Jump to content


Photo

Jupiter October 9...and yes, the seeing was good.

  • Please log in to reply
19 replies to this topic

#1 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 12 November 2012 - 01:58 PM

Hi all,

Here are Jupiter images shot in good seeing.

http://www.sunspot51...39_RGB_pmax.jpg

Paul

#2 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 5228
  • Joined: 18 Jun 2006
  • Loc: Valley of the Sun

Posted 12 November 2012 - 08:04 PM

Nice to have good seeing! Good work Paul but those limb effects are troubling. Any ideas why your getting those?

Glenn

#3 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 12 November 2012 - 09:11 PM

Glenn,

The limb affects are troubling and I have no idea how to gtet rid of them. Perhaps I'm too aggressive with wavelets? I wish I knew other processing ideas to add sharpness without adding the limb affects. At the ASA Expo in Tucson this past weekend I talked to Adam Block about producing a planetary processing tutorial. He seemed open to it, but not sure when he can get to it. Maybe if enough of us bug him about it....

Otherwise, I'd love suggestions.

Paul


Nice to have good seeing! Good work Paul but those limb effects are troubling. Any ideas why your getting those?

Glenn



#4 astrovienna

astrovienna

    Surveyor 1

  • -----
  • Posts: 1770
  • Joined: 04 Dec 2006
  • Loc: The NoVa White Zone

Posted 12 November 2012 - 10:30 PM

Paul, if you post your pre-wavelets stacks, I'll take a look at the data. Are you stacking in AS?

Kevin

#5 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 13 November 2012 - 09:30 AM

I'll do that. I'll get them out later today. I am using AS for my stacking.

Thanks!!
Paul

#6 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 13 November 2012 - 09:37 AM

Hard to think it is too aggressive wavelet use as the rest of the disk looks very nice.

And the limbs are a little off in the individual channels, so it shouldn't be the RGB compilation. From that, it seems to me it must be in the aligning and stacking routine. But I generally find AS2 to be the most reliable program in my workflow. However, you might try it in R6 or avistack (or something else, just to see if it is still there). I'll look forward to seeing the pre-wavelet stacks. It's too bad there isn't a decent way for us all to share avis. (Is there?)

#7 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 13 November 2012 - 09:39 AM

Here is a link to the zipped tif raws. I've labeled them and included the 685nm data as well. One thing of note is that my mount has a significant random vibration. I am going to get the mount hyper tuned in the next month or so which should remove that annoyance. The mount is a CGE-Pro.

http://www.sunspot51.../jupiterraw.zip

Paul

#8 Rankinstudio

Rankinstudio

    Mercury-Atlas

  • *****
  • Posts: 2595
  • Joined: 05 Apr 2008
  • Loc: Southern UT

Posted 13 November 2012 - 12:44 PM

Nice work Paul :)

#9 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 14 November 2012 - 09:18 AM

Paul, I haven't been able to reproduce an image without the edge effects. Looking at the pre-wavelet images, they definitely show the effects as well, so it must be either in the stacking program or in the acquisition.

Do you always see this? If so, it might be those vibrations, though I'd think that would mess up the center of the disk as well.

#10 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 14 November 2012 - 12:33 PM

Paul,

Glenn Jolly and I have thought maybe it is my acquisition process, I capture BMP's rather than AVI or SER. I captured AVI this morning and I'm not sure if it has made a difference. I need more testing. I also manually set gamma at 1024, normally I leave the box unchecked. This has shown up for several years and with the Mewlon as well. I should be getting my EM200 mount back in a couple of weeks and I can do some testing there as well. I do think a little of the issue has to do with the vibration, but won't know how much until I get the CGE-Pro hyper tuned. Frustrating to be so close, but so far!

Thanks!!
Paul


Paul, I haven't been able to reproduce an image without the edge effects. Looking at the pre-wavelet images, they definitely show the effects as well, so it must be either in the stacking program or in the acquisition.

Do you always see this? If so, it might be those vibrations, though I'd think that would mess up the center of the disk as well.



#11 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 5228
  • Joined: 18 Jun 2006
  • Loc: Valley of the Sun

Posted 14 November 2012 - 02:57 PM

When running the Flea3 in FireCapture the appearance of the planet should not change when checking and unchecking the gamma box if it is set to 1024. You can see the gamma effect by noting whether or not the histogram shifts left or right. For example a setting of 1000 will shift it a little to the left. This should read gamma 0.977 in Flycap. A setting of 1024 should give a gamma of 1.0. So 1024 should be ok, but I'd recommend unchecking the gamma box most of the time.

I have mistakenly left the gamma in a low setting used for focussing, like 0.8, and taken data on Jupiter. I did get limb or onion rings in that case. What having gamma set to greater than 1.0 would do I cannot say, but I don't think it would be good for Jupiter.

Now whether a setting of gamma 1.0 on a given Flea3 is actually gamma 1.0 would have to be bench tested. To do this requires a calibrated light source to do properly. It can also be a function of gain so its not an easy thing to measure. I tried once using f stops on a camera lens, but even the best Nikon lenses are not geometrically accurate enough.

