Smart Histogram always way too dark, no matter what

[Menno555]

Brain and Robin

There is no timeline, the question on the Smart Histogram was there from when I started with SC. But I delayed using it because I wanted to understand SC a bit better first. So now I was on the point of using and understanding it.
And nothing has changed, other then that I am trying more and more things now that I am starting to understand more. Pushing the envelope is something I do for sure but that is on purpose, so no frustration. Just want to understand things
I am happy with SharpCap, with my images, I love the experimentation with all of this. Seeing objects and capturing them, with succes or failure.
And I ask a lot of questions, I know. And if I don't understand, I keep on asking. Way back I drove my teachers mad with that.
And sometimes I maybe overdo that due to my enthusiasm. But it's never an attack or playing dumb or trolling or something like that.
Will try to diminish it a bit.

Menno

[oopfan]

Menno,

Robin will correct me if I am wrong but I believe that the Brain gives you a way to select how much noise you are willing to tolerate. If you move it in a direction that indicates that you have no tolerance for noise then it will suggest a longer exposure. So this is built-in to the tool. When you read about people advocating for longer exposures, perhaps it is best to translate this to mean that they have a low tolerance for noise. We all need to find the settings that work best for us. If you practice EAA then you will make a trade-off and accept more noise. On the other hand if you have a great mount and guiding then you can lower the noise tolerance level.

Brian

[oopfan]

Frankly I never understood why people get so worked up over exposure selection. You don't find this among practitioners of conventional photography. They are more focused on composition as we should be.

If I were writing the rules this would be my advice:
1. Congratulations! You bought a telescope.
2. First things first, discover how long of an exposure you can tolerate before round stars become oblong. You can tolerate discarding 25% of the frames.
3. OK great! So you can take a 60-second exposure.
4. Choose an "easy" target, something relatively bright.
5. Take one 60-second exposure. How does it look? Can you see it on a single frame, stretched? If not then choose a brighter target.
6. OK great! How do the stars look? Are they unattractively large? If so then cut the exposure in half and repeat. Find a happy medium.
7. Using your final exposure, capture 1 to 2 hours worth of frames using LiveStack or capture now and process later.

It is no more difficult than that. Depending on your Bortle class you may find that increasing exposure beyond a certain point yields washed out frames. If that is the case then lower your exposure.

Brian

[donstim]

Frankly, I don't see anyone directly responding to Menno's question, which as I understand it is, why, when he uses the settings recommended by the Brain tool, are the resulting images so dark as to be completely unusable? I am sure he understands he can use any gain/exposure/brightness settings that he wants, but why is it that he apparently gets nothing in the image if he uses the Brain recommended settings? I don't see him "getting worked up" over it, he just would like to better understand what it going on with this feature.

He's not even asking for all that much in terms of the read noise limit of 10%. And since the Brain is coming up with a solution, why does it not seem to provide anything usable? I've been using the Brain tool for some time now and have had pretty good experience with it. But then, I am live stacking within SharpCap and not trying to save frames to stack with other software.

Menno, can you provide a specific example in terms of the object you are trying to image and the Brain recommended settings? When live stacking, how many frames do you allow to accumulate before determining that it is too dark? You said that when you use the auto stretch, it is still too dark. How about if you do further stretching and allow at least 5 frames to accumulate? Can you provide screen shots showing what you are viewing when requesting the Brain to measure the sky brightness, what the Brain recommended settings are, and then what you see when you are trying to image an object with those settings before and after stretching with at least a few frames stacked?

Don

[oopfan]

Don,

I didn't say that Menno got "worked up" over it. I said that people in general get worked up over it. I hope you have better luck with Menno. We are all here to help but we need more information. I'm outta here.

Brian

[Menno555]

@ Brain
I am trying. English is not my native tongue plus I have difficulties explaining myself. I can't get to a point, there is some "wiring" in my brain that keeps on firing info around and I use that info in communicating. Even had medication for that years ago but did more wrong then right :p It's very difficult for me to write (and speak) what the problem or question is. Just have patience with this poor, bit old, man

@Don
You explained my points well Indeed I use gain, exposure, brightness and white balance and getting better results with that each time I experiment. Also because of the info and answers here. Had a nice session last night with the Cat's Eye Nebula and M92.
I am indeed just curious on why I get the results with the Brain that I am getting.

