selecting the 'best' iso setting - sorry here we go again

[dph1nm]

I think we should use more realistic numbers: 9,10,11 for the ISO 800 run, ok. If you instead run 2 exposures at ISO 1600 (with the same photons arriving) you will catch the equivalent of 4+5=9, 4+6=10, 6+5=11 (a likely split for the arrival of the photons) which will be doubled to 8,8,12 and 10,12,10 (because we've doubled the gain by doubling the ISO and before read-noise messes them up, but let's ignore that for now). Averaging these two still gets you the same information that you had originally = 9,10,11. The reason you don't really see the nice gap of 2 (as in 10,12,10) in the actual images in the higher ISO setting is the read noise which smears the quantisation.

I think this is right providing you are at unity gain or better (actually I suspect you really only need to be at a gain which well-samples the read noise). You might get some benefit if you did this at low ISOs, where you may be requring several photoelectrons before you get one ADU.

NigelM

[simon hicks]

Hi Themos,

This is really making me think! I think I said at the beginning that there is a whole load of folklore because of the huge number of variables...and I may have been spreading some of that folklore.

Themos I think you are right to a point with your analysis...I see your point. I guess the point I should have been making is that given a given exposure time (i.e. I can guide for a certain amount of time) then at ISO1600 I can get the signal (and the noise in the signal) up relative to the read noise vs doing the same exposure length at ISO800.

If you have a look at the second paper of the two I linked to earlier it takes the sky background and the gain and the read noise and the total time you have available to image etc into account and it gives you a 'best case' exposure value. There are other papers that do it other ways.

But I've applied this to some of my narrowband images from a very dark location (Galloway Dark Sky Park) and it gives me sub durations of something like 45 minutes! (A guy on the PixInsight forum made it into a script) This was based upon an analysis of some 15 minute subs that I fed into it. And to be honest, I think the subs could have done with longer exposures to get more out of the very darkest parts of the image....the 45 minute answer 'felt right'....they should have definitely been longer than 15 minutes.

Now if I was at ISO800 the calculated sub exposure time would be even longer. So going to ISO1600 at least gets me part of the way there.

So your explaination of dynamic range not really changing is quite convincing. But I think I can still get better S/N in the faintest parts by upping the exposure duration and the ISO, taking read noise, sky glow, etc into account.

Cheers

Simon

[themos]

Yes, I think high ISO has its uses and hunting faint narrowband objects in dark skies is probably one of them (that's a stab in the dark, as it were) but the reason for it, I think, is ,ultimately, the relationship between read noise and sky "background".

[simon hicks]

I agree (I think )

[James4]

Steve, the highest ISO will capture the faintest objects in the shortest time. Reducing ISO will improve image quality. James