[Dr_who]

Dr_who -- that's a nice tip!

I've yet to get good debayering from PI on my K10D. I've found a good system of calibrating via Maxim, then stacking the undebayered fits through DSS - including the debayering at this step, then doing all the other work in PI. Nothing else seems to handle the non-linear amp glow of the K10D properly.

If you can post 3 raws online somewhere I can download them I can fiddle with it and see if I can get it to debayer properly.

[k5astro]

woah! Great image krp! Thanks for showing what can be done. I'm just getting started in astrophoto and bought a Vixen Polarie. I highly recommend it for those who don't have a telescope - and I don't think you need a scope for a lot of beautiful astrophotos. Anyway, it will give pinpoint stars at 70mm with 2 min exposures, without any special alignment.

[VisualDarkness]

You guys are crazy skilled. I can't even get close to your work despite trying and trying!
This was the best I could do with a Sigma 20/1.8 and the K-5 on a tripod this clear night. I don't get the cone shaped stars I got, lens issue? It's a stack with 40 images and no dark frames (I actually forgot to take them). Is 20mm too wide to get anything interesting?

20mm
F1.8
40x10s
ISO6400

[Dr_who]

Oval shaped stars are normal depending on the lens used. Here's a image I did of M31 https://www.pentaxforums.com/gallery/images/17786/1_M31_10_21_12_F.jpg

http://en.wikipedia.org/wiki/Coma_(optics) is a description of coma distortion.

It did turn out well enough you can see the Belt of Orion. What software are you using to stack, to me it mostly looks like your blacks are clipped in processing.

*Edit - I just did a stack from some images I took tonight and has similar results and just modefied my "s curve" in the luminance tab and it got way better. Hope that helps you

---

Last edited by Dr_who; 01-05-2013 at 01:38 AM.

[k5astro]

@VisualDarkness - yeah, thats about the extent of my astrophotography too. I'm hoping for a clear night to stack as many 1-2 min exposures as possible and see what I get.

(I use Deep Sky Stacker. Free, Fully automatic and works great!)

[VisualDarkness]

Oval shaped stars are normal depending on the lens used. Here's a image I did of M31 https://www.pentaxforums.com/gallery/images/17786/1_M31_10_21_12_F.jpg

http://en.wikipedia.org/wiki/Coma_(optics) is a description of coma distortion.

It did turn out well enough you can see the Belt of Orion. What software are you using to stack, to me it mostly looks like your blacks are clipped in processing.

*Edit - I just did a stack from some images I took tonight and has similar results and just modefied my "s curve" in the luminance tab and it got way better. Hope that helps you

@VisualDarkness - yeah, thats about the extent of my astrophotography too. I'm hoping for a clear night to stack as many 1-2 min exposures as possible and see what I get.

(I use Deep Sky Stacker. Free, Fully automatic and works great!)

I also use Deep Sky Stacker, wonderful little program! What is the highest ISO you guys dare to go with the K-5 when stacking about 40 pics?
The blacks shouldn't be clipped but my processing was frustrating to say the least so I guess I need to redo it. Actually the original pics are almost blown out so I guess I went way passed the limit for the light pollution. About the cone/oval stars I thought about coma but shouldn't they point to the center? Maybe my lens got a centering issue?

[Dr_who]

I also use Deep Sky Stacker, wonderful little program! What is the highest ISO you guys dare to go with the K-5 when stacking about 40 pics?
The blacks shouldn't be clipped but my processing was frustrating to say the least so I guess I need to redo it. Actually the original pics are almost blown out so I guess I went way passed the limit for the light pollution. About the cone/oval stars I thought about coma but shouldn't they point to the center? Maybe my lens got a centering issue?

