DSS Help

[Alien 13]

I have seen a few posts that mention that the final stacked image in DSS being very dark but mine always get very bright even to the point where its completely white i am using the default settings but not sure why this is happening any ideas?

The second question is how do i add several sessions into an image or do i have to stack the whole lot manually?

I do have a lot of LP in my area so all my subs are very bright is this a factor with the first question.

Thanks in advance.

Alan

[lensman57]

I have seen a few posts that mention that the final stacked image in DSS being very dark but mine always get very bright even to the point where its completely white i am using the default settings but not sure why this is happening any ideas?

The second question is how do i add several sessions into an image or do i have to stack the whole lot manually?

I do have a lot of LP in my area so all my subs are very bright is this a factor with the first question.

Thanks in advance.

Alan

Hi Alan,

There could be many reasons for this , LP is certainly one of them particularly if the level is high and you are not using an LP filter and the subs are just too long so that the sensor is satureated with the skyglow. The other problem is inclusion of bad frames with clouds running across them, even the slightest mist will cause the sky background value to shoot up so examine the frames before stacking and get rid of the bad ones, junk in - junk out, be ruthless. Also try and not save the stack with adjustments appied, you can stretch it later.  I am not sure what you mean by stacking several sessions together but if you look at the bottom of the page you see that once one set of subs are loaded DSS will give you another grouping option, group1, you can load the second subs into the group1 and DSS will stack the lot together, the other way if the subs are of different  length is to first satck one lot, open the second lot and and uncheck the lights from the first stack but right click on one of the first lights and mark this as a reference frame, then stack the second lot and these will be aligned to the reference frame. This is useful if you for example want to align a stack of short exposures to a long exposure for an HDR treatment of lets say M42. I don't think that DSS has the ability to rescale the subs or compensate for meridian flip as NEB3 or PI or Registar. Hope that this helps.

Regards,

A.G

[wxsatuser]

I assume you check your histogram for each session?

Try to keep the 'mountain' between the 20/40% marks on the histogram.

Sky brightness is going to be your limiting factor.

If your sky is very bright you will no doubt have to have shorter exposures.

Shorter exposures will need more exposures.

If you have'nt got a good LP filter, get one.  I use a Hutech IDAS P1.

[Alien 13]

I assume you check your histogram for each session?

Try to keep the 'mountain' between the 20/40% marks on the histogram.

Sky brightness is going to be your limiting factor.

 

If your sky is very bright you will no doubt have to have shorter exposures.

Shorter exposures will need more exposures.

 

If you have'nt got a good LP filter, get one.  I use a Hutech IDAS P1.

Thanks for the reply, my histograms tend to be between 60-80% with a 45 second exposure (f4.7) and 90-95% on the outputed DSS tiff which as you say is due to LP but unfortunately most of the localised LP is white LED street lamps.

The individual subs although bright do have a lot of detail and with M42 there is a lot of vivid colours too most of which gets washed away during processing by DSS.

I guess im hitting the limits of my back yard astro work but i will be trying a LP filter to remove some of the distant orange glow from the city.

Alan

[lensman57]

Thanks for the reply, my histograms tend to be between 60-80% with a 45 second exposure (f4.7) and 90-95% on the outputed DSS tiff which as you say is due to LP but unfortunately most of the localised LP is white LED street lamps.

The individual subs although bright do have a lot of detail and with M42 there is a lot of vivid colours too most of which gets washed away during processing by DSS.

I guess im hitting the limits of my back yard astro work but i will be trying a LP filter to remove some of the distant orange glow from the city.

Alan

I like yourself am suffering from this new menace, called white light pollution in particular since our neighbour a doctors surgery has now prepared for Sunday opening according to  the government wishes and installed a dozen lights to keep the neighbourhood lit like MU FC football ground. I am afraid that LP filters will not help you much. Your best bet is to try UHC-S CCD imaging  fiters for nebulea as these might prove more effective with the caveat that the exposure times will be longer. The idea of keeping the histogram to about 20% may or may not be helpful as in an effort to attenuate the LP you are also attenuating the signal so you have to find a compromise point where LP is kept to a minimum but with affecting the signal to the least amount possible, not easy to be honest. Please keep us informed as to how you progress as there is a lot of interest in this topic.

Reagrds,

A.G

[wxsatuser]

What your looking for is correct exposure, I'm not trying to reduce LP with the histogram setting.

Somewhere between 20>40% should be the correct exposure for your sky conditions and camera settings.

This may vary night to night and for different sites.

BTW

I suffer from Sodium lights plus football floodlights and leisure centre court flood lights.

It's not perfect but the Hutech does a good job in my location.

[Sandancer10]

Hi Allen, just a thought but have you tried to move the RGB channels on your histogram in DSS. Pulling them to left will darken your image.

[MikeMc]

Hi Allen,

I am new to all this but my DSS stacks range from very bright, to very 'orange' to very dark, I am still doing loads of tests (I may actually put them up at some point?).

This was my first go at stacking mutiple groups of different exposures (30, 60 and 120 secs at 800 ISO) - using the group method mentioned above and most of mine look like this. 

