Jump to content

Narrowband

M94-Please have a go


Rodd

Recommended Posts

I should I thought of this before now.  Please find raw, calibrated, registered stacks of M94.  The stacks have not been cropped yet.  I used SFS to inspect the subs, and I removed subs with high FWHM, eccentricity, median sky levels, low (or super high) SNR, and noise. Calibration worked well--no issues with dust bunnies moving like with M63.  As recommended, no bias frames were used in calibration, just darks, flats and flat darks.  Every sub and calibration frame was shot at -20, and the darks were of the same duration as the lights and flats.  While there is no doubt that my processing skills are not helping things, there are also aspects of this data that I do not think are indicative of decent data.  But I can't figure out what the problems are.  Each sub was 2 min, with the number of subs being incorporated in the stack name.  The SL stack is a synthetic luminance consisting of all subs.  I ended up not using it, but I figured I would provide it just in case.  

I do not know why there seems to be fog, or nebulosity everywhere--dropping off near the edges.  The imaging circle of the TOA 130 at F7.7 is 90mm, so there can be no viginetting on a chip that has a 21mm diagonal.  This haze, for the lack of a better term, is not removed by DBE.  Also-note the background of the SL stack--314 subs is not a paltry number--a degree of smoothness is to be expected as data grows.  But not so here--there is a graininess, or coarseness to the background that I don't understand.  

Gain was unity (139), temp was -20, dithering was used after every sub.  I understand that many subs are required with this camera--but this should be enough to render half way decent results (100+ in each channel).  There is nothing for me to do but try manipulating gain and exposure length.  But unity (139) is a sort of midway set point that is commonly used to great effect.  Maybe 2min is too long to expose with unity gain for RGB?  Unity gain did work quite well for Ha subs of 5min duration.   Having trouble collecting decent RGB data is nothing new for me....I just did not expect it to be SO bad with this camera.   Is it my gain settings?  My exposure times?  The sky conditions? Or my processing? Have fun!  

r114.fit

g96.fit

b109.fit

SL-314.fit

SL-314.fit

Link to comment
Share on other sites

  • Replies 63
  • Created
  • Last Reply

An unexpected response.  Usually folks jump on data to process.  Any thoughts at all would be most welcome.  You don't have to process them...just look.  Even a question would be helpful. 

Rodd

Link to comment
Share on other sites

On 25/04/2018 at 19:28, Rodd said:

I should I thought of this before now.  Please find raw, calibrated, registered stacks of M94.  The stacks have not been cropped yet.  I used SFS to inspect the subs, and I removed subs with high FWHM, eccentricity, median sky levels, low (or super high) SNR, and noise. Calibration worked well--no issues with dust bunnies moving like with M63.  As recommended, no bias frames were used in calibration, just darks, flats and flat darks.  Every sub and calibration frame was shot at -20, and the darks were of the same duration as the lights and flats.  While there is no doubt that my processing skills are not helping things, there are also aspects of this data that I do not think are indicative of decent data.  But I can't figure out what the problems are.  Each sub was 2 min, with the number of subs being incorporated in the stack name.  The SL stack is a synthetic luminance consisting of all subs.  I ended up not using it, but I figured I would provide it just in case.  

I do not know why there seems to be fog, or nebulosity everywhere--dropping off near the edges.  The imaging circle of the TOA 130 at F7.7 is 90mm, so there can be no viginetting on a chip that has a 21mm diagonal.  This haze, for the lack of a better term, is not removed by DBE.  Also-note the background of the SL stack--314 subs is not a paltry number--a degree of smoothness is to be expected as data grows.  But not so here--there is a graininess, or coarseness to the background that I don't understand.  

Gain was unity (139), temp was -20, dithering was used after every sub.  I understand that many subs are required with this camera--but this should be enough to render half way decent results (100+ in each channel).  There is nothing for me to do but try manipulating gain and exposure length.  But unity (139) is a sort of midway set point that is commonly used to great effect.  Maybe 2min is too long to expose with unity gain for RGB?  Unity gain did work quite well for Ha subs of 5min duration.   Having trouble collecting decent RGB data is nothing new for me....I just did not expect it to be SO bad with this camera.   Is it my gain settings?  My exposure times?  The sky conditions? Or my processing? Have fun!  

r114.fit

g96.fit

b109.fit

SL-314.fit

SL-314.fit

Thanks Rodd. I'll try and have a go tonight.

ps - The last 2 files are labelled the same. Are they different, or have you uploaded the same file twice?