Example is is M106. Don't have screenshots or captures. Threw them away. Will describe it, sorry in advance for the long text
I followed all the Brain steps. The settings I provided were 15 seconds minimum, 90 seconds maximum, 10% Read Noise Limit, 60 minutes stacking time and Max Dynamic Range. Result was 40 seconds exposure, Gain 61, Brightness 11. Entered those and the histogram was all but there: totally on the left and super thin.
Made my darks and applied them together with my flats. I don't use the Live Stacking in SC because it doesn't work for me most of the time because my FOV (Meade ACF 8" f/10 scope with 0.62x reducer) is small, so there are not enough stars for SC to work with. I do use the stretch of the Mini Histogram though to make everything brighter on screen. But even that did not show much: some stars and M106 was visible but not much.
I did make 24 captures and then ... clouds.
Normally I use DeepSkyStacker but it only detected 4 stars so used AutoStakkert for stacking of the 24 captures. Had to set the Brightness to maximum though to see and select the stars for alignment points. So, a stack of 24 captures.
Then in Photoshop the 4 brightest stars were faintly visible but M106 and the rest not. Just black. Used Curves and Levels but way, way more then normal and the result was that M106 was visible but it had to be brightened SO much before anything became really visible, that the amount of noise was ridicules. The details of the arms and clouds was just not there because of that. And the background was a big, noisy soup.
Tried a test capture last night of M106 with the suggested 40 seconds but with Gain 200 and Brightness 30, so way more then the Brain suggested. The histogram now looked fine and the captures, stacking and Photoshop showed everything just fine.

And just curious now why the Brain gives such low values for me. I know that when Max Dynamic Range is used, the Gain will be low most of the time. But to counter that effect, I would expect that the exposure time and brightness would make up for that to show at least the faintest objects. But maybe that is a fault in my logic.

[admin]

Hi,

things to check and things to do which will help us understand what is going on ...

* Make sure that you are getting the brain to measure the *darkest* part of the frame with the least nebulosity/sky glow
* If you are using filters, use the same filter to make the brain measurement as you intend to use for imaging.
* Capture a screenshot showing the results of the brain measurement/calculation and post it here - make sure it shows the main camera settings on the right too
* Capture a couple of image frames at the suggested settings and share them. Please share the .capturesettings.txt files too
* Capture a couple of image frames at the settings that seem right to you and share those too - again including the .capturesettings.txt
* post a full description of your optical equipment - scope, reducers/extenders/filters/camera so we can be sure of that
* Remind us of your bortle conditions

thanks,

Robin

[Menno555]

Robin

Will do that.
Btw, that first step. Till now I did choose the darkest spot near the object I wanted to capture. Seemed logical for me. Is that okay or must it be the darkest part of sky I can see, even when it is not even close to the object?

Thanks in advance,
Menno

[donstim]

I don't have the same camera you do, so can't comment directly on the Brain's suggested gain/exposure/brightness settings other than it has picked the gain where the read noise has a sudden "step" dropoff and so corresponds with the highest dynamic range, lowest read noise gain for your camera. (You can see this from examining the read noise and dynamic range plots in your camera owners manual.)

For the histogram, do you have the Logarithmic box checked? This will spread out the shape of this histogram so it might not be so compressed to the left hand side.

Have you tried the same thing without darks or flats? (Just in case there might be an issue with your darks and/or flats.)

If you look at the Brain's smart histogram while imaging at the suggested settings, where does the peak of the histogram fall relative to the red, amber, and green bars across the top of the histogram? Can you provide a screenshot?

[turfpit]

Menno

In File > SharpCap Settings > General tab under Saving, is Save 10/12/14 bit images in FITS files without scaling to 16 bit checked or unchecked? For normal imaging, it should be unchecked.

Dave