The EQ3 DSO Challenge

[Nigel G]

51 minutes ago, The Admiral said:

Failed indeed! What file did you put where? Bias frames taken at same ISO? Was it totally blank?

Ian

I registered my lights, flats and bias as normal

Then took the master bias .tiff file- renamed it 2 times and entered that as a dark and dark flat. So when stacking I had 70 light, 1 dark ( renamed master bias .tiff), 30 flat, 1 dark flat ( second renamed master bias .tiff file ) and 50 bias.

The outcome is a 206 mb black image.

I have since re stacked using the above without the dark. still blank.

I am now stacking without the dark flat input but with the dark file. Edit ( failed, full maximum stretch showed a few stars )

What is going on !

Edit: all 800 ISO

Nige.

Edited March 31, 2017 by Nigel G

[Stub Mandrel]

This diagram off the DSS ssite may help understand how the different frames are used.

 

From: http://deepskystacker.free.fr/english/index.html

Click the tab 'how to create better images'

[The Admiral]

31 minutes ago, Nigel G said:

I registered my lights, flats and bias as normal

Then took the master bias .tiff file- renamed it 2 times and entered that as a dark and dark flat. So when stacking I had 70 light, 1 dark ( renamed master bias .tiff), 30 flat, 1 dark flat ( second renamed master bias .tiff file ) and 50 bias

Nige.

Now I'm confused. You shouldn't need the bias as well, surely? I think I would have just used the separately stacked bias instead of the flats dark, and a copy, re-named, as the dark. In AstroArt, I only have catrgories for flat dark and dark, both of which I replace by the master bias.

Ian

Edit. Can you mix raws and tif files in DSS? Try using DSS to stack the bias frames (raws?) and save as a fits. Dunno, might help!

Edited March 31, 2017 by The Admiral

[Nigel G]

Give me a bit of timber and a knife and I'll make anything. Fits, tiff's, dithering, drizzling, flats, darkflats, darks, bias, my brain gets a bit like a scope out of focus. blurry.

I'm now trying what I know works. just lights flats and bias. To see if I have a problem elsewhere.

Nige.

[The Admiral]

Nige, look at the Alternate Calibration Process II. Unfortunately, it is a bit convoluted to find. Use the link, go to 'Users Manual', go to 'lights, flats,......how to make them', then click on the first 'here' link.

Sensible approach, do it the simple way first . Interesting to hear what you get.

Ian

[Nigel G]

40 minutes ago, The Admiral said:

Nige, look at the Alternate Calibration Process II. Unfortunately, it is a bit convoluted to find. Use the link, go to 'Users Manual', go to 'lights, flats,......how to make them', then click on the first 'here' link.

Sensible approach, do it the simple way first . Interesting to hear what you get.

Ian

Got it  so I enter the master bias file which DSS saved as a tiff, as a dark and a dark flat but don't use the bias frames to stack.   I guess DSS does mix files as it saves master files as tiff's. All my input frames are raw.

Edit: I have entered my 70 light, 1 dark, 1 dark flat and 30 flat. DSS is saying don't set the black point to 0 ....not sure what to do so I'm going to try both.

Edited March 31, 2017 by Nigel G

[mikey2000]

I from reading that DSS page, I got the idea that it is perfectly ok to use lights, darks, flats and bias.   I've not tried a dark flat frame yet.

 

i also got the idea that bias masters are OK to reuse time and again.   Darks are easy - just put the cap on the scope and let the camera fire away while I'm tidying up, having final cup of Tea etc. Then  I make the flats the following morning (takes only a few minutes)

 

Alternate Process 1 says that will produce a proper calibration.

 

so, is there a benefit to dark flats?  I imagine they are quite quick to make too.

 

i shoot raw on my camera  but use the TIF master bias file.   When registering all the lights/darks/flats, it also saves a master dark TIF and master flat TIF   I then rebuild the file list with the raw light files and TIF dark and flat master.   It saves a lot of computing time if you want to rerun the stacking with different parameters....

 

 

 

[Stub Mandrel]

Those who read the depths of the DSS documentation will find that DSS keeps a record of all the frames it uses for a stack, and looks to see if the matching master TIFFs for the provided control frames already exist, and if they do it will use them instead of rebuilding them.

