Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

An Out-Of-Focus Widefield of the Scorpio Region

  • Please log in to reply
17 replies to this topic

#1 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 26 May 2020 - 11:14 AM

Anyone have some tips for nailing focus with a wide, fast lens?

 

I shot this a couple of weekends ago with my modded Canon 600D, Skyguider Pro, and Sigma Art 50mm f/1.4 at f/1.8. I thought I did a decent job of focusing... but nope. I got 108 90-second light frames at ISO 800, calibrated with 30 each darks, flats, and dark flats. Stacked and processed in Astro Pixel Processor.

 

Unfortunately when I stretched the image, donut-shaped stars were evident in the middle of the frame, and were smeared out into little comets at the corners. It isn't too bad scaled down for the forum, but at full resolution... it's just bad.

 

I really like how this lens performs when I can hit focus, but my success rate hasn't been great. My Bahtinov mask does not create large enough spikes to be useful at 50mm, so I've been focusing on dim stars with trying to maximize their brightness. I just don't think the way I'm doing it is precise enough below f/2. Everything looks fine blown up on the back-of camera LCD, but not once displayed on a full-size monitor. I also think the focus slipped through the night on this run, so I'll have to try to tape it down.

 

I'm sure many of you folks have used wide, fast lenses remotely where a full-size display is not practical. What do you do to make sure you haven't wasted a clear night by shooting a bunch of out-of-focus light frames?

 

Here's the quick processing, warts and all.

RO_50mm_rescale2.jpg


  • Starman27, aa5te, mikewayne3 and 4 others like this

#2 aa5te

aa5te

    Genial Procrastinator

  • *****
  • Posts: 1,871
  • Joined: 30 Aug 2008
  • Loc: Clinton, TN

Posted 26 May 2020 - 11:53 AM

I'm not sure what your Canon has, but my Sony A6300 has a "focus peaking" feature where it'll display dots (you pick the color - white, red, or yellow) around the edges and contrast lines of objects to help you focus. The closer to focus you are, the more dots it displays. I couple this with the Live View zoom feature to zoom in to 11.9x (5.9x is also an option) and then use dimmer stars to maximize the dots. I bump ISO to 1600, exposure to 30 sec., aperture wide open, do what I mentioned, then take a test shot. Works with my Rokinon 12mm f/2 and the 135mm f/2. If I can't get a good focus there for whatever reason, I'll try a tower light or a street light that's at least a few thousand feet away.


  • DubbelDerp likes this

#3 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 26 May 2020 - 12:36 PM

I'm not sure what your Canon has, but my Sony A6300 has a "focus peaking" feature where it'll display dots (you pick the color - white, red, or yellow) around the edges and contrast lines of objects to help you focus. The closer to focus you are, the more dots it displays. I couple this with the Live View zoom feature to zoom in to 11.9x (5.9x is also an option) and then use dimmer stars to maximize the dots. I bump ISO to 1600, exposure to 30 sec., aperture wide open, do what I mentioned, then take a test shot. Works with my Rokinon 12mm f/2 and the 135mm f/2. If I can't get a good focus there for whatever reason, I'll try a tower light or a street light that's at least a few thousand feet away.

Thanks for the suggestion! I don't think I have focus peaking on my Canon, but it looks like if I install Magic Lantern it'll give me that option. The bahtinov mask works great with my 135 f/2. Maybe I should give it another try with the 50mm lens and just up the exposure time.


  • aa5te likes this

#4 StarBurger

StarBurger

    Viking 1

  • -----
  • Posts: 505
  • Joined: 06 Mar 2018
  • Loc: North Country NY

Posted 26 May 2020 - 02:16 PM

Well DD, I think it looks great! I have attempted similar images with my 600D with varying results. Taping down the focus ( on far off daylight terrestrial objects) was the best improvement as was a higher f ratio. The comets on the edges of the field are likely the lens aberrations. I also noticed that the lens iris vanes (if your lens has the same) do tend to worsen the tightness of the stars with the flaring they create.

