[DavidOliver]

For testing purposes I shot the night sky for the first time the other night, with a 16mm Samyang lens. For the last shot I tried the astro tracing feature, and here's the 300-second exposure's result. I can see that the sensor was moved as the tree's branches, lower right, are blurry. (The other non-traced shots have sharp branches as it wasn't windy, and here's an 856-second non-traced exposure to show the sharp branches and trail directions.)

I've since realised that the star trails in the traced shot are not consistent with each other in length or direction; instead, the (erroneous) trails seem to visually originate/zoom out from somewhere near the centre of the frame. I then wondered if this is due to my only being able to select 15mm when inputting the focal length - not the actual 16mm of the Samyang.

I'll do more testing, but do you know if it's going to be impossible for me to use the astro tracing feature with a lens that doesn't communicate its focal length and that doesn't happen to match one of the focal lengths available for selection by the camera?

[Na Horuk]

I think you could get away with f4 or something; f8 makes even the 300second image rather dark.. You can probably use a higher ISO, as well. Yes, f8 and ISO 100 should give "best quality", but with astrophoto gathering as much light as possible is more important than that. And the lens and camera you are using are both good enough to use lower f number and higher ISO.
Anyway, make sure you are using the latest firmware on your gear.
DC shooter got the rest

I then wondered if this is due to my only being able to select 15mm when inputting the focal length - not the actual 16mm of the Samyang.

This could be part of the problem. For SR it is usually recommended you input the nearest wider focal length so the SR will undercompensate (be less effective) rather than overcompensate (potentially introduce extra blur). I am not sure if the astrotracer behaves in the same way, but it probably does? Maybe someone else knows more specifics

[Nass]

Agree with dcshooter about the sidereal aspect - I think the lens is so wide it captures paths that look like they go in different directions, so can never be 'stopped' as such. Here are some old diagrams I once did for another forum explaining the shape of trails you get and the direction.

[Adam]

There are a couple of issues. First, the astrotraceer does appear to be mis-calibrated, since there a trails even at the center of the image. 300 seconds is a fairly long exposure, even with such a wide angle, so if the calibration is off even slightly, you will get errors. You might get better results boosting the ISO and doing several shorter exposures (<1 min) and stacking to reduce noise.

This is what I would do. Pushing the exposure duration to the limits of the astrotracer will increase the likelihood of visible trails.

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:

[maxfield_photo]

I then wondered if this is due to my only being able to select 15mm when inputting the focal length - not the actual 16mm of the Samyang.

I think that's quite possible since 15mm is about 7% wider than 16mm, even though it's "only 1mm". That's a real bummer. Maybe something can be done from the debug menu?

[DavidOliver]

Thanks, everyone.

First, the astrotraceer does appear to be mis-calibrated, since there a trails even at the center of the image. 300 seconds is a fairly long exposure, even with such a wide angle, so if the calibration is off even slightly, you will get errors. You might get better results boosting the ISO and doing several shorter exposures (<1 min) and stacking to reduce noise.

However, even when calibrated, you will still see trails on the corners of an astrotraced image with a lens this wide since the corrections it provides are only in a flat XY axis along the sensor plane, while the apparent star movement is sidereal i.e. along the curved inner surface of a rotating sphere.

I'll watch some calibration videos and try performing it again. Apparently it can be inconsistent? I'm not even sure what exposure times the astrotracer is supposed to provide for. Anyway, more reading and experimenting required. Thanks for the note on the limitations, too.

I think you could get away with f4 or something; f8 makes even the 300second image rather dark.. You can probably use a higher ISO, as well. Yes, f8 and ISO 100 should give "best quality", but with astrophoto gathering as much light as possible is more important than that. And the lens and camera you are using are both good enough to use lower f number and higher ISO.
Anyway, make sure you are using the latest firmware on your gear.

Okay, I'll definitely try out those kinds of settings next time. There weren't any firmware updates available for the K-3 II last time I checked. (Only the K-3.)

Agree with dcshooter about the sidereal aspect - I think the lens is so wide it captures paths that look like they go in different directions, so can never be 'stopped' as such. Here are some old diagrams I once did for another forum explaining the shape of trails you get and the direction.

Thanks for sharing those diagrams. I've seen some shots showing those paths, like Elia Locardi's Church of the Good Shepherd shot. However, I think in this case it's not due to that as the paths of the stars in the non-traced shot aren't like that.

I think that's quite possible since 15mm is about 7% wider than 16mm, even though it's "only 1mm". That's a real bummer. Maybe something can be done from the debug menu?

I've done a little searching on Pentax DSLR debug menus and I haven't found any reference to setting focal lenghts, unfortunately.

My wife has the Pentax 12-24mm, so I'll also try some traced shots with that and compare. Thanks again.