Pastel stars?

[DJT]

I was interested in seeing that part of the Malins focus this year was on Pastel Stars.

So what does that mean?

Clusters can be pretty mundane to some and many of us tend to push saturation to give them a bit of a pop so as part of a project I am doing this year on clusters I am holding back and relying on PI for photometric calibration and not touching saturation at all.

Heres a couple I have done with minimal processing and in both cases the images were captured in Sydney at new moon.

M12 and M4

Dynamic crops
Photometric calibration including background neutralisation
HST
SCNR
MT

Stars are "fat as" due to shocking seeing this year so a little MT was applied.

so..Pastels..Hot or Not?

GSO RC10
Moravian F2
Astronomiks RGB filters

[Placidus]

Hi, David,

Thanks for the interesting and thought-provoking question. Here's what I think:

Our 20 inch scope collects 6400 times the photon flux of the human eye. With a one hour exposure, it's got perhaps 54,000 times the integration time of a living-room adapted human eye, and still a thousand times that of the darkest-adapted eye. So we are increasing the total number of photons collected by 64 million times. Dr Malin might think that is artificial, and cheating.

The scope has a resolution limited only by seeing, at say 2 sec arc, whereas the living-room adapted eye has a resolution of about 60 sec arc, 30 times worse. The dark adapted eye's resolution is pathetic. Using the scope is starting to sound like a disgusting conspiracy to pervert the beauty of the eye. Perhaps David might like to throttle down the aperture of the AAT to a quarter inch, to make the images more natural looking.

If Dr Malin uses a cheap spectrometer, he can distinguish colours separated by a few nanometers, and a good spectrometer hilariously better again (down to a femtometer is commercially available) whereas colourbind Mike can barely distinguish red from green.

As to the use of unsharp masking, well that is definitely unnatural, bordering on cheating.

I think it is completely sensible to enhance the colour differences to the point where they are easily distinguishable without getting a headache. It is more than sensible, it is a moral obligation to enhance them to the point where they are telling a valid story about the astrophysics, for example showing that the core of a galaxy is generally more orange, and the OB regions in the spiral arms generally more blue, or for example that ellipticals are generally more orange than tidally disrupted spirals.

But to go beyond that, to the point where they are lurid, like a National Flag, is at best art, and not necessarily even good art.

So I would argue strongly for artificially increasing the saturation just to the point where we are saying something sensible about what is up there, but no further.

If Dr Malin's definition of "pastel" is "no greater colour distinction than that of the dark adapted (but in my case colour-blind) human eye", then he is being wildly inconsistent and unscientific, making a personal aesthetic choice that is totally unjustifiable. He should set down his thermometer, and use only the palm of his hand; he should set his spark plugs by eye; he should diagnose a heart attack on symptoms and not cheat using an electrocardiogram or cardiac enzyme analysis. He should drive using only his gut feeling of how fast he is going, and avoid the unnatural distinctions that Constable Bookem might use with his Lidar.

Of course, they are the Malin Awards, not the Mike Awards, and he is perfectly justified in making a personal aesthetic choice, but we must not fall into the trap of thinking that it is based on logic, or that it is the last word, or that they should become official IIS Deep Sky Policy.

I personally shall continue to avoid lurid like a butcher's apron, lurid like a kindergarten landscape, lurid like a national flag, and aim for saturated enough to tell the one, true story about what is up there.

Very best,
Mike Berthon-Jones

[Placidus]

…. and …..

they are two beautiful globular clusters, beautifully photographed.

Best,
Mike n Trish

[Atmos]

I’ve imaged both of these at 600mm and this tells me one thing... I do need more focal length haha

They look great but I’m with Mike, a little more saturation as it helps distinguish between the different star colours. There are many variables that go into what you would see through a telescope. One night with the same equipment can have pastel colours while the next night due to better seeing and transparency leaves rich and vibrant colours.

[DJT]

Snort..I read the original unedited response with the request for a tad more saturation. Always go with your gut. It’s invariably right.

I really liked the big response. Well thought through. To your point, everyone has different takes on this stuff .


Without the saturation, as you and Colin says, you loose the distinction between the different stars and their temperatures and you end up pixel peaking to pick them up. No great data here so that brings about its own problems.


Thanks again.

[Ryderscope]

Thanks as well David for starting a thread on this topic. It is always bound to start a vigorous conversation on aesthetics v science and where the two meet. I think that MnT’s response is on the money when you think about the practicality of what we are doing with the photons from the point of collection until they appear on our screens. I the case of star colours I am of the view that it is ok to push the saturation such that one can easily identify the differences but without overdoing it.

[strongmanmike]

Very interesting Dave and Mike

You might like to be at Star Stuff II next Sun 8 July, to hear David's take on this sort of thing, here is the abstract for his presentation :):

11:15am – Dr David Malin
Fake photographs, False Colour – The ethics of photographic image manipulation are ill-defined, and any working definition depends largely on the genre of the images and the intent of their creators. Astro-photographers are the most versatile and knowledgable of image makers, but are faced with the fundamental problem that they are generally unable to see the colours of the objects they acquire. I intend to explore this murky topic a little and suggest some guidelines, implicitly recognising that astro-images reveal an unseen part of the natural world whose colours are unseen, but not unknown.