And yeah, I could not see good Baht mask spikes on even the brightest of stars.

Stock camera lenses are not the best for this kind of wide angle work. 

You should be proud of the overall result. Without extreme zooming on the image (if there is a focus problem? My zooming was not enough to reveal it on the jpg. posted) it looks fine. The colors show well and the framing excellent.

Well done. I don't believe it can be improved much more.


  • DubbelDerp likes this

#5 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 26 May 2020 - 02:46 PM

SB, thank you for the kind words! The stars do look a lot better when I reduce the image size. The focus problem really shows when I tried to integrate some lights from my Rokinon 135mm f/2 with the 50mm data, and even with the background colors matching, the tighter stars from my better focus with the 135mm really makes this part of the image stick out like a sore thumb. Here's a screenshot from the minimally processed version:

 Annotation 2020-05-26 153711.jpg

 

If I stop the lens down to f/2.8 or so, it is much easier to focus since it isn't quite so touchy, but I've been trying to learn how to use this lens to its potential by opening up the aperture more and more. I figure there's not much sense in paying for a Sigma Art lens and then tossing photons by stopping it down.



#6 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 6,147
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 26 May 2020 - 06:10 PM

It's a digital camera...take a shot...review it at full 100% playback....Is it sharp ? Rinse and repeat until it's as sharp as you can get....It only takes one-two minutes and will save a lot of wasted hours shooting OOF.



#7 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 26 May 2020 - 06:27 PM

Yup... like I said in the OP, Everything looks fine blown up on the back-of camera LCD, but not once displayed on a full-size monitor.

#8 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 6,147
  • Joined: 21 Mar 2013
  • Loc: Colorado,USA

Posted 26 May 2020 - 11:49 PM

Yup... like I said in the OP, Everything looks fine blown up on the back-of camera LCD, but not once displayed on a full-size monitor.

Time for a new camera then...

If my review is sharp on my LCD it's sharp on my monitor.......100% is 100%

I'd taken my laptop out with me to take this photo , on my very first test run at my now darksite.. I forgot a cable, and couldn't connect to the laptop, so everything had to be done the "Old Fashioned" way, focus on BOC lCD and and a wired intevalometer that I luckily had in my tote box...90 min total, f4....

27528737207_dab175d329_c.jpg


Edited by 17.5Dob, 26 May 2020 - 11:58 PM.

  • ewave likes this

#9 t-ara-fan

t-ara-fan

    Apollo

  • -----
  • Posts: 1,218
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 27 May 2020 - 01:19 AM

Anyone have some tips for nailing focus with a wide, fast lens?

 

 donut-shaped stars were evident in the middle of the frame, and were smeared out into little comets at the corners

The LCD on my Canon cameras doesn't zoom in far enough to see if focus is slightly off. And if it did, at 288 pixels-per-inch I couldn't see it clearly anyway.

 

With my 200mm f/2.8 lens, I just focus looking at a bright star in live view.  Your faster and shorter FL lens is going to have a tiny zone of perfect focus.

 

Do you have a laptop with you when in the field? APT or BackyardEOS are great, and make focusing very easy because (a) electronic control of the lens' focus motor from the laptop moves in much smaller steps than you can do by hand and (b) FWHM and HFR tools can help you dial it in.

 

Bonus: view the image at 400% on your laptop to really pixel peep.

 

I suspect that 50mm f/1.4 will never have pinpoint stars in the corners.
 


Edited by t-ara-fan, 27 May 2020 - 01:20 AM.

  • DubbelDerp likes this

#10 BQ Octantis

BQ Octantis

    Mercury-Atlas

  • *****
  • Posts: 2,706
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 27 May 2020 - 05:25 AM

The trouble with really fast lenses is that the axial CA wide open leaves too much to chance while focusing. The solution? Aim at the brightest star nearby, set the aperture to f/8, zoom to 10x, and press the aperture preview button while focusing. Instead of trying to turn a white star into a magenta donut, you just bring the white star to a point. Even if you don't need them, use a pair of reading glasses to pixel peep on the screen while you do it. Then open up the aperture and shoot. Camera lenses are parfocal, so optimum focus is practically the same across apertures.