Link to comment
Share on other sites

22 minutes ago, Xiga said:

Thanks Rodd. I'll try and have a go tonight.

ps - The last 2 files are labelled the same. Are they different, or have you uploaded the same file twice?

Sorry--they are the same...I did not think it uploaded I guess and tried again.  Didn't notice that.  Thanks for trying

Rodd

Link to comment
Share on other sites

8 hours ago, Rodd said:

An unexpected response.  Usually folks jump on data to process.  Any thoughts at all would be most welcome.  You don't have to process them...just look.  Even a question would be helpful. 

Rodd

Haven't had time yet. But I'll jump this weekend. No astrodarkness here, just hail and rain, so your extra data is welcome.

Link to comment
Share on other sites

42 minutes ago, wimvb said:

Haven't had time yet. But I'll jump this weekend. No astrodarkness here, just hail and rain, so your extra data is welcome.

My heart goes out to you.  I don't know what I would do with no darkness (I image in Nautical darknss with Ha sometime--seems ok for the portion closest to astrodark.  Thanks for giving it a go,

Rodd

Link to comment
Share on other sites

42 minutes ago, vlaiv said:

Here is my take on it (probably wrong color balance, I just used one star to map color), and core slightly blow (maybe, maybe not), I even did not do any denoising:

GIMP-2.9.thumb.png.2ca991c97941ff15803bcae0878f4a44.png

 

:hello2::hello2::hello2: Wow--how do you do that.  Definitely did good....on the inner part.  I should have said that the outer ring was my target.  Without it, I don't think it works.  But for what it worth--your inner portion is regally nice--subtle, yet detailed.  I like the color.  OK--so maybe allot of it IS me.  I don't know whether to be happy or sad.  Mt efforts look like a blob of clay.

Rodd

Link to comment
Share on other sites

5 minutes ago, Rodd said:

:hello2::hello2::hello2: Wow--how do you do that.  Definitely did good....on the inner part.  I should have said that the outer ring was my target.  Without it, I don't think it works.  But for what it worth--your inner portion is regally nice--subtle, yet detailed.  I like the color.  OK--so maybe allot of it IS me.  I don't know whether to be happy or sad.  Mt efforts look like a blob of clay.

Rodd

I don't think you should be sad, data is good, maybe a bit of LP gradient or passing high cloud creates unevenness in the background. As for processing - that is something that can be learned and one can get more skillful with time.

Link to comment
Share on other sites

Hi Rodd

Here's my take:

combine-RGB-image-mod-lpc-cbg-csc-NoSt-Arcsinh.thumb.jpg.acf30feb84643dcde5d928490be62daf.jpg

I found it quite tough to process i must admit, but not because of the quality of the data (which is excellent btw), just because RGB can be hard and i suspect this might be a tougher target than one might think. 

Here's a very brief run-down of what i did (the order might not be exact):

1. Combine the channels in Astro Pixel Processor. For this i assigned the R,G, and B to their respective channels, and i then assigned the Luminance and the R and G all to the Luminance channel. APP only lets you select a maximum of 6 channels at the moment, hence i wasn't able to use the G in the Luminance as well (perhaps a good thing, as it was the noisiest channel - although still far from noisy!). 

2. Gradient Reduction and star calibration also done in APP. Saved as a TIFF for finishing off in PS.

3. Multiple arcsinh stretching done in PS to preserve colour (this is still to be added to APP, otherwise i would have just used a DDP-stretched version from it). I don't have PI so i use @sharkmelley 's excellent method for doing similar in PS.

4. Highlight recovery via Shadows & Highlights to bring the core back a good bit. 

5. High Pass Filter with Soft Light to sharpen the core. 

6. Colour saturation using Jay GaBany's technique. 

7. Colour noise reduction using the Camera Raw filter. 

8. HLVG to remove the green.

9. Horizontal & Vertical banding reduction using Noel Carboni's actions.

10. Subtle curves to increase the contrast of the outer ring. Set the black point, and up the Vibrancy. 

11. Add a small amount of NR using Carboni's Deep Space NR. 

12. Crop out the slightly darker edges. 

I haven't done much RGB work so thanks for letting me play with it. :thumbright:

Link to comment
Share on other sites

38 minutes ago, Xiga said:

. Combine the channels in Astro Pixel Processor. For this i assigned the R,G, and B to their respective channels, and i then assigned the Luminance and the R and G all to the Luminance channel. APP only lets you select a maximum of 6 channels at the moment, hence i wasn't able to use the G in the Luminance as well (perhaps a good thing, as it was the noisiest channel - although still far from noisy!). 