Initial stack in DSS with no processing and with adjustments 'applied' when imaged saved - that is another thing! 

How does that compare?

[lensman57]

Hi Allen,

I am new to all this but my DSS stacks range from very bright, to very 'orange' to very dark, I am still doing loads of tests (I may actually put them up at some point?).

This was my first go at stacking mutiple groups of different exposures (30, 60 and 120 secs at 800 ISO) - using the group method mentioned above and most of mine look like this. 

Initial stack in DSS with no processing and with adjustments 'applied' when imaged saved - that is another thing! 

combination 30, 60 and 120 sec ver 1 no processing.jpg

How does that compare?

Hi,

I have taken the liberty of removing the LP by software,  to the best of my ability,  and this is the result. It is not ideal as it leaves a lot of gradients to be taken out and during the process it also takes a lot of the detail out too even with masking.

Regards,

A.G

[MikeMc]

Wow that was quick!

Nice one, It took me several days and various attempts (this was ver 7) to manage the following in PS,

I am starting to compare all the different methods and processes but it is a very steep learning curve, keeps me busy :-)

[Alien 13]

Thanks everyone for the replies so far and everything will get taken on board but dont get me wrong i can still manage to pull out reasonable data even when the sub is over exposed but it does loose all the colour impact of the original subs.

To be honest i am amazed sometimes even with bad LP and 10x 45 second subs what comes out at the end ive even considered batch processing the RAW files to give them a better exposure setting prior to DSS.

I still have to learn the technique of using multiple layres to add detail to the over exposed bits like the core and maybe a seperate layre with the colour information but its fun learning this stuff.

What my practice efforts have so far shown is you can do astrophotography with heavy LP so what i suppose i realy need now is lots more data and go for it.

Here is a sample of progress so far with settings mentioned earlier. Alan

[Alien 13]

Hi Allen,

I am new to all this but my DSS stacks range from very bright, to very 'orange' to very dark, I am still doing loads of tests (I may actually put them up at some point?).

This was my first go at stacking mutiple groups of different exposures (30, 60 and 120 secs at 800 ISO) - using the group method mentioned above and most of mine look like this. 

Initial stack in DSS with no processing and with adjustments 'applied' when imaged saved - that is another thing! 

combination 30, 60 and 120 sec ver 1 no processing.jpg

How does that compare?

Thats pretty much how my subs look before DSS and its worse when stacked still i think a lot of the data is in there somewhere.

Alan

[wxsatuser]

Yes there is no doubt you can get reasonable data out of a bad site but get every thing on your side.

Compared to a darker site you will just need more total exposure to get the same quality.

If your site is twice as bright as a dark site you will need twice the total exposure.

[MikeMc]

Nice M42,

I know what you mean about the colour 'loss', I am still having a few problems with my M42 but have found that doing the RGB curves in PS separately tends to bring the colour out a bit.

I also followed this to start with,

http://flintstonestargazing.com/2009/06/26/my-quick-deepskystacker-tutorial/

The first part deals with stacking and the second some basic post processing in DSS.  -  I know this has caused lots of discussion but as a starter it can show you what sort of data you have, it has a good example of a very bright DSS stack and you can always save the initial stack before you mess with it in DSS.

I am trying not to do this so much and work on my PS skills but I still get my best M42 by doing at least some messing with the colours and saturation in DSS.

30 x 30 sec @ 800 ISO (Lights)

23 Darks

25 Bias

Some DSS post processing (as per the above link)

Additional work PS, mainly levels & curves

[Alien 13]

Nice M42,

I know what you mean about the colour 'loss', I am still having a few problems with my M42 but have found that doing the RGB curves in PS separately tends to bring the colour out a bit.

I also followed this to start with,

http://flintstonestargazing.com/2009/06/26/my-quick-deepskystacker-tutorial/

The first part deals with stacking and the second some basic post processing in DSS.  -  I know this has caused lots of discussion but as a starter it can show you what sort of data you have, it has a good example of a very bright DSS stack and you can always save the initial stack before you mess with it in DSS.

I am trying not to do this so much and work on my PS skills but I still get my best M42 by doing at least some messing with the colours and saturation in DSS.

30 x 30 sec @ 800 ISO (Lights)

23 Darks

25 Bias

Some DSS post processing (as per the above link)

Additional work PS, mainly levels & curves

M42 - Orion - 1st stack DSS & PS processing 30 L 23 D 25 B 30 sec 800 ISO.jpg

Like your M42 by the way im going to try and tackle the loss of colour with mine by selecting 1 or 2 subs that have lots of colour in them then use the noise tool in PS to remove the stars (well as much as i can) so im left with a colour only layre that i can blend in.

I too have found that i get better results by using the basic level control in DSS to center the histogram before sending to PS.

[dph1nm]

If you have RGB Channels Calibration set  in DSS (which you should have to avoid an orange background, and so you can use sigma clipping) then the resulting colour will be very destautrated (says so in the DSS manual). You can get it back in DSS by upping the saturation slide to 20% or so.

NigelM