[cam_abyss]

Hey pentax people,

I'm currently waiting for my first pentax to arrive... so I figured I'd join the forum.

I am selling my iOptron Skytracker tonight but before I do so, I wanted to ask a question about star-trailing with the astrotracer function of the K1:

It seems pretty well documented that the optimal focal lengths for the astrotracer are somewhere between 50-300mm...

So, for anyone using an ultra wide angle lens (i.e. Rokinon 14mm on the FF K1), my question is will the astrotracer still give me any extra time without producing trails? I've seen a lot of images of the astrotracer being used with the 14mm and what you see is star trailing in the corners. According to the 500 rule, I could avoid star trails all together with an exposure of < 35 sec (so let's just say 30 sec). If I turned on the astrotracer and set the exposure for one more stop (60 sec), will I still start to see the trailing in the corners? Or will the astrotracer function work well enough to give me maybe a 60 sec exposure without trails?

Hoping someone has some experience with this...

[Adam]

You'll likely always see a bit more movement in the corners when shooting that wide, but at 60s you should still have a perfectly usable photo.

[cam_abyss]

You'll likely always see a bit more movement in the corners when shooting that wide, but at 60s you should still have a perfectly usable photo.

Do you know if a panorama would help fix that?

Let's say I'm doing a pano of the whole milky way... If I left a little more overlap between frames, would photoshop or LR or other pano software replace the trailed stars with the stars from the center of the frame on the adjacent image?

[Adam]

Do you know if a panorama would help fix that?

Let's say I'm doing a pano of the whole milky way... If I left a little more overlap between frames, would photoshop or LR or other pano software replace the trailed stars with the stars from the center of the frame on the adjacent image?

I think even a slight crop would alleviate the issue. You could certainly try stacking but there may be some manual work involved since everything is constantly moving.

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com server and development costs are user-supported. You can help cover these costs by donating or purchasing one of our Pentax eBooks. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, KEH, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

[cam_abyss]

I think even a slight crop would alleviate the issue. You could certainly try stacking but there may be some manual work involved since everything is constantly moving.

Good point... slight crop and then process for panorama seems like a viable option. I just don't find myself lugging around my equatorial mount as much as I'd like to so I wanted to make sure I could still get some solid results before getting rid of it. Thanks

[photoptimist]

Astrotracer works best on rectilinear lenses. If the lens has barrel or pincushion distortion, astrotracer's movements will create slight trailing in proportion to the amount of lens distortion and the proximity of the pointing angle to the ecliptic. If you're pointing directly at the North Star, astrotracer should prevent trailing even with a fisheye lens.

[cam_abyss]

Astrotracer works best on rectilinear lenses. If the lens has barrel or pincushion distortion, astrotracer's movements will create slight trailing in proportion to the amount of lens distortion and the proximity of the pointing angle to the ecliptic. If you're pointing directly at the North Star, astrotracer should prevent trailing even with a fisheye lens.

It is a rectilinear lens... and distortion is supposedly corrected for pretty well... though I'm guessing there still probably is some

[photoptimist]

It is a rectilinear lens... and distortion is supposedly corrected for pretty well... though I'm guessing there still probably is some

Photozone (Samyang 14mm f/2.8 IF ED UMC Aspherical - Full Format Review / Test Report - Analysis) measured the distortion to be 5% which to a first approximation will leave about a 15 pixel trail* in the corners after a 1.5 mm astrotracer traversal of the sensor. That said, I'm sure astrotracer will enable somewhat longer exposures without a noticeable trail but it won't handle as long an exposure as one might think given the focal length and range of sensor motion.

(* That's just a quickie calculation)

[CDW]

I've found my iOptron tracker to deliver better images, particularly with longer focal lengths. The Astro Tracer is nice for situations where you don't want to carry the iOptron but I use both, depending on the circumstances.

[JimmyDranox]

IMO, I would postpone the selling of iOptron. At least until you will see if 60 seconds of exposure are enough to reveal what you want in your photos. If you want to capture the whole beauty of the Milky Way, 60 s could be not enough. And staking a large panorama could prove to be difficult.

[Digitalis]

It depends on lattitude and longtidute. The 500/f is only a rule of the thumb and due to the way our planet rotates, with exceedingly wide focal lengths star trails in the corners will be inevitable. The formula I use is as follows: 666/14 =47.571 seconds before trails would appear.

[cam_abyss]

It depends on lattitude and longtidute. The 500/f is only a rule of the thumb and due to the way our planet rotates, with exceedingly wide focal lengths star trails in the corners will be inevitable. The formula I use is as follows: 666/14 =47.571 seconds before trails would appear.

I've always err'd on the shorter shutter speed but I just came across a shutter speed calculator today that takes into account pixel tolerance, sensor, and focal length.

Check out this calculator: https://www.tl-photography.at/stars/night-sky-photography-shutter-speed-calculator/

[Gimbal]

It is a rectilinear lens... and distortion is supposedly corrected for pretty well... though I'm guessing there still probably is some

The thing with wide rectilinear lenses are that they do distort, allot. They have to in order to be rectilinear. Everything gets enlarged at the sides and corners of the picture compared to the center. Take a picture of a group of people and those standing at the very edge of the frame will be partially enlarged and look, well weird.

This distorsion will make stars that move across the frame appear to move at different speeds, The stars in the corners of the lens will move faster then the stars in the center of the lens. Astro tracer can't compensate for this, the image sensor can only rotate/move and compensate for one speed, and that is the speed in the center of the frame.

To compensate fully the whole camera and lens will have to move so that the stars don't appear to move, so the stars in the corner stays in the corner of the lens.

[lister6520]

It is not just a matter of rectilinear or not. With linear sensor shift it is never possible to accurately track the angular motion of the subject over the whole sensor. The issue is not only with Astrotracer but even with shake reduction. The only reason we don't worry about it with shake reduction is that with wide angle lenses shake is not usually much of an issue anyway.

It can be geometrically shown that shifting the sensor can never fully compensate for an angularly moving target regardless of how perfect the lens is. Briefly explained, if a target at the centre of the image moves by x degrees it will correspond to a linear movement of y pixels on the sensor. However a target towards the edge of the image moving the same x degrees will shift more than y pixels on the sensor. (The amount is determined by the trigonometric 'tangent' function)

The further (in degrees) the target is from the lens axis the larger will be the linear shift (in terms of pixels or microns) on the sensor for a given focal length. The longer the focal length (and hence the narrower the field of view) the less will be this difference, so for instance at 300mm there is hardly any difference between the centre and the edges of the image.

This all applies to a theoretically perfect rectilinear lens - it is a function of physics and geometry and not imperfections in the lens.

Any lens distortions and imperfections will of course add onto this problem as well.


To get good very long exposures with wide angle lenses you will need to limit the individual exposure times and re-aim the camera before each image of the stack. The re-aiming doesn;t need to be extremely accurate but if you can get an easy to see star at the centre of the view finder it makes it a lot easier.




PS: now only after posting I realise that there were more posts after the one I was responding to. My apolgies for repeating a lot of what has already been said and explained in those subsequent posts

[cam_abyss]

PS: now only after posting I realise that there were more posts after the one I was responding to. My apolgies for repeating a lot of what has already been said and explained in those subsequent posts

No apologies needed... well-explainedhttps://www.pentaxforums.com/forums/images/smilies/wink.png