 

As to comet stars at the edges, you'll need to use the clone tool with a donut-shaped brush to paint them into shape.

 

BQ


  • DubbelDerp likes this

#11 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 27 May 2020 - 08:30 AM

Time for a new camera then...

If my review is sharp on my LCD it's sharp on my monitor.......100% is 100%

I'd taken my laptop out with me to take this photo , on my very first test run at my now darksite.. I forgot a cable, and couldn't connect to the laptop, so everything had to be done the "Old Fashioned" way, focus on BOC lCD and and a wired intevalometer that I luckily had in my tote box...90 min total, f4....

27528737207_dab175d329_c.jpg

Nice image. But it's not the camera, it's the f/1.4. That must be an 85mm lens at f/4?



#12 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 27 May 2020 - 08:38 AM

The LCD on my Canon cameras doesn't zoom in far enough to see if focus is slightly off. And if it did, at 288 pixels-per-inch I couldn't see it clearly anyway.

 

With my 200mm f/2.8 lens, I just focus looking at a bright star in live view.  Your faster and shorter FL lens is going to have a tiny zone of perfect focus.

 

Do you have a laptop with you when in the field? APT or BackyardEOS are great, and make focusing very easy because (a) electronic control of the lens' focus motor from the laptop moves in much smaller steps than you can do by hand and (b) FWHM and HFR tools can help you dial it in.

 

Bonus: view the image at 400% on your laptop to really pixel peep.

 

I suspect that 50mm f/1.4 will never have pinpoint stars in the corners.
 

Thanks! I didn't think about using the lens AF motor to focus through BYEOS. I've gone through my 30-day trial with the software, but if I can focus using the AF motor, that by itself would justify the full license. Can you autofocus through BYEOS, or do you have to manually do it from time to time? I suppose if the motor is activated, it would keep focus locked in position. Just a slight drag from a dew heater I suspect caused focus to shift in the field.

 

The data I got with my 135mm f/2 came out much cleaner, thanks to the bahtinov mask...  https://www.astrobin.../full/917bja/0/

 

It probably wouldn't have helped with this shot since I had to hike across my property to get a clear view to the south, but I can certainly give it a try when I set up my tracker within range of AC power.



#13 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 27 May 2020 - 08:42 AM

The trouble with really fast lenses is that the axial CA wide open leaves too much to chance while focusing. The solution? Aim at the brightest star nearby, set the aperture to f/8, zoom to 10x, and press the aperture preview button while focusing. Instead of trying to turn a white star into a magenta donut, you just bring the white star to a point. Even if you don't need them, use a pair of reading glasses to pixel peep on the screen while you do it. Then open up the aperture and shoot. Camera lenses are parfocal, so optimum focus is practically the same across apertures.

 

As to comet stars at the edges, you'll need to use the clone tool with a donut-shaped brush to paint them into shape.

 

BQ

Thank you for the suggestion! I'll give that a try next time in the field. Part of the problem with getting a new lens are figuring out the nuances of it. I'm just getting the hang of the 135mm, but so far this one has been a struggle.

 

When I actually managed to hit focus, the corner stars aren't misshapen enough to bother me. But if the center of the field isn't sharp, the edges just look so much worse.



#14 t-ara-fan

t-ara-fan

    Apollo

  • -----
  • Posts: 1,218
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 27 May 2020 - 10:02 PM

  Can you autofocus through BYEOS, or do you have to manually do it from time to time? I suppose if the motor is activated, it would keep focus locked in position. Just a slight drag from a dew heater I suspect caused focus to shift in the field.

