DSLR Sub lengths ?

[Davey-T]

Hi Guys,

Haven't done any serious DSLR imaging for a while but thinking of just taking portable set up and DSLR on holiday .

When using CCD camera I find the longer the subs the better but using the DSLR I find that subs quickley become over exposed, so is there an optimum setting ie long exposure low ISO or high ISO  short exposure ?

Any thoughts appreciated.

Dave

[ollypenrice]

I would just experiment. An awful lot depends on your skyglow and, I expect, on the ambient temperature when the camera isn't cooled.

Olly

[ChrisLX200]

The thing about DSLRs is you get the results immediately, so look at the screen   My 'usual' is 10min @ ISO800 for DS objects.

ChrisH

[RikM]

I tend to go for 5min at ISO1600 unless there are bright stars in the field when I will drop to 7min at ISO800. This is with an f/5 Newt.

[jgs001]

I tend to use 5 minutes at iso 800. Unless I'm using my RC at f/9 when I use 10 minutes. But it all depends on what I'm trying to image really.

Sent from my GT-I9300 using Tapatalk

[wxsatuser]

As Olly points out, the brightness of the sky will limit the exposure.

The best way is to set your exposure length with the histogram.

Get the histogram mountain somewhere between 20 and 40%.

The darker your sky the longer the exposure.

[Bizibilder]

I use 5 mins at ISO 800 and a LP filter.  It's not just the sky brightness that you need to consider - losing a 5min sub or two is no great loss but 10 mins could be.  With satellites and planes to contend with (as well as the neighbours "security" lights goinf on every time a cat walks past!) I find its better to stick with many short subs. The passage of small clouds can also lead to lost subs too.

[Davey-T]

Thanks for the info, I'll be using William Optic 110 FLT with f/r making it f5.6 and hopefully somewhere dark so should be able to leave the LP filter out for once , course it'll probably rain for a week so this may all be purely academic

Dave

[Physicist13]

If the question is "what is the optimum ISO" then you can easily find this out by doing test exp on the same object and comparing the signal and noise levels. For DSLR the opt is usually around ISO800-ISO1600, but varies slightly. Due to the non-linearity of the onchip processing higher ISO tend to give propotionaltly more noise than just what you expect from the gain increase. Obv limit your ISO such that you are not saturating any pixels (although saturating some stars is usually acceptible)

P

[Alien 13]

My camera is happiest in the ISO800-1600 range i tend to stick at 1600 in the winter months and 800 when its warmer.

Alan

[wxsatuser]

Tried the various ISOs but now going with ISO 800 on my 60Da.

[dph1nm]

Due to the non-linearity of the onchip processing higher ISO tend to give propotionaltly more noise than just what you expect from the gain increase

Do you have any evidence for this? Most studies I have seen of Canon cameras contradict it.

NigelM

[StuartJPP]

I would just experiment.

Olly

That is the best answer but I can add to it slightly, have a look at what other people achieve with similar kit and use that as a baseline, then experiment.

Asking a question like this always leads to debates Personally I try not to suggest settings because everybody has a different opinion. To me what counts is a captured/processed image and wanting to produce similar or better results, not analysing a graph of expected results.

It is akin to an audiophile looking at an oscilloscope rather than listening to the music and letting the ears decide...

[Physicist13]

Do you have any evidence for this? Most studies I have seen of Canon cameras contradict it.

NigelM

both my Canon 350D and 500D showed this, I believe C.Stark had similar results.

[Sam]

I found 5mins was the best trade off between increasing thermal noise and exposure in my 1000D. I used ISO400, basically because I read it on here somewhere.

[Pete Presland]

I use mostly 3 mins with my 200pds and 5mins with 80ed both at iso800. Light pollution stops much more than that, although processing can get rid of a lot sky glow. When i did my roesetta i could hardly see it in an idividual sub.

Sent from my GT-I9100 using Tapatalk 2

[Luke]

I'm more into webcam these days, but I normally used 5 minutes at ISO1600 with a 450D and 1100D and various scopes from about F5 to 7.5. I didn't find much of a difference between ISO800 and 1600, but that's just my own experience and not very scientific! I preferred 1600 partly because I could more easily check the subs on screen as they came in, with the details being brighter!

One thing I did convince myself of is that tons of very short exposures (e.g. 30s) did not give the same level of detail as a much smaller number of longer exposures (e.g. 5 mins) of equal total exposure time. I don't know if that is because with the short shots important detail is left low down in what I guessed may have been "read noise," a very streaky area of noise which I found impossible to pull details out of.

I'm no technical expert, I just experimented and chose what seemed to work for me. Half the time I have no idea whether something to do with imaging makes a difference or not, with multiple factors I just stick with what seems to work!!

[dph1nm]

Do you have any evidence for this? Most studies I have seen of Canon cameras contradict it.

NigelM

both my Canon 350D and 500D showed this, I believe C.Stark had similar results.

Having stated that ISO400 was optimal, Craig Stark rather back-tracked on this in the subsequent Cloudy Nights discussion. In the end he decided ISO 800 was probably better than ISO400, and there was little or no difference between 800 and 1600.

However, to get back to the original question, one should remember that changing ISO does not alter the sensitivity of the camera. So doubling ISO does not mean you can halve the exposure time and get the same result.

NigelM