Glenn

#12 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 14 November 2012 - 02:59 PM

Hmm. I usually set it to 0.95. I should try one with it unchecked.

#13 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 5228
  • Joined: 18 Jun 2006
  • Loc: Valley of the Sun

Posted 14 November 2012 - 03:35 PM

A gamma setting of 950 should yield a gamma of 0.928. I'm not sure I would recommend it. But experimentation may be called for. I only set the gamma down for focussing, and hopefully remember to uncheck the box before capture.

The settings I'm stating depend on whether or not you are in absolute mode.

If you bring up FlyCap from power on without FireCapture ever running, you will find the default gamma is set to 1.25 (or a register setting of 1280 in non absolute mode). So if somehow the camera is in that condition it would not be useful for planetary imaging.

The settings for gamma we put in FireCapture are the non absolute register settings, just to be clear. So 1024 is gamma 1.0. A setting of 950 is gamma 0.928.

Glenn

#14 Sunspot

Sunspot

    Fly Me to the Moon

  • *****
  • Posts: 5650
  • Joined: 15 Mar 2005
  • Loc: Surprise, AZ

Posted 14 November 2012 - 04:41 PM

That matches what I noticed during my tests this morning. Had to try it just to be sure.

I have a couple of dummy filters called "red-focus" and "blue focus" where the gamma is set at 512. I adjust focus off of that then before imaging I click back to the R filter to start the RGB. At midnight for me it's a good thing I don't have to remember to uncheck the gamma box. :lol:

Paul


When running the Flea3 in FireCapture the appearance of the planet should not change when checking and unchecking the gamma box if it is set to 1024. You can see the gamma effect by noting whether or not the histogram shifts left or right. For example a setting of 1000 will shift it a little to the left. This should read gamma 0.977 in Flycap. A setting of 1024 should give a gamma of 1.0. So 1024 should be ok, but I'd recommend unchecking the gamma box most of the time.

I have mistakenly left the gamma in a low setting used for focussing, like 0.8, and taken data on Jupiter. I did get limb or onion rings in that case. What having gamma set to greater than 1.0 would do I cannot say, but I don't think it would be good for Jupiter.

Now whether a setting of gamma 1.0 on a given Flea3 is actually gamma 1.0 would have to be bench tested. To do this requires a calibrated light source to do properly. It can also be a function of gain so its not an easy thing to measure. I tried once using f stops on a camera lens, but even the best Nikon lenses are not geometrically accurate enough.

Glenn



#15 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 14 November 2012 - 04:46 PM

I think I read somewhere that the gamma setting in FireCap was a little different than what was "real" on the camera. But it sounds like I had which direction it is off backwards.

#16 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 5228
  • Joined: 18 Jun 2006
  • Loc: Valley of the Sun

Posted 14 November 2012 - 05:37 PM

Fear not, Torsten has all this figured out for us already!

When the gamma box is unchecked it sets gamma to 1.0 or a register setting of 1024. Manually setting the register to 1000 will give gamma 0.977 (1000/1024).

In average seeing which is most of the time Jupiter looks pretty bland in the red channel so there is a temptation to lower gamma. When the seeing is really good however Jupiter starts looking pretty snappy at gamma 1. For focussing lowering the gamma sometimes helps, if you lower a lot then you may have to increase the brightness. All that can lead to confusion when you are shivering in the dark.

Gamma 1 means linear response. Thats what you want for capture. You can always manipulate the histogram later.

Point Grey documents variously state the default gamma as 1.0 and 1.25. I have not checked the dates on these docs so maybe there was a correction later. I have confirmed that for my Flea3 the default gamma is 1.25 (register 1280). Thats ok for security where you want to see stuff in the shadows. However when you startup Torsten's FireCapture it overrides that default with whatever you have selected.

Glenn

#17 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 14 November 2012 - 06:03 PM

So...leaving it unchecked in FireCap will produce a gamma 1 image? Or do I need to set it to 1024?

Sorry to hijack the thread.

#18 DesertRat

DesertRat

    Fly Me to the Moon

  • *****
  • Posts: 5228
  • Joined: 18 Jun 2006
  • Loc: Valley of the Sun

Posted 14 November 2012 - 06:52 PM

Yes unchecked is gamma 1 which sets the register to 1024.

Glenn

#19 bunyon

bunyon

    Gemini

  • *****
  • Posts: 3323
  • Joined: 23 Oct 2010
  • Loc: Winston-Salem, NC

Posted 14 November 2012 - 08:09 PM

Awesome, thank. I'll try it out next time it's clear.

#20 BKBrown

BKBrown

    Gemini

  • *****
  • Posts: 3296
  • Joined: 22 Aug 2009
  • Loc: Northern Virginia, USA

Posted 15 November 2012 - 10:09 AM

That's a respectable set Paul, well done. I have occasionally run into the ringing issue when I stray from my regular gamma or gain setting parameters...

Clear Skies,
Brian






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics