I think Andrew may have presented the math and how it has been simplified to the 5x rule with regard to the ideal f/ ratio ( if not Andrew then this background has been presented and discussed by others on a number of occasions I can recall).
Anyway I am happy to follow this guidance and do.
I was interested in the comment that Mitchell made above i.e. that “higher fps have diminishing returns”. I have often wondered about this and would use the reasoning that the distortion of the image through processes that degrade the image must, most of the time, occur at some typical rate for much of the time and with higher rates for lesser amounts of time. This would imply that there would be a limit to the usefulness of very high frame rates. If they start to exceed the typical rate of distortion then it would follow that many frames would be similar to the previous one.
Is this reasoning reasonable!!
Cheers Paul
I come from the long exposure side of things where the camera spec highlight is on full well depth and Qe where if the photons fill up a pixel and they spill over into the adjacent pixels that are dark or empty, the star ‘bloats’ and requires reduction in processing. I suppose the same can be said for areas of contrast where planets are concerned. Smaller pixels do fill up with photons faster and color cameras are more prone to spillover due to the arrangement of the bayer matrix.
Maybe what Jan is suggesting is just an observation based on the fast optics used in his use case where at f5 there are 5x the photons hitting the sensor than at f10 instead of what we understand as a mere 2x due to the increase image scale. In this case, reducing the frame rate combined with larger pixels would in fact provide increased contrast. There’s also Qe to consider as well which will be a determining factor on contrast.
I think there’s more analytical work to be done with lucky imaging and how those parameters work within the methods we use to capture planets. I suspect just as in deep sky AP, the REAL answers lie in analyzing the individual sub exposures or frames. But such analysis would be difficult in real case use as astronomical seeing is never consistent. One would need a consistent simulation of a light source with significant surface contrast to capture and analyze.
When the seeing is poor and the target is jumping around, i increase the frame rate based on the observation…..but I can only increase that frame rate by increasing the gain if the histogram is given consideration.,…with an F5 scope this wouldn’t be necessary….or even at F10 compare to the F20 and higher we often use. Hmmmm……I feel like the real answer lies in those pixel well depths and Qe.
Edited by mayhem13, 25 February 2023 - 04:53 AM.
