sh2-HaSHO

[Rodd]

The experiment continues.  Here is the Hubble Palette image of this target to compare with the bicolor I recently posted.  I was only able to collect 32 SII subs, so I think it can be improved upon.  It is unlikley to happen anytime soon, however as the weather is turning quite bad and then the Moon will be back.  So, here's a start anyway.

TOA 130 with ASI 1600 and 3nm Astrodon filters

Ha 124 300 sec

OIII: 107 300 sec

SII: 32 300 sec

I binned 2x2 in software prior to processing, then I resampled up to be almost the same size as an unbinned image--4,200 pixels instead of 4,500 pixels wide.  The resolution of the original image was .78 arcsec/pix and the FWHM of the combined stacks was just over 2.0--so I did not lose any details I don't think.  

Question:  Do I need more SII?  I originally wanted about 120 subs of each channel.  60 more would suffice.....but maybe unnecessary?

One other thin--Usually I replace the stars in a SHO image with the Ha stars, but I did not do that here,.  That is why there are faint coloration in some stars.  No doubt I will rework this data quite a bit--but often the original is the best version....this is the original.

Thanks,

Rodd

 

[dave_galera]

Oh yes, I like that, perfectly handled and coloured SHO

[Rodd]

1 minute ago, dave_galera said:

Oh yes, I like that, perfectly handled and coloured SHO

Thanks Dave.  i was a bit worried about the limited amount of SII.  This hobby never ceases to surprise me

Rodd

[Kinch]

I don't think it needs any more data - It shows nice depth, the colour is really appealing and the stars are great. Overall -  a great job I would say!

[dave_galera]

You don't

40 minutes ago, Rodd said:

I was a bit worried about the limited amount of SII.  This hobby never ceases to surprise me

I don't think you need any more SII as anything above 35/40 subs is a law of diminishing returns and you are not going to improve noise reduction that much. A longer exposure time maybe, but not more subs.

[Star101]

Looks like a very nice image to me. 

Well done.

[Rodd]

1 hour ago, Kinch said:

I don't think it needs any more data - It shows nice depth, the colour is really appealing and the stars are great. Overall -  a great job I would say!

Thanks Brendan.  Your opinion means a lot.

Rodd

[Rodd]

56 minutes ago, Star101 said:

Looks like a very nice image to me. 

Well done.

Thanks Star!

Rodd

[Rodd]

1 hour ago, dave_galera said:

You don't

I don't think you need any more SII as anything above 35/40 subs is a law of diminishing returns and you are not going to improve noise reduction that much. A longer exposure time maybe, but not more subs.

Well--whats important is total exposure time.  There is a running argument between those who feel 1 30 min sub is better than 6 5 min subs.  In truth, they are pretty much the same with respect to photons (less noise in the stasck of 6 subs).  Photons come in a steady stream--it does not matter how they are collected becuase once you collect them you add them together.  It doesn't matter if photons are collected early in one long sub, or 6 days later in  short subs--becuase the photons are identicle.  There may be a tiny differences but those differences will strike the sensor using both appraches.  but for all intents and purposes, you can get just as much depth from a hundred short subs as youcan from 25 long ones (assuming equal total exposure).  I used to think the other way...but these high efficient, low noise CMOS cameras showed me different.  I used to think that somehow the photons striking teh sensor in a 30 min sub were different than teh photons stricking teh sensor in a 20 sec sub--they are the same.

As far as the diminishing point of return--in my sky that comes significantly past 30-40 subs.  For me its around the 70-80 mark using 300 sec subs with the ASI 1600 and this scope.  More if the Moon is out.

Rodd

[JamesF]

47 minutes ago, Rodd said:

Well--whats important is total exposure time.  There is a running argument between those who feel 1 30 min sub is better than 6 5 min subs.  In truth, they are pretty much the same with respect to photons (less noise in the stasck of 6 subs).  Photons come in a steady stream--it does not matter how they are collected becuase once you collect them you add them together.  It doesn't matter if photons are collected early in one long sub, or 6 days later in  short subs--becuase the photons are identicle.  There may be a tiny differences but those differences will strike the sensor using both appraches.  but for all intents and purposes, you can get just as much depth from a hundred short subs as youcan from 25 long ones (assuming equal total exposure).  I used to think the other way...but these high efficient, low noise CMOS cameras showed me different.  I used to think that somehow the photons striking teh sensor in a 30 min sub were different than teh photons stricking teh sensor in a 20 sec sub--they are the same.

If I've understood the argument correctly (which is not a given :), the idea (roughly) is that as well as the signal you collect, there are source of noise which increase with exposure time (such as background noise from the sky) and sources of noise that are fixed (eg. read noise in the camera).  Once the fixed noise is "drowned out" by the rest, your SNR is effectively fixed, so there's no point exposing for longer.