2. Gradient Reduction and star calibration also done in APP. Saved as a TIFF for finishing off in PS.

3. Multiple arcsinh stretching done in PS to preserve colour (this is still to be added to APP, otherwise i would have just used a DDP-stretched version from it). I don't have PI so i use @sharkmelley 's excellent method for doing similar in PS.

4. Highlight recovery via Shadows & Highlights to bring the core back a good bit. 

5. High Pass Filter with Soft Light to sharpen the core. 

6. Colour saturation using Jay GaBany's technique. 

7. Colour noise reduction using the Camera Raw filter. 

8. HLVG to remove the green.

9. Horizontal & Vertical banding reduction using Noel Carboni's actions.

10. Subtle curves to increase the contrast of the outer ring. Set the black point, and up the Vibrancy. 

11. Add a small amount of NR using Carboni's Deep Space NR. 

12. Crop out the slightly darker edges. 

I haven't done much RGB work so thanks for letting me play with it. :thumbright:

Edited 31 minutes ago by Xiga

Well-I do not speak your dialect, but there is no arguing with the result.  The outer ring is visible.  The background is a bit yellowish and noisy--but only upon zoom.  The detail in the core is very nice.  This brings me to a very strange place.  I am using the processing tools in the order and the manner they were designed, and yet it is not working.  I tend to process my images over and over until I get them as good as I can get them, then I wait a month and process them all over and over again add nauseam.  If I had more clear nights I am sure I would collect new images.  Or, if I was satisfied with them.  So, I have been over this data incessantly and still can't get it to look decent.  I don't know what else to try.  I have had some success--M33, The triplets, M1, M101--even the tadpoles in LRGB--but I did the same processing I always do--its wierd how sometimes it works and sometimes it doesn't.

Rodd

Link to comment
Share on other sites

6 hours ago, Rodd said:

Well-I do not speak your dialect, but there is no arguing with the result.  The outer ring is visible.  The background is a bit yellowish and noisy--but only upon zoom.  The detail in the core is very nice.  This brings me to a very strange place.  I am using the processing tools in the order and the manner they were designed, and yet it is not working.  I tend to process my images over and over until I get them as good as I can get them, then I wait a month and process them all over and over again add nauseam.  If I had more clear nights I am sure I would collect new images.  Or, if I was satisfied with them.  So, I have been over this data incessantly and still can't get it to look decent.  I don't know what else to try.  I have had some success--M33, The triplets, M1, M101--even the tadpoles in LRGB--but I did the same processing I always do--its wierd how sometimes it works and sometimes it doesn't.

Rodd

Today this looks very different on my work monitor vs my cheapo home monitor. It looks MUCH darker than I was expecting. I really need to sort that out!

I agree with you, it is noisier than I would like in places too. This was a result of me pushing it quite hard in order to bring out the outer ring. My stretching skills are probably not the best tbh so I'm sure someone will be able to improve on this.

What are your skies like? RGB really needs a dark sky, unlike with NB.

How aggressive was your dithering between subs?

Depending on what your skies are like, it's possible you aren't using the most optimal settings. Have you measured the ADU of your subs? You can use a bit of Maths to determine what level you need to get to, then you will know if you are under or over (most likely over) exposing. Here's a couple of good sources of information (Jon Rista has done a lot of testing with this camera, he knows his stuff):

https://www.cloudynights.com/topic/570020-gain-settings-for-asi1600mm-cool/

https://www.cloudynights.com/topic/559771-asi1600mm-c-cheat-sheet-no-math/

To be clear though, the settings you are currently using would not be hindering the quality of your images, save for possibly over-exposing some brighter areas.

And finally, I would personally be VERY happy with this level of image capture (I'm a cheap date though, lol). I think you're doing great as it is. And you're not alone in producing a different outcome with each re-process, I think we all can understand where you are coming from on that one.

ps - What sort of result did you get on this one yourself? Would be good to see as a comparison. Also, doing a quick search on Astrobin shows quite clearly that this is not an easy target by any means.

Keep it up!

Link to comment
Share on other sites

2 hours ago, Xiga said:

Today this looks very different on my work monitor vs my cheapo home monitor. It looks MUCH darker than I was expecting. I really need to sort that out!

I agree with you, it is noisier than I would like in places too. This was a result of me pushing it quite hard in order to bring out the outer ring. My stretching skills are probably not the best tbh so I'm sure someone will be able to improve on this.

What are your skies like? RGB really needs a dark sky, unlike with NB.

How aggressive was your dithering between subs?

