The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Help Understand What I'm Seeing - IQ3100 & a7rii Star Photos

Craig Stocks

Well-known member
First of all, I'm not trying to do a "which is better" type of comparison, I'm simply trying to understand the differences in the two systems so I can get the most out of each one. I'll be travelling this winter with just the a7rii so I want to play to its strengths.

Note that my a7rii is still on firmware 3.10 which is before the "Star Eater" issue was introduced.

I went out last week to do a star trail shot under a bright waxing gibbous moon and since I'd have plenty of down time I set up both cameras side by side and nearly identical settings. a7rii had a Sony 12-24 zoom @ 12mm and was at ISO 500, 60 seconds @ f/5.6. The IQ3100 had a P1 28mm and was at ISO 400, 60 seconds @ f/5.6. I shot 60 frames with each to stack into star trails. I like the a7rii version better so that one has had more work done on it.





What I noticed is that the IQ3100 version has lots more trails from small stars. I don't know that it necessarily makes a better star trail images, in fact I think I prefer the a7rii version with fewer trails, but it's a difference worth noting.

That led me to wonder about standard night sky images. Since I use both systems I want to understand the differences. I normally do night sky photos with a tracking mount so that's where I started last night before the moon came up. The a7rii had the 12-24 @ 12mm, ISO 1600, 30 seconds @f/4.5. The IQ3100 had the 28mm, ISO 1600, 30 seconds @ f/4.5. Both used the tracking mount. I processed both in Capture One (primarily brightening about 1.8 stops and adjusting white balance). They were then opened in Photoshop where I stack the two frames and transformed the IQ3100 frame so the stars were aligned. This is a crop of the center area of the result.





I'm sure there are other differences but the two that jump out to me are that the IQ3100 shows many more stars but they are all smaller. This is consistent with what I saw in the star trail image. Also there is quite a bit more detail and nuance in the Milky Way in the IQ3100 image. (Also, yes, I know I should have zoomed the Sony to around 17mm for a better match but I didn't think of it at the time. In practice I'll use it at 12mm.)

I then switched to 24mm equivalent lenses and did the same shots only at f/4.0 this time. I also added my a6000 with a 16-70 zoom at 16mm. I also processed and stacked them the same way. The big differences is that (unbeknownst to me) the batteries in my tracking mount gave up in the cold so I only have comparison frames untracked. 30 seconds is a little long for a 24mm eq. focal length but the difference is what's important.







I have noticed in the past that untracked star photos look better than tracked ones, at least until you blow them up. The stars get a little smudged and show up more readily and that seems to be the case here. I still see the difference in the fine details of the Milky Way but note that there is now much less difference in the number and character of the stars, in fact almost no difference at all.

What I don't understand is why. I'd like to explain the difference by the blurring of the stars without tracking minimizing any star-eating tendencies of the a7rii but that should have applied to the original star trail image.

I think in reality I could probably be happy with any one of the photos (except maybe the a6000 because of noise) if that's all I had. I like that the a7rii makes bigger stars and shows more variation in brightness (size) of the stars. I like that the IQ3100 captures more stars, and I can induce the variation in size with my Minimum and Maximum filter application in Photoshop.

I'm curious to get other folks thoughts.
 

tjv

Active member
I've never taken a star trail photo in my life, but to my eyes I prefer the IQ images. The stars look more natural, with less blooming (if that's the right word?) around the actual stars. I think that's what makes the sony stars appear bigger.

Anyway, I also think the Sony images look noisier. But that's just me.

I'm stuck in the relative 'dark ages' with a 60mpx CCD back, so long exposures like this aren't even on my radar.
 

JohnBrew

Active member
I've made star trail images and I definitely prefer the IQ star trails image. The other shots, imo, are all unimpressive enough to simply dismiss.
 

anwarp

Well-known member
Hi Craig

I've never tried start trails myself, and in any case, I'm still in the dark ages too with an IQ260.

Playing the armchair detective - the fact that you are getting more visible stars with the "smudged" image, makes me wonder if it's something to do with single pixel noise reduction for the Sony profile in C1?

Anwar
 
Last edited:

thrice

Active member
Yes that's true, but my A7r2 has not been updated to the firmware that created the "star eater" problem.
Hi Craig,

I tested every firmware which offers uncompressed RAW and they all have spatial filtering (star eater).
For some reason everyone thinks it was introduced in 3.30 but it was there since 2.0!
 

Craig Stocks

Well-known member
I suspect a lot of the difference I'm seeing has more to do with lenses than sensors. The 28mm Phase One is a pretty poor lens for night skies. It's pretty sharp in the center but has lots of coma aberration in the outer 1/3 of the frame. In my comparison it was compared to the Sony 12-24mm which is a pretty good lens with much less coma but does seem a little softer - which has the effect of creating slightly larger stars (my theory) The other lenses I compared were a Phase One 35mm BR and the Sony 24-70mm GM, both of which are great lenses. The comparison with those lenses showed virtually no difference in the number of stars but since my tracking mount had stopped the results aren't really representative of preferred shooting conditions.

So far in my limited experience I seem to get the best results from the IQ3100 with the 28mm on a tracking mount. I then stack 5 to 10 frames for noise reduction and manually edit the stars with significant coma aberration. I also run a sequence of minimum and maximum filters to create more differentiation between bright and dim stars which also makes most stars slightly larger and more round.
 
Top