 

The data I got with my 135mm f/2 came out much cleaner, thanks to the bahtinov mask...  https://www.astrobin.../full/917bja/0/

 

It probably wouldn't have helped with this shot since I had to hike across my property to get a clear view to the south, but I can certainly give it a try when I set up my tracker within range of AC power.

I believe NINA can do AF with a DSLR. I have heard a lot of great things about NINA, haven't tried it yet because SGP works.

 

With ByEOS the lens must be in AF mode so the motor can focus. Then I switch the lens to MF so the camera doesn't try to focus when taking an exposure.
 



#15 deepanshu29

deepanshu29

    Mariner 2

  • *****
  • Posts: 262
  • Joined: 16 Jul 2017

Posted 27 May 2020 - 10:27 PM

You can use zoom in live view, that will give you max 10x on your 600D. I would also suggest pressing down and holding the depth of field button while you are focusing. By design, lens does not stop down to set aperture until the exposure is started. So even if you set it to f/2, your lens is still at f/1.4 while you are adjusting the settings on camera. I would also suggest using ISO 3200 or 6400 for focusing, then you can step down. Exposure simulation work good enough for stars to show up at f/2, ISO 6400, 30s Manual mode. 

 

Also, NINA can not autofocus with DSLR & lens using the internal drive. APT and ByEOS both can do that, you will need to have lens in AF mode like mentioned above. 


  • DubbelDerp likes this

#16 t-ara-fan

t-ara-fan

    Apollo

  • -----
  • Posts: 1,218
  • Joined: 20 Sep 2017
  • Loc: 50° 13' N

Posted 27 May 2020 - 10:28 PM

 Aim at the brightest star nearby, set the aperture to f/8, zoom to 10x, and press the aperture preview button while focusing.

OK. But at f/8 don't you have a wider zone where focus is good? And at f/1.4 the zone is so narrow, you could be out of it yet look great at f/8?

 

 



#17 BQ Octantis

BQ Octantis

    Mercury-Atlas

  • *****
  • Posts: 2,706
  • Joined: 29 Apr 2017
  • Loc: Red Centre, Oz

Posted 28 May 2020 - 03:35 AM

OK. But at f/8 don't you have a wider zone where focus is good? And at f/1.4 the zone is so narrow, you could be out of it yet look great at f/8?

You certainly have a bigger depth-of-field (to use a familiar term) at at higher focal ratios—my f/15 Mak-Cass is way easier to focus than my f/6 SCT in spite of it being 3.6 times longer—but the Bahtinov is the crucial indicator for shorter optics. If the optic is too small for a Bahtinov to work, then the circles of confusion from being slightly out of focus are small enough to be below the detection threshold of the sensor. Indeed, I can feel the stepper steps in my 50mm f/1.8, and at f/8 a bright star comes to a point at one and only one step.

 

BQ


  • t-ara-fan likes this

#18 DubbelDerp

DubbelDerp

    Viking 1

  • *****
  • topic starter
  • Posts: 699
  • Joined: 14 Sep 2018
  • Loc: Upper Peninsula of Michigan

Posted 28 May 2020 - 09:04 AM

You can use zoom in live view, that will give you max 10x on your 600D. I would also suggest pressing down and holding the depth of field button while you are focusing. By design, lens does not stop down to set aperture until the exposure is started. So even if you set it to f/2, your lens is still at f/1.4 while you are adjusting the settings on camera. I would also suggest using ISO 3200 or 6400 for focusing, then you can step down. Exposure simulation work good enough for stars to show up at f/2, ISO 6400, 30s Manual mode. 

 

Also, NINA can not autofocus with DSLR & lens using the internal drive. APT and ByEOS both can do that, you will need to have lens in AF mode like mentioned above. 

Well that's something I didn't think about. I guess that's what I get for only using manual and vintage lenses - it never occurred to me to have to use the depth-of-field preview. And it looks like APT has the capability of autofocus with EOS lenses. That's pretty interesting.


  • deepanshu29 likes this


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics