How do you guys judge what ISO/Exposure etc...

[stan26]

Hi Guys, just wondered what sort of image/detail I should be looking to get in my individual subs before stacking and editing.

For example, last night I had ago at the M51. Im using a EQ3-2, startravel 102 frac so I'm limited realisticaly to exposures of around the 30sec mark to get a good amount of keepers (till i perfect my polar align) I set my canon 500d to ISO 800 being this seems to be good balance between noise and pulling out the detail (used on M42 with great results) took a test shot and could clearly make out M51 after a 10sec exposure. I managed to get 15 well tracked shots at 30sec and took 15 darks at the same time. After loading them into DSS and reg/stacking all I'm left with is a black final image with a few of the bright stars. I am able to pull out the M51 in PS but its ever so faint........

Did I load in to many darks? OR is because I should use a higher ISO in the first place? It always seems that my single subs have more detail than my stacked image..........

How bright/exposed should my subject be when I take my test shots to find ideal exposure...?

Can you help...

Cheers

Stan

[Johnboy2011]

I'm only just beginning so I'm sure someone on here will explain much better than me but I had a similar problem. I found that my test jpeg was much better than my stacked tiff. I got some advice and found that initial processing in DSS was critical, especially pulling out brightness with the lower slider on the middle of the three options, probably best to leave the others alone. I went back and revisited my 25sec Sombrero and M51 and was really pleased what I managed to pull out.

I hav'nt managed to get anything out of GIMP yet (free processing software)

John

[ollypenrice]

I use neither DSLR nor DSS but it sounds like you are being given a linear TIFF or FITS file after the stack. This is normal in stacking software. The image then needs to be 'stretched open' which means taken from a linear range of brightness values to something more or less logarithmic.

Like many others I do this in Photoshop using Levels and Curves. What you try to do is brighten the low brightness values (coming from faint nebulosity etc) while not brightening the bright stuff more than you can help. That way the fullest range of brightnesses can be rendered visible in the image.

Rob, excellent fellow that he is, has done a tutorial on his website. It is an absolutely first rate explanation of this fundamental imaging technique and skill.

pic and description page

Olly

[jgs001]

To be honest, with that setup, I'd go for ISO1600, and probably at least 100x30second exposures. Stick with the 15 darks... This will take a long time to stack, but it'll make a difference.

As Olly said, this is normal, when I did M51 a month or so ago, I could not see M51 on the stacked tif, and it was only with the stretching the image appears.

[stan26]

Thanks guys,

Oly that site looks a good read, I will have a proper look later, thanks...

I think you are right jgs001 that a higher ISO is needed to compensate for the restricted exposure time. My best ever M42 edit was done at ISO3200!

So 15 darks should be enough for 100 subs @ 30sec each...?

The bit im really struggling with is brining out the spiral arms etc without blowing the background. It ends up looking like a horrible noisey orange sky with a bright blown out blur in the middle.....

Thanks

Stan.

[davew]

Hi Stan,

First I'll tell you I haven't got a 500D ! But I know someone who has.

The results from 30 seconds ISO800 look very clean in the main. The ISO 1600 look good too. The images aren't full of hot pixels.

Give this a go perhaps and see what happens -

Try 30 seconds ISO800 or 1600. Take more than 30. Take 50 bias. In DSS use Kappa-Sigma clip to stack the lights and Median for Bias. Here's the bit that some will disagree with..... In Cosmetic click Detect hot pixels and set to 1 pixel, detection around 90/85%. This is like using a bad pixel map in the expensive stack programmes. No darks in sight.

Also process the normal way and compare. Better / worse ?

It'll cost nowt to try. Then stretch the results in PS as normal.

Dave.

Edit.... This doesn't get rid of an Orange sky.

[dph1nm]

Higher ISO won't make much (if any difference) to the final image. What matters is the number of photons you collect, which is the same at any ISO. So you do need more subs. Here is a shot with a Celestron Nexstar 4" f6.5 scope (Canon 1000D, ISO1600 1 min subs - 30 mins in total, stacked and stretched in DSS. No darks) - and I have a lot of light pollution, which makes the images a much noisier than from a dark site.

NigelM

[jgs001]

Stan, you really want exposure time... and the only way to get that is many many subs.. if you're feeling brave and have a long time to process, take 200... That's still only 100 minutes, which really isn't very long. The tracking accuracy is actually going to help somewhat, as each frame will not be in exactly the same place, so using the Sigma stacking methods will actually get rid of hot pixels etc, and any planes, satellites etc.

[stan26]

Thanks for you help guys. Slowly getting there. I have managed to get aroung 30mins worth of "ok'ish" subs from my setup so far. Thats deleting about 60-75% of the subs i take to be expected from EQ3-2 I suppose......

Here is what I have so far using advice above. I reackon once I have a bout 2hrs worth it will start looking a little more like M51....

Alot of learning to do with image processing as you can see......