Depending on what your skies are like, it's possible you aren't using the most optimal settings. Have you measured the ADU of your subs? You can use a bit of Maths to determine what level you need to get to, then you will know if you are under or over (most likely over) exposing. Here's a couple of good sources of information (Jon Rista has done a lot of testing with this camera, he knows his stuff):

https://www.cloudynights.com/topic/570020-gain-settings-for-asi1600mm-cool/

https://www.cloudynights.com/topic/559771-asi1600mm-c-cheat-sheet-no-math/

To be clear though, the settings you are currently using would not be hindering the quality of your images, save for possibly over-exposing some brighter areas.

And finally, I would personally be VERY happy with this level of image capture (I'm a cheap date though, lol). I think you're doing great as it is. And you're not alone in producing a different outcome with each re-process, I think we all can understand where you are coming from on that one.

ps - What sort of result did you get on this one yourself? Would be good to see as a comparison. Also, doing a quick search on Astrobin shows quite clearly that this is not an easy target by any means.

Keep it up!

It looks different on my 4k processing monitor too....BETTER!  I would be thrilled with this.  Thanks for the links--I will have a look.  Your probably right about my exposure times. Maybe 2min is too long for this camera for my skies.  Here's my best version.  No where near what has been done by you and others on this thread.

https://www.astrobin.com/343641/0/?nc=user

 

 

Rodd

Link to comment
Share on other sites

1 hour ago, Rodd said:

It looks different on my 4k processing monitor too....BETTER!  I would be thrilled with this.  Thanks for the links--I will have a look.  Your probably right about my exposure times. Maybe 2min is too long for this camera for my skies.  Here's my best version.  No where near what has been done by you and others on this thread.

https://www.astrobin.com/343641/0/?nc=user

 

 

Rodd

I think you were maybe trying too hard to bring out the outer ring. It's quite faint, so it's simply not possible to accentuate it a lot without bringing up the noise level.

I would also recommend always shooting True Luminance (with shorter subs) rather than combining the R,G, and B into a synthetic luminance. Obviously in a perfect world, one would always dither between every frame, but with shorter subs, the lost time due to frame download and dithering can become more significant than insufficient dithering, so you need to find a balance. A good rule of thumb is to dither approx. every 1 in 20 subs. So if you only capture 20 subs then you need to dither every frame, but if you are capturing 100 subs (a good number for this camera, to help recover bit depth, which is important as it's only 12 bit) then you only need to dither every 5 subs. This will save you some overhead in lost time, so you can then afford to up the aggressiveness of the dithering. Don't be afraid to really crank it up, the more dithering the better. It's one of the most important steps any imager should take. Also, I'm not sure how you are dithering, but if you are using PHD then make sure it is set to dither randomly, and not in a spiral fashion.

You probably know all that already, but if not then I hope it's of some help.

Link to comment
Share on other sites

So I had a go at it in Photoshop. I think the data was quite nice, maybe a bit noisy. Interestingly, when I stretched the colour channels together (about 10 curve stretches) I ended up with a rather different colour tone than Ciaran - more blue. Have no idea why since PI processing is for the most part a black hole for me. Then I sharpened it a bit mainly by using a high pass filter. In any case, this is what I came up with:

PS. Thanks Rodd for this unexpected entertainment!

PS2. I noticed that adding the luminosity increased the noise in the background sky, so using layers I only added lum to the galaxy and its shell.

 

 

Rodd 114 RGB PS15 GNilsson.jpg

Link to comment
Share on other sites

Sorry if I wasn't clear. The majority of my own processing (i.e from step 3 onwards) was also done in Photoshop. I don't own Pixinsight myself. I do have APP, which I mostly just use for stacking and gradient reduction.

ps - That's lovely Goran. When I click on the picture though it doesn't display properly for some reason.

Link to comment
Share on other sites

1 minute ago, Xiga said:

Sorry if I wasn't clear. The majority of my own processing (i.e from step 3 onwards) was also done in Photoshop. I don't own Pixinsight myself. I do have APP, which I mostly just use for stacking and gradient reduction.

ps - That's lovely Goran. When I click on the picture though it doesn't display properly for some reason.

Thanks!

Try again - I was probably uploading a new version since I detected some green in it on my screen.

Link to comment
Share on other sites

2 hours ago, Xiga said:

so it's simply not possible to accentuate it a lot without bringing up the noise level.

Have a look at Barry Wilson's version on DSO browser.  Amazingly clean and bright.  And with limited data....and with lousy skies.  

