I've been working on this theory for months, but I had to wait till I had clear skies under the full moon. Then we had a large high pressure system making for great skies under a super moon. The super moon was the worst possible condition, which was perfect for final experimentation.
The black point cuts off little other than the low luminosity where noise tends to be found. The mid point cuts off nothing because the white point icon is on the right side of the histogram and is unmoved. This method merely sets a black point about 0.1% into the light curve. Pop over to the Sky Story Astrobin and check out all the images that have been "killed" using this method. Everyone of them has been made using this method. www.astrobin.com/users/SkyStory/
I hope this comment comes out the way I intended it to. I've been watching your videos ever since you taught Affinity photo many years ago. Your brain is somewhat sooo advanced then us laymen people and you're so fast, that I have to replay your videos several times to get the info to stick into my lil brain !! LOL. But once I get what you're saying, everything works great. My processing of my astro images have never been better. This histogram video especially. I've always had trouble stretching my images, even tried many different methods, but this one works the best by far. I also need to learn Affinity photo better too. One thing that also distracts form my learning is the background music you have running while you're talking. If you feel you need it, please make it quieter. Thanks and keep up the helpful videos. Your future one of the full moon will be helpful, plus I live in a bortle 7 sky area, so I hope the video includes how to get rid of the effects of light pollution. Clear skies..
Thank you! Be aware that there is no one size fits all histogram strategy, but this one usually works. However, if I am shooting a DSO that has very bright and very dark components, I will do two stretches. For the bright part, I will place the black point just left outside the light curve, and the mid point will be adjusted till the bright information looks good. Then a second light curve is run on the dimmer data using the Evolved method. The problem with background sound is that it's all digital these days, and some speakers or headphones will emphasize back channels more. In my headphones, the music is louder. On my very high quality test speaker set, the background sound (including music) can barely be heard. I will turn it down a little and see if that can reach a better average for different users.
It is only a tiny fraction of a single percent of information that is clipped, and only at the base of the effective light curve. Around that point, you are often in the range of noise and camera offsets so it's barely a loss that matters. In fact, it can serve to clean an image.
A couple months ago, I posted a complete layer-based workflow playlist from start to finish. Hope this helps. ruclips.net/video/ru_WvGhiDm8/видео.html&pp=gAQBiAQB
I am not sure if you have tried using the gradient mosaic tool in pi. I made some haoiii, siioiii and rgb images but the ho and rgb, had some artifacts on 2 stars on side bright the other dark. Is there some other tool you would recommend, as I played with some settings but those stars where always strange. SO looked fine. I did process this anyway as a first trial for a mosaic and overall really like it, aside from those 2 stars. I will upload this to astrobin later today of the Heart + soul nebulas.
Because I live at a dark sky site, I almost never have to deal with gradients. When I do, it's only due to moonlight. In an extreme case, or a case where I am less concerned about aesthetic perfection, I use GraxPert. Its AI tends to give better results than the PI tools. I have developed means to shoot effectively through moonlight that don't involving gradient management which I think is advantageous because gradient tools create compromises in data, however, I think my method of dividing high and low frequency information only works well with dark sky data. I tried to look up your Astrobin upload but couldn't find it. Is your Astrobin name the same as your RUclips handle?
@@SKYST0RY No my handle is 'Lukas Mattern' on astrobin, probably should have mentioned that >.>. I have only uploaded the jpg, it does show the issue I am talking about there, but I do still variants in tif/fit format as well.
@SKYST0RY appear my comments did not make it through. If they appear on your end, apologies for the spam, but I'm guessing yt deleted them. The title of the image is Heart & Soul and it's quite blue and yellow. If you search that it's the 2nd one with both the hand s nebula together.
@@lcmattern YT auto deletes any posts with links except for the content creator. Content creators cannot change it; it's YT policy. You can work around this by spelling out [dot] or [slash].
True, but why bother? It's needlessly complex to provide an outcome that doesn't lend itself as well to finished processing. By doing the finish work in a layer-based photo editor, you can go back and discreetly edit any component of the information at any time.
Provocative cliffhanger. Looking forward to the next video!
I've been working on this theory for months, but I had to wait till I had clear skies under the full moon. Then we had a large high pressure system making for great skies under a super moon. The super moon was the worst possible condition, which was perfect for final experimentation.
Look forward to that 'topic for another time' video about the moon.
Dude you are killing it. Youre my new go to for this stuff, great work.
this way you cut off a lot of the signal in the dark and light parts - you killed it.
