More time or up the ISO?

[earth titan]

Still playing with M13. Trying not to blow the core by running too high an ISO. The image following was all I managed to grab without star trails, as I forgot to do up one of the key mount retaining screws. This was ISO400 for 240 seconds, unguided (I'm awaiting some cash to arrive which will fund a guidecam).

I'm guessing 300 seconds or more? Can I leave the ISO at 400 or up to 800?

Any help much appreciated.

[Tinker1947]

There's people that run the ISO at the max there camera will 1600/3200

[Stephen]

Above 800 I suspect will start to get noisy.

Is that a single sub or a stack?

[earth titan]

It's a single sub. There is data in it, as I've been playing with Gimp this evening. This is not a serious image yet, just feel upping the ISO to anything above this will result in noise.

[dph1nm]

ISO does not affect the sensitivity of the camera. So from a signal-to-noise point of view it doesn't really matter what ISO you set (for Canon DSLRs higher ISO is actually better).

NigelM

[themos]

Nigel, as long as you're not read-noise limited, right?

Here's the decision tree:

1) Set ISO to lowest

2) are you read-noise limited? If yes goto 3 else you're done.

3) increase ISO , then goto 2.

The reason you start with the lowest is that low ISO gives you the most dynamic range.

[Jessun]

Your PA is looking good, but you might want to try shorter subs - and thus more, in order to nail the details in the dust lanes which I find the most fascinating aspect of this target.

To guide you can get a webcam and temporarily use your finder scope for virtually no cash at all. :-)

Cheers

/Jesper

[dph1nm]

Nigel, as long as you're not read-noise limited, right?

Here's the decision tree:

1) Set ISO to lowest

2) are you read-noise limited? If yes goto 3 else you're done.

3) increase ISO , then goto 2.

The reason you start with the lowest is that low ISO gives you the most dynamic range.

Well there is always read noise, even if it is not a major contribution. Personally I like to minimise it, so always use ISO1600 on my Canon 1000D.

The worry with very low ISO is that you go below unity gain (1e-==1ADU) and so start running into quantisation issues. These are difficult to quantify (ahem!), but your probably want to avoid a situation where you have a gain of 10e- per ADU but only 1e- coming from your object (you may still detect it by the way, but I suspect the s/n will be reduced).

So I guess it really depends on whether you are more interested in the faint parts of your image or the bright parts!

NigelM

[LukeSkywatcher]

I would take shorter subs and lots of them and stack. There is already loads of good detail in your image (dust lanes etc). You will be amazed at how well things will look even with only a few dozen short stacked subs. It may also be worth splitting the subs into 2 groups:

1/ The core

2/the rest of it

then you can layer the images in PS. Its a technique many people use on M42 so they do not blow out the "Trap". I have not explained it well. I'm sure someone knows what i mean and can explain it better.

Basically less subs of the bright core and more of the rest of the galaxy................same settings for each (i think), just less subs of one part of galaxy and more of the rest.