[StormAurora]

Ok don't laugh!

My first attempt tonight with 150pds on eq3 pro mount and canon 100d.

M13 and the Orion nebula, straight from the camera, no processing.

I'm really happy with them considering I've never done astrophotography before ?

Couldn't get the synscan to work but a good night despite that.

 

[Stub Mandrel]

A good start @StormAurora!

How long are you exposing for?

If you shoot RAW and stack (plenty of advice on doing this here) you will soon start pulling out loads more detail.

[wimvb]

I agree with Neil, that's a good start. The easiest way to proceed from here is to expose more frames, download deep sky stacker to combine them, and use whatever image processing software you have to increase contrast.

Good luck, and have fun

[The Admiral]

17 hours ago, Nigel G said:

Got it  so I enter the master bias file which DSS saved as a tiff, as a dark and a dark flat but don't use the bias frames to stack.   I guess DSS does mix files as it saves master files as tiff's. All my input frames are raw.

Edit: I have entered my 70 light, 1 dark, 1 dark flat and 30 flat. DSS is saying don't set the black point to 0 ....not sure what to do so I'm going to try both.

What was the outcome Nige?

Ian

[Peco4321]

Went to a dark site last night, my sisters house, I was mainly getting shots of her house with the stars in the background as I have been promising her it for a while now, but before packing up, I got my scope set up as I noticed Vega was out therefore the Ring Nubula, and also Hercules.  I got the Ring Nebula last year and M13, but since then I have motors and the improvement is great (to me anyway!).  Only about 5 mins of 30 sec exposures on each, ISO 800, as it was getting late, I will be back for more soon, dark sites with wide open horizons are amazing.  Added darks, bias and flats for once! Stacked in DSS and stretched a bit in GIMP.

[Stub Mandrel]

1 minute ago, Peco4321 said:

Only about 5 mins of 30 sec exposures on each, ISO 800, as it was getting late, I will be back for more soon, dark sites with wide open horizons are amazing.  Added darks, bias and flats for once! Stacked in DSS and stretched a bit in GIMP.

That's the way to do it! Better results than I was getting with low numbers of subs, but I have more LP.

Get your polar alignment spot on, go up to a minute and buy an intervalometer off eBay for about a tenner - then you can go in and drink tea/read SGL etc.* while it takes a hour or two's subs!

 

*I know you are supposed to be standing next to the scope with your bins or something....

[Peco4321]

And another one from last night, M92, again, only about 5 mins of lights.

[Nigel G]

8 hours ago, The Admiral said:

What was the outcome Nige?

Ian

Ian, I made several stacks.

1 Normal stack - Lights, proper darks, flats and bias.

2 Lights, flats & bias.

3 Lights, master bias as a dark, flats & master bias as a dark flat.

4 Lights, master bias as a dark, flats and bias.

5 Lights, master bias as a dark, flats, master bias as a dark flat & bias.

Any stack that had bias stacked with a copy of the master bias I got a blank image except for number 4 where I had only the brightest stars visible after full stretch.

Number 1-2 & 3 stacks all produced a good image which I think number 1 being very slightly better maybe.

Number 3 stack DSS suggested not to have the black point set to 0. I tried both and setting the black point to 0 worked.. Unchecking the black point to 0 tab produced no image.

This DSS cheat has not worked for me  but I know I'm not hurting my images with darks and I can get away with no darks.

3 images #1 L-D-F-B      #2 L-F-B       #3 L-MBd-F-MBdf. All background extracted and auto stretched only in PixInsight.

Nige.

[The Admiral]

15 minutes ago, Nigel G said:

Ian, I made several stacks.

1 Normal stack - Lights, proper darks, flats and bias.

2 Lights, flats & bias.

3 Lights, master bias as a dark, flats & master bias as a dark flat.

All background extracted and auto stretched only in PixInsight.

Nige.

Strange isn't it? As you know I've moved from DSS to AstroArt in order to stack my Fuji RAFs directly without having to convert to DNG, so I haven't tried any of this bias for darks malarkey in DSS, but it does seem to work fine in AA, and saves me a lot of time. I can't see any difference between your three outputs, what do you think?

Aah, I see you've taken the plunge with PI! Now you really have moved over to the very, very dark side .

Then again, if you're using PI, why are you stacking in DSS?