The black point cuts off little other than the low luminosity where noise tends to be found. The mid point cuts off nothing because the white point icon is on the right side of the histogram and is unmoved. This method merely sets a black point about 0.1% into the light curve. Pop over to the Sky Story Astrobin and check out all the images that have been "killed" using this method. Everyone of them has been made using this method. www.astrobin.com/users/SkyStory/
I hope this comment comes out the way I intended it to. I've been watching your videos ever since you taught Affinity photo many years ago. Your brain is somewhat sooo advanced then us laymen people and you're so fast, that I have to replay your videos several times to get the info to stick into my lil brain !! LOL.
But once I get what you're saying, everything works great. My processing of my astro images have never been better. This histogram video especially. I've always had trouble stretching my images, even tried many different methods, but this one works the best by far. I also need to learn Affinity photo better too.
One thing that also distracts form my learning is the background music you have running while you're talking. If you feel you need it, please make it quieter.
Thanks and keep up the helpful videos. Your future one of the full moon will be helpful, plus I live in a bortle 7 sky area, so I hope the video includes how to get rid of the effects of light pollution. Clear skies..
Thank you! Be aware that there is no one size fits all histogram strategy, but this one usually works. However, if I am shooting a DSO that has very bright and very dark components, I will do two stretches. For the bright part, I will place the black point just left outside the light curve, and the mid point will be adjusted till the bright information looks good. Then a second light curve is run on the dimmer data using the Evolved method.
The problem with background sound is that it's all digital these days, and some speakers or headphones will emphasize back channels more. In my headphones, the music is louder. On my very high quality test speaker set, the background sound (including music) can barely be heard. I will turn it down a little and see if that can reach a better average for different users.
Your results are awesome. Do you worry about clipping the dark end of the histogram?
It is only a tiny fraction of a single percent of information that is clipped, and only at the base of the effective light curve. Around that point, you are often in the range of noise and camera offsets so it's barely a loss that matters. In fact, it can serve to clean an image.
Would love to see your processing steps in Affinity Photo (from this point). I struggle with AF most days.
A couple months ago, I posted a complete layer-based workflow playlist from start to finish. Hope this helps.
ruclips.net/video/ru_WvGhiDm8/видео.html&pp=gAQBiAQB
@@SKYST0RY Thank you, must of missed this one.
Superb narration like always, please do tell us a bit about your solution to moon light: is it one specific filter or a combination of solutions?
All will be revealed soon lol
Very interesting. How to make the video of the final image with the forward movement of some stars?
I created CGI of a starfield using a planetarium called Space Engine and screen composited the video over the final image via Davinci Resolve.
@@SKYST0RY Gracias! Lo intentaré!
I am not sure if you have tried using the gradient mosaic tool in pi. I made some haoiii, siioiii and rgb images but the ho and rgb, had some artifacts on 2 stars on side bright the other dark.
Is there some other tool you would recommend, as I played with some settings but those stars where always strange. SO looked fine.
I did process this anyway as a first trial for a mosaic and overall really like it, aside from those 2 stars.
I will upload this to astrobin later today of the Heart + soul nebulas.
Because I live at a dark sky site, I almost never have to deal with gradients. When I do, it's only due to moonlight. In an extreme case, or a case where I am less concerned about aesthetic perfection, I use GraxPert. Its AI tends to give better results than the PI tools. I have developed means to shoot effectively through moonlight that don't involving gradient management which I think is advantageous because gradient tools create compromises in data, however, I think my method of dividing high and low frequency information only works well with dark sky data. I tried to look up your Astrobin upload but couldn't find it. Is your Astrobin name the same as your RUclips handle?
@@SKYST0RY No my handle is 'Lukas Mattern' on astrobin, probably should have mentioned that >.>. I have only uploaded the jpg, it does show the issue I am talking about there, but I do still variants in tif/fit format as well.
@SKYST0RY appear my comments did not make it through. If they appear on your end, apologies for the spam, but I'm guessing yt deleted them.
The title of the image is Heart & Soul and it's quite blue and yellow. If you search that it's the 2nd one with both the hand s nebula together.
@@lcmattern YT auto deletes any posts with links except for the content creator. Content creators cannot change it; it's YT policy. You can work around this by spelling out [dot] or [slash].
Where would I find the image? I searched for lcmattern on Astrobin but it didn't pull up anything.
What about GHS? You can achieve very good results with only a few clicks :)
True, but why bother? It's needlessly complex to provide an outcome that doesn't lend itself as well to finished processing. By doing the finish work in a layer-based photo editor, you can go back and discreetly edit any component of the information at any time.
Great post! So at 8:02, I see satellites or small animated stars moving from the center of the image to the bottom left. How did you do that??
I created a motion star field by using a virtual planetarium called Space Engine, then screen composited the starfield onto the background image.
I was wondering the same thing.
this is sadism not telling us how to exterminate the moon glow
Hang tight! LOL The next video will be out soon. I am still finalizing some data, but the next full moon is a whole month away.