Jump to content

SkySurveyBanner.jpg.21855908fce40597655603b6c9af720d.jpg

Dark Orientation Fail?


Notty

Recommended Posts

I had another go at M31 last night, got lost of data and then I think I may have performed a supreme schoolboy error and parked my scope before running my dark shoot plan before going to bed (because it was going to take about an hour).

My subs are all portrait and my darks are all landscape.  I had not forseen this :mad: !  Unfortunately I've completely forgotten what orientation the camera was in the tube either when it was doing the subs or parked... I think I've made my darks therefore unusable... am I right?  

Link to comment
Share on other sites

Could you not simply rotate them either left or right?  Try both, and see which works better.

As I understand it, the purpose of taking the darks is so that you have an image of the noise in the sensor at approximately the same temperature as when you were taking the real exposures, so that an algorithm can identify the points and remove the noise.  Whether the sensor is turned left or right or upside-down, it's still the same sensor -- the only reason they show as landscape rather than portrait is because the camera enters in metadata somewhere that it was so.  The sensor itself is the same, so the noise should be the same...I'd try it and see rather than throwing away.

Link to comment
Share on other sites

Darks don't need any orientation! They are taken with the camera shutter closed. You can even do them with the camera off the telescope (some people put DSLRs into the fridge to get the temperatures the same!).

As long as you shot them at the same temperature as the light frames then they will be fine.

Link to comment
Share on other sites

I get this problem when packing away my DSLR and leaving it to take darks. I just rotate the rogue ones before stacking - I can tell which way to go by looking at the pattern of hot pixels. You might be able to do the same with your lights, in order to match up the darks in the same orientation.

Darks don't need any orientation!

Are you sure about that? What if part of the sensor produces more noise than another part? I'd have thought the darks would need to be in the same orientation as the lights.

Link to comment
Share on other sites

Are you sure about that? What if part of the sensor produces more noise than another part? I'd have thought the darks would need to be in the same orientation as the lights.

Positive.

When taking darks the optical train is blanked off. Therefore the camera orientation makes no difference. How is the camera supposed to know (or even care) which way is is mounted on the scope or which part of the sky it is pointing to?

For some reason, Windows displays some images in portrait and some in landscape when you view the thumbnails. I've seen this when I use a DSLR to shoot wide-angles. Some will be displayed in landscape, some in portrait. They are all in the same orientation though.

Link to comment
Share on other sites

I am aware the camera can't see anything with the lens cap on. ;)

For some reason, Windows displays some images in portrait and some in landscape when you view the thumbnails. I've seen this when I use a DSLR to shoot wide-angles. Some will be displayed in landscape, some in portrait. They are all in the same orientation though.

It depends on how the camera stores the images. If the orientation is just metadata then you're right. If it encodes them differently then they need to be rotated the same way as the lights in order for the patterns to match.

Link to comment
Share on other sites

It depends on how the camera stores the images. If the orientation is just metadata then you're right. If it encodes them differently then they need to be rotated the same way as the lights in order for the patterns to match.

That I can't comment on as I am not familiar with the intricacies of how data is stored in-camera. I know that I've never rotated a dark and have never seen a problem.

I think that we are having two different conversations though as the OP's point was that he thought the darks were unusable as he had parked the scope. My point is that it makes no difference which way the orientation of the scope is. The camera doesn't care where in the sky it is pointing at, or indeed if it is still connected to a scope.

Link to comment
Share on other sites

Thanks for the responses guys. I have the same problem with my bias frames though, and I am guessing that orientation is important. I get that they would all have their varioushotons on the right sensors regardless etc but I'm concerned at how DSS will decide to apply them. So in windows the lights are in portrait and half the bias and all the darks are landscape. If I rotate them in the canon photo editor (which is the only app I have that seems able to edit theses raw canon files before DSS turns them into tifs) and save them but windows still insists they are landscape. Looking at the data attached to each picture the only difference is for some reason the lights have a value "aspect ratio 3:2" with all the other details common with the mis-oriented darks except this, which for some reason is missing from these files.

Anyway I'm sort of glad it's not that clear cut I don't feel so daft now!

Link to comment
Share on other sites

You cannot change the orientation of the chip. The chip is the chip. It's long side remains its long side. When you rotate the camera you rotate the chip but you don't change the chip's orientation with regard to itself, so the long side remains the long side. There is no way the software can make the long side the short side!

Now, some DSLRs do rotate the screen shot if you turn the body from landscape to portrait. They may store the shot in portrait as well. But they are just rotating the image. The same 'pixel x' which was rather hot and read a value of 3800 ADU too high is still the same pixel and needs 3,800 ADU subtracting from it. That pixel cannot change place on the chip.

The only problem I can forsee is that the darks might go into the software in the wrong orientation but this will be easy to avoid. Make a master dark on its own, first, and give it a hard screen stretch. It is bound to have some distinguishing features, like amp glow in one corner or a column defect. Now do a hard screen stretch of one sub and look for that feature. If necessary rotate your dark to ensure that the feature appears in the same place as your light frame and you can use the master dark confidently.

Olly

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.