes, I see the “autoguider is settling” message along with the countdown from 10 seconds. So it is definitely SGP doing this, and IMHO it is completely unnecessary if the previous exposure was not dithered.
I’m experiencing the same issue. It’s making short exposures very inefficient. Unfortunately I’m in a severely light polluted location with a fast scope so broadband exposures are very short even at zero gain. I could use direct mount and go unguided. I’ve had very short delay times between images by doing that in the past but I’m using ONAG and FocusLock which needs PHD2 to actually be guiding in order to maintain the focus continuously. I spoke with the Lead Dev at Optec yesterday and he said he would try to implement focusing when a star is selected in PHD2 but without guiding, just in a green box… That will work but the star would need to be selected without guiding on target change or sequence start or automation goes out the window. Maybe that functionality could be implemented in SGP?