I'd be willing to shoot ISO 800-1600 however I almost always shoot ISO 400, as I have a method of tracking and can get away with longer exposures without startrails. Doing some research I found this link for you VisualDarkness. If you scroll down to distortion, If you are shooting wide open it's more apparent then if you stop down to say F4. Anyhow might be worth a read/try. Astrophotography Without a Telescope

[VisualDarkness]

I'd be willing to shoot ISO 800-1600 however I almost always shoot ISO 400, as I have a method of tracking and can get away with longer exposures without startrails. Doing some research I found this link for you VisualDarkness. If you scroll down to distortion, If you are shooting wide open it's more apparent then if you stop down to say F4. Anyhow might be worth a read/try. Astrophotography Without a Telescope

"Give up, startrails are cool too" sounds like the most promising one.
Though I really need to find a place with low light pollution that doesn't mist up every winter night around here.

[smigol]

Here's my latest work using a converted and cooled K10D.




This is the total of about 13 hours of integration time over two months. The horsehead here is a tight 100% crop of the full field. Most of the sessions were used to test for flexure. That's why I gathered so much data on the same target.

[Ex Finn.]

Here's my latest work using a converted and cooled K10D.




This is the total of about 13 hours of integration time over two months. The horsehead here is a tight 100% crop of the full field. Most of the sessions were used to test for flexure. That's why I gathered so much data on the same target.

That is some dedication to the hobby, nice image. How is the cooling done, is it a peltier directly connected to the sensor or is the whole camera cooled?

Cheers..

[k5astro]

I'm wondering about boosting the ISO. Wouldn't it be best to just capture at the native ISO of the camera (which I think is usually the lowest ISO)? Since higher ISO's just amplify the light that has been already captured. Post processing could be used to bring up the image brightness, no?

[smigol]

That is some dedication to the hobby, nice image. How is the cooling done, is it a peltier directly connected to the sensor or is the whole camera cooled?

Cheers..

I have a peltier device pressed against the LCD with some aluminum sheeting wrapped around the bottom of the body as well. This covers the hot spots on the outside of the camera. Helps to stabilize the camera temps and brings it down nicely. Doesn't get lower than ambient - or below freezing. The biggest advantage is the steady temp - makes using darks much easier.

[smigol]

I'm wondering about boosting the ISO. Wouldn't it be best to just capture at the native ISO of the camera (which I think is usually the lowest ISO)? Since higher ISO's just amplify the light that has been already captured. Post processing could be used to bring up the image brightness, no?

Native ISO is not necessarily the lowest setting. You would need to do some testing to find the unity gain value (where the value of electrons vs ADU is about 1). This also assumes that read noise is also amplified which is common among CMOS sensors.

There's been extensive testing of Canon and the general consensus is that ISO 400 to 800 is the ideal range because of the balance of unity gain vs the shenanigans that happens to the RAW files by the in camera software. Someone would have to do the same testing on the Pentax cameras to determine the proper settings.

In my testing for the K10D, I've settled on ISO 400 because it's a reasonable balance of noise and exposure length for my locations. A 10 minute exposure usually shows skyglow and thus there is no need to go deeper.

[Dr_who]

My 1:1 crop of the horsehead nebula has quite a bit more noise then yours. Thou I'm still quite happy with it as it is for only 77minutes, and used no darks/flats/or biases. I look forward to doing my second attempt on the nebula sometime in the future!






I did some more tweaking and posted another image of my IC 434 region. It's in this thread: https://www.pentaxforums.com/forums/pentax-dslr-discussion/210829-first-astro...ml#post2230696

[k5astro]

Native ISO is not necessarily the lowest setting. You would need to do some testing to find the unity gain value (where the value of electrons vs ADU is about 1). This also assumes that read noise is also amplified which is common among CMOS sensors.

There's been extensive testing of Canon and the general consensus is that ISO 400 to 800 is the ideal range because of the balance of unity gain vs the shenanigans that happens to the RAW files by the in camera software. Someone would have to do the same testing on the Pentax cameras to determine the proper settings.

In my testing for the K10D, I've settled on ISO 400 because it's a reasonable balance of noise and exposure length for my locations. A 10 minute exposure usually shows skyglow and thus there is no need to go deeper.

Thanks for the info! I'll try some different ISOs (on my K20) to see what works best. --Cheers