Ian

Edited April 1, 2017 by The Admiral

[Nigel G]

8 minutes ago, The Admiral said:

 

Aah, I see you've taken the plunge with PI! Now you really have moved over to the very, very dark side .

Then again, if you're using PI, why are you stacking in DSS?

Ian

Ian. I am trying PI  with the 40 day trial  not quite moved completely to the dark dark side  

I have not tried stacking with PI yet, ( the old "I know DSS " thing )

15 minutes ago, The Admiral said:

. I can't see any difference between your three outputs, what do you think?

 

I think on very close inspection there's a fraction less noise in the first image which is my normal stack, and a fraction more gradients around the edge in the 3rd but really not much in it , It obviously works for some but not all. If I had no dark frames I think I would just stack L-F & bias. I might try it with another image, I'm swapping cameras to the unmodified for a few images so maybe that might be different.

Nige.

[Stub Mandrel]

I downloaded APlab yesterday. I analysed a couple of 2-minute images at iso800 taken in single-figure ambient temperatures. It worked out that my read noise (bias frames) was five times that from my darks.

That means the Bias frames are more important that darks up to 10 minutes of exposure at such temperatures.

That probably explains a lot!

I've also found that LOTS of darks (50-100) appear to make make more difference to noise levels than exact temperature matching.

[wimvb]

4 hours ago, Nigel G said:

Ian. I am trying PI  with the 40 day trial  not quite moved completely to the dark dark side  

I have not tried stacking with PI yet, ( the old "I know DSS " thing )

I think on very close inspection there's a fraction less noise in the first image which is my normal stack, and a fraction more gradients around the edge in the 3rd but really not much in it , It obviously works for some but not all. If I had no dark frames I think I would just stack L-F & bias. I might try it with another image, I'm swapping cameras to the unmodified for a few images so maybe that might be different.

Nige.

Since you're testing PI, here's a tip from a dark lord:

PI has a noise evaluation script, and you can do image statistics (process -> image -> statistics), which you can use to evaluate the three stacked images. The difference between the images should be in the background noise, and any hot pixels. You should also examine the stacks for any "walking noise" or streaks left over after stacking.

If you want to test stacking in PI, I suggest you use the batch preprocess script, since you're only on the trial version.

The feature that should win you over to the dark (dark) side is DBE. Crop the image, apply DBE, and see the magic force.

[Peco4321]

School boy error last night on the Leo Triplet, I forgot to change to RAW!!  Looks like I had polar alignment quite good as well looking at the middle star.  Anyway, stacked 20 lights, 30 sec ISO 800, darks and bias, no flats.

[wimvb]

Been there, done that.  Nowadays my camera is always on RAW. Even when used as family camera. (Which it is less nowadays, as mobile phone cameras are so much easier.)

Nice pic, btw.

I'm not sure that calibration frames make sense when shooting jpeg, since the calibration process needs single pixel information, which is lost in the conversion process. Flats may work, removing vignetting. Bias and darks may deteriorate the lights due to added noise.

There's only one way to find out ...

[Nova2000]

I once was awake all night shooting constellation and when I went home 120km away I realized they were jpeg!!!!! 

[Stub Mandrel]

Owl Nebula and Surfboard Nebula from last night using the 130P-DS, tricky one up near the zenith so after seeing that about half the 120s subs were spoiled I went to 60s ones, in the end I was able to stack 1 hr 38 minutes out of just over two hours data.

If it's clear tonight I might use the 135mm lens on the Virgo Cluster.

[mikey2000]

I've had some kind of breakthrough with my guiding.   I can now reliably control RA to within +/-2" and have and RMS of <1 for extended periods!  With a half moon up, I thought I'd look for a globular cluster.

 

I stumbled across NGC2419 the Intergalactic Wanderer in Stellarium.  I was inspired by the image on the wikipedia page https://en.wikipedia.org/wiki/NGC_2419 and thought I'd give it a go.  It turns out, it is *very* far away and consequently small and faint.   This is a heavy crop from my aps-c sensor...

Sadly, I missed focus slightly (close inspection shows doubled up spikes on the bright stars) and my processing seems to have 'ringed' the bigger stars. 19x4mins  I just realised I included a single sub that went wrong.  oops!