[Stonius (Markus)]

I did a bad thing as tested my gear out of interest, and now it started me thinking. I should have known better than to go looking for issues where there were none :-D

I did some tests shooting a photographer's neutral grey card with my 1600MM and a set of ZWO RGB and NB filters.

Compared to the Luminance filter, exposure times increased by a factor of 3.25 (R), 3(G) and 3.75 (B) respectively - a range of roughly 1.5-1.8 stops.

I notice most people do sets of RGB subs at the same exposure times for all filters regardless of the chip's QE in that frequency, or the individual filter's ability to transmit light.

Given that the blue channel is noisier anyway, I'm thinking maybe I should compensate exposures for R and G to get the same amount of data density on each sub. That way, the colour balance when RGB channels are combined will be more accurate and per-sub SNR levels should be more similar.

BTW, the NB filter exposures increased by 43.5 (Ha), 32.5(O3) and 50(S2) times relative to Luminance (which from my reckoning is about 5-5.8 stops).

I suppose a OSC would suffer from these inconsistencies too, given the bayering filter and chip would have the same associated problems.

Maybe this is all irrelevant given that some subs will get tossed so you're never going to have exactly the same data for each channel, but surely that is the initial objective?

It seems to me that without compensating I will have a 'sensitivity footprint' across all my images where the Blue channel is consistently poorer than the Red or Green.

I know most people bump the gain up for NB imaging, but even with that in mind, the same thing applies - should I expose S2 for slightly longer than O3 (with Ha somewhere in between the two)?

Am I completely overthinking it? Maybe I'd be better off just getting more Blue subs to compensate, for example.

What do you guys think?

Markus