This means that if you have a light-polluted sky, you'll potentially reach your maximum achievable SNR more quickly than if you have a very dark sky when it takes much longer for the fixed noise to be drowned out.  It's therefore worth taking longer exposures under a dark sky, but if you live near a town you might as well take more, shorter ones.

CMOS cameras with very low read noise make it very easy to drown out, so are even less demanding of exposure time, but for "ideal" data collection, you'd still probably want longer exposures at a very dark site than at a light polluted one.

James

[Rodd]

2 minutes ago, JamesF said:

If I've understood the argument correctly (which is not a given :), the idea (roughly) is that as well as the signal you collect, there are source of noise which increase with exposure time (such as background noise from the sky) and sources of noise that are fixed (eg. read noise in the camera).  Once the fixed noise is "drowned out" by the rest, your SNR is effectively fixed, so there's no point exposing for longer.

This means that if you have a light-polluted sky, you'll potentially reach your maximum achievable SNR more quickly than if you have a very dark sky when it takes much longer for the fixed noise to be drowned out.  It's therefore worth taking longer exposures under a dark sky, but if you live near a town you might as well take more, shorter ones.

CMOS cameras with very low read noise make it very easy to drown out, so are even less demanding of exposure time, but for "ideal" data collection, you'd still probably want longer exposures at a very dark site than at a light polluted one.

James

There is no question a dark site trounces a light polluted one.  All I know is Bortle 6--M31...never, Milky way...nah.  Its frustrating.  thank heaven for narrow band.  I agree with what you say, execept your "exposure time" is total exposure time irrespective of individual sub length.  10 hours of eposure is 10 hours of exposure regardless of the number of subs (roughly....as you say)

Rodd

[JamesF]

Just now, Rodd said:

There is no question a dark site trounces a light polluted one.  All I know is Bortle 6--M31...never, Milky way...nah.  Its frustrating.  thank heaven for narrow band.  I agree with what you say, execept your "exposure time" is total exposure time irrespective of individual sub length.  10 hours of eposure is 10 hours of exposure regardless of the number of subs (roughly....as you say)

Oh, certainly.  The variable is what SNR you have achieved against what you might possibly achieve.  The potential SNR is higher at a dark site, and that's what makes the longer exposures worthwhile there, if you want to chase that extra potential.

James

[Rodd]

2 minutes ago, JamesF said:

Oh, certainly.  The variable is what SNR you have achieved against what you might possibly achieve.  The potential SNR is higher at a dark site, and that's what makes the longer exposures worthwhile there, if you want to chase that extra potential.

James

All I am saying is at a dark site, long exposures, or many short exposures will result in the same thing.  Possibly more if you consider that one can throw away 3 5 min subs and keep 3, but you either have to throw out a 30 min sub or keep it--good and bad.  If a cloud passes over and 15 of the 30 min was not as good--you lose either way--if you toss the sub you lose the good 15 min and if you keep the sub you gain the poor 15 min in your stack  But this is another issue, but it relates to the first.  Assuming a perfect sky, say on the Moon, the difference between 1 60 min sub and 10 6 min subs would be negligiable.

Roddd

[Rodd]

I couldn't help myself.  A few tweaks--nothing major, but the detail and contrast in the nebula is a bit better. 

[JamesF]

4 minutes ago, Rodd said:

All I am saying is at a dark site, long exposures, or many short exposures will result in the same thing.

I think the problem here is talking about exposure times as if they're what leads the argument.  What determines how much useful signal you have is the SNR.  The better the SNR, the more data you have to play with.  The SNR is maximised when the sources of noise that increase over time drown out the fixed noise.  At a site where there is very little time-related noise, that takes much longer to happen.  Once it does happen, there's no point going for anything longer.  Prior to that point, longer exposures will give a higher SNR than short ones.

So, if for some combination of kit, the maximum SNR at a specific location is achieved at exposures of 4 minutes, using ten six-minute exposures or six ten-minute exposures will make no difference.  But ten three-minute exposures will be better than fifteen two-minute exposures because the fixed noise makes a bigger contribution to the SNR.

This is actually easier to show mathematically than explain in English :)

So the real answer to "Are shorter exposures as good as long ones if the total time is the same?" is "It depends." :D

James

[Rodd]

19 minutes ago, JamesF said:

"It depends." :

Naturally--like almost everything in this hobby.  One of these days I will have to get to a Bortle 2 site.

Rodd

[graemlourens]

@Rodd Help me if i missed that somewhere: Whats the designation of this target?

Great picture.... very nice colours! You did a great job.

[Rodd]

6 minutes ago, graemlourens said:

@Rodd Help me if i missed that somewhere: Whats the designation of this target?

Great picture.... very nice colours! You did a great job.

Thanks Graem.  I know it as sh2-132.  I am sure it has another designation, maybe an IC number or NGC number. Some call it the Lion Nebula.

Rodd