Link to comment
Share on other sites

1 hour ago, gorann said:

So I had a go at it in Photoshop. I think the data was quite nice, maybe a bit noisy. Interestingly, when I stretched the colour channels together (about 10 curve stretches) I ended up with a rather different colour tone than Ciaran - more blue. Have no idea why since PI processing is for the most part a black hole for me. Then I sharpened it a bit mainly by using a high pass filter. In any case, this is what I came up with:

PS. Thanks Rodd for this unexpected entertainment!

PS2. I noticed that adding the luminosity increased the noise in the background sky, so using layers I only added lum to the galaxy and its shell.

 

 

Rodd 114 RGB PS15 GNilsson.jpg

Wow--nice job.  Normally I do shoot lum, but you think this data was noisy--the luminance has been really bad--also, by the time I finished the RGB, the Moon was starting to get invasive.  Maybe I will return to this target and get some proper lum.  I think I like your version the best,

Link to comment
Share on other sites

2 hours ago, Xiga said:

Don't be afraid to really crank it up, the more dithering the better.

At present, the pixel size of the Lodestar 2 is 8.2um and the imaging sensor pixels are 3.8.  The resolution difference is 6.8arcsec/pix vs .78 arcsec/pix.  So a dither setting of .5 pixels moves the imaging frame 5-8 pixels on the imaging sensor.  This setting allows me to use a dither delay of 15 sec.  When the pixels are closer in size (like the STT-8300 and self guiding filter wheel), I set it at 5 pixels and the dither delay is 70 sec--but 70 sec every 10 min (30min for narrowband) is not bad.  Based on the images you folks have posted with my data--I am hesitant to change much.  I just need to learn how to process.  I might try switching to PHD2 and SGP--but that will require at least a night to test and learn.  Nights are so few and far between, I find it almost impossible to justify making big changes like that.  

Rodd

Link to comment
Share on other sites

5 minutes ago, Rodd said:

At present, the pixel size of the Lodestar 2 is 8.2um and the imaging sensor pixels are 3.8.  The resolution difference is 6.8arcsec/pix vs .78 arcsec/pix.  So a dither setting of .5 pixels moves the imaging frame 5-8 pixels on the imaging sensor.  This setting allows me to use a dither delay of 15 sec.  When the pixels are closer in size (like the STT-8300 and self guiding filter wheel), I set it at 5 pixels and the dither delay is 70 sec--but 70 sec every 10 min (30min for narrowband) is not bad.  Based on the images you folks have posted with my data--I am hesitant to change much.  I just need to learn how to process.  I might try switching to PHD2 and SGP--but that will require at least a night to test and learn.  Nights are so few and far between, I find it almost impossible to justify making big changes like that.  

Rodd

I don't get it, don't you get clear nights when there's a full moon like all of us?

On another note, your stars look tight enough, but have you considered at least throwing a barlow between the guidescope and the lodestar?

I dither usually each 10-15 minutes and aim for a main imaging camera field movement of at least 10 pixels. I also offset the target during multiple nights.

Link to comment
Share on other sites

45 minutes ago, moise212 said:

I don't get it, don't you get clear nights when there's a full moon like all of us?

On another note, your stars look tight enough, but have you considered at least throwing a barlow between the guidescope and the lodestar?

I dither usually each 10-15 minutes and aim for a main imaging camera field movement of at least 10 pixels. I also offset the target during multiple nights.

Unfortunately no....since I got this camera, I have imaged 4 times (in over 2 months).  and 2 of those times were shakey.  Its been really bad.  As far as the barlow--at present, at 1,000mm fl, its the longest FL I will use with this guidescope.  And guiding seems very solid.  I originally got this camera to use with the FSQ 106 at F3 (.6x reducer)--which has a FL of 318mm.  So a barlow won't be needed there.  The small pixels of the camera will yield a resolution of 2.46--which is pretty good for really wide field.  I am using the camera now with the TOA 130 because it was the scope that was up and I wanted to get used to it on a scope I know (I have not used the FSQ 106 yet either!).  As I type this it is pouring out--and will be lousy for another week.  I have never seen it like this.  That was the other reason I got the FSQ 106 and reducer for F3--maybe I will be able to finish an image in 1 night (or 2) instead of 5 or 6 (which is really almost a month of actual time).

Link to comment
Share on other sites

19 minutes ago, Barry-Wilson said:

Rodd you certainly have the data as the very fine processing versions show.

It is a fiendishly difficult target to capture and process: as Rodd mentioned, my version is here, Barry's SmugMug site, and I plan to improve on this.

Nice one Barry!

But I am confused about the "right"colours of this galaxy and a quick look on the web does not help. Versions go from yellow to red to pink to blue. Just like ours...

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.