Last night I was shooting at the moon and just for the heck of it I tried my 80-320mm on both cameras, first at 320mm and then added a Vivitar 2x TC, quite similar results, I know they are not too different of cameras but still i wanted to see how a 16mp sensor differs from the 24mp sensor. I edited in LR with the same values for each pic. Shot on a tripod. The last two pictures varies the speed of the shutter, my bad, i wanted to shoot both of them at 1/100s and i did not see that in the k3ii i moved the dial to 1/80s. I used the 2 second delay mode but it was a windy night so might have some movement from the wind affecting the cameras. The four images are cropped.

Pentax K-3ii
Aperture f/8
Time 1/125s
ISO 100
320mm


Pentax K-5ii
Aperture f/8
Time 1/125s
ISO 100
320mm


Pentax K-3ii
Aperture f/8
Time 1/80s
ISO 200
600mm


Pentax K-5i
Aperture f/8
Time 1/100s
ISO 200
600mm

Interesting exercice, I just bought a second hand K-70 and wanted to do just this with my K-5. Maybe if you do a 100% crop on both images it would be easier to compare. The K-3 moon will appear larger just because of its smaller pixels but I wonder if we could see more details.

These are the same images from above in the same order, just heavily cropped.

You will need to get better focus if you want to compare resolutions of the sensors with the same optics.

I second Lew's statement. The focus seems a bit off on all of them. How did you focus?

Wind is a killer and you will need a rock solid tripod and head to avoid shake from it so that may be a source of the problems.

Also you may have pushed the sharpening in light room too much since at ISO 100 the noise should be that noticeable. One thing I got pointed to a while ago was using the high pass to perform sharpening for astro images. Pete_XL was the one who told me about it and I like the results I get more than using what ever is done in LR or Photoshop when using their built in sharpening, the unsharp mask, or even rawtherapee's RL deconvolution. Here is the youtube video that Pete provided:

Thanks, at 600mm it's very difficult and atmospheric conditions come into play. I did a quick try in my backyard during lunch break and found there is a significant difference between my K-5 and my K-70, even with a 18-55 kit lens. And pixel shift brings about as much sharpness as the 24 Mpx brings compared with the 16 Mpx.

I was never able to get pixel shift to work on the moon. It moves too fast, even with short exposures. My understanding is that it works well if you have a well-aligned tracking mount, but otherwise no go.

The focus was achieved auto and manual, but i think the wind was affecting the stabilization, hopefully these coming days the wind will slow down a bit.

Maybe see about using a Bahtinov mask for focusing. I use them even when focusing on the moon. For things with a resolvable disk like the moon or planets focus on a star and then recompose on the actual subject. For generating one I use this site which will produce a SVG file. Each mask works best for a specific open aperture and focal length but will work well enough for ones between 1/2 and 2x the focal length but get beyond than and they really don't work well at all. It is best to have one specific to each lens. If you have access to a 3d printer you can convert the svg to a STL file which is what most 3d printers want by using this site and depending on the size of the mask you will want to extrude it to a thickness of 3mm to 5mm for strength.

If you are concerned that the star you used for focusing is much farther away than the moon is, don't be because from the perspective of your lens they are both at infinity. Since the moon is about 240,000 miles away it is well beyond the general rule of everything beyond 300x the focal length of the lens is at infinity. When looking this up as I had read somethign like it some where I saw values being used between 200 and 500 so 300 seemed good enough Even if one were to up that by an order of magnitude to 3000 times the focal length, just to be extra sure, the moon is still well beyond even that greatly inflated number. So to make the math easy lets use 10,000 times the focal length and your 600mm setup, this would make things at 6,000,000mm, 6KM or about 4 miles, and beyond all be at infinity from the perspective of the lens and here we used a number 33 times that of the general rule. Now I realize that this was purely an exercise in because I can but we can take away that anything that is at least as far away earth orbit should be treated as being at infinity with ease unless you have some silly big optics but no one here has any scopes with focal lengths in the 10s of meters so it doesn't matter.

These are from tonight´s full moon, again comparing the k-3ii with the k-5ii, the lens used 80-320mm. The first two pictures at 320mm and the last two using a Vivitar 2x TC.

K-3 II
Focal distance 320mm
Aperture f/8
Time 1/500s
ISO 100


K-5 II
Focal distance 320mm
Aperture f/8
Time 1/500s
ISO 100


K-3 II
Focal distance 640mm
Aperture f/8
Time 1/125s
ISO 100


K-5 II
Focal distance 640mm
Aperture f/8
Time 1/125s
ISO 100