Both globs look ok to me Dave, I'd perhaps push the saturation a tad more but they look fine as is too :thumbsup:

Mike

[Andy01]

“But to go beyond that, to the point where they are lurid, like a National Flag, is at best art, and not necessarily even good art.”

Wonder what the critics & art community thought of Picasso, Van Gogh, Hockney & Pollock when they first appeared on the scene?

Imo we’re not bound by science when creating our images.
Respect it by all means but the outcome & process depends on one’s intent.
Ie:are we using and interpreting the AP medium to create pretty pictures/ dynamic images or attempting to generate a scientifically accurate record?

Each to their own I guess. :D

I would be very interested to hear Dr. Malin’s take on this :)

[multiweb]

That's why it's called astro photography and not wedding photography. :P

[Paul Haese]

I think as conveyors of science the photography we should be fairly accurate but unless you are contributing images for scientific analysis it does not matter much.

Have said that, if you enter David Malin's competition you need to play by his rules and wants. Having conversed with him on several occasions now (he comes to visit ASSA a bit) his greatest criticism of amateur images is that galaxies are not purple, they are blue and soft yellow and this is driven by pastel stars. I interpret this to mean he likes to see subtle saturation on stars, but is not averse to saturation on things like nebulae. There might be bright nebulae within galaxies that are shown as red or magenta. Given his back ground you can hardly argue with him but I gather he is not always agreed with even in professional circles. So, if you want to pick up anything in his competition you have to produce his type of images, whether you agree or not, and even then you might not actually get a win. If you're not interested in entering then you are free to take on whatever level of saturation you like.

[Peter Ward]

Ah...this old chestnut.

I agree with Marc on this one. Your first, and last job, as a wedding photographer would be when you show the bride your interpretation of her complexion as green and make her white dress a lovely pastel.

Of course you can keep the hues accurate and bump the chroma up a tad (like commercial TV) and still keep the punters happy. I have no problem with RGB astrophotographers doing the same. There is clearly a continuum ranging from accurate colour to Las Vegas neon, but no surprises that some people like a refined single malt, and some like slabs of XXXX.

With narrow band imaging things change, as you are no longer dealing with the same wavelengths, hence to better show the interplay of ionised gasses remapping to another part of the spectrum is fair enough. I’ve noticed there seem to be fashion trends on how best to do this. Hubble palettes have been slowly replaced with gold green and blue mapping. I have no doubts some other combination will be more popular in years to come.

That said, the underlying details remain important: composition and framing, resolution (focus!) , lack of noise, dynamic range and no induced processing artefacts...eg you wouldn’t want to emphasise dark circles under the bride’s eyes in her otherwise glowing portrait....just as stars should not have orbiting black rings either.

[gregbradley]

Nice shots but the stars aren't pastel.

Pastel has an English definition:

a soft and delicate shade of a colour.
"the subtlest of pastels and creams"

Examples:

https://www.bing.com/search?q=examples+of+pastel+colours&form=EDNTHT&mkt=en-au&httpsmsn=1&refig=71564407591d4abfefc846e6a3b0d 86d&sp=-1&pq=examples+of+pastel+colours&sc=8-0&qs=n&sk=&cvid=71564407591d4abfefc846e6a3b0d8 6d

I don't see what the big deal is. David is not saying no colour when he says pastel. He means pastel per that definition. Sounds reasonable to me.
You can see the colours of some bright stars at a dark site visually. So there is that as a reference point. Keeping in mind the ones you can see are the top largest and brightest stars and not necessarily usual stars.

We can also see our sun. During the middle of the day it appears mostly white but at times early in the morning it is really yellow. Its not garish yellow but more subtle overall. But its definitely yellowish. Morning first rays can be more of a shade of gold.

An example of not pastel are in images from about 5-10 years ago say from R Jay Gabany (who had a run-in with David on this colour subject up at the Gold Coast about 5 years ago). Quite often those early pioneer imagers made their stars pinpoints and quite colourful. Don't get me wrong I admire R Jay's many fabulous images and he has his own distinct "style" as does Tony Hallas.

In your Glob images these stars are usually old and thus often yellowish which isn't really portrayed so you are not there yet in terms of colour.

Some optics give chromatic aberration and put blue rings around bright stars that aren't blue.

This is similar to the argument about colour temperature for your camera doing nightscapes. One person advocates using daylight white balance. If I do that on my Sony the images look artificially yellow/brown. Auto white balance though works well on Sony and Fuji cameras as well as Nikon.

Greg.

[Shiraz]

well said - agree entirely with this. Have sometimes found it necessary to saturate beyond the point of comfort to emphasise a particular feature in an image - not ideal, not for a purist, but necessary to get across a point. Cheers Ray