Astrophotography Observations - IQ3100 & Sony a7R2

[Craig Stocks]

I've been doing some testing as I get ready for the Milky Way season. I admit I'm somewhat overly focused on star rendering as I'm trying to replicate that feeling you get when standing under a dark sky filled with stars.

I also recently ran across a series of discussions about the Sony's "star eater" problem where it implements single-pixel noise reduction for any exposure of 4 seconds or longer regardless of the LENR setting. The result is that it can see small stars as single pixel noise and erroneously reduce or eliminate them.

The examples here were shot back to back on my tracking mount with the XF - IQ3100 with 28mm and the Sony a7R2 with Canon 14mm for 30 seconds @ ISO 800, f/4.5 and f/4.0 respectively. Both cameras are very ISO invariant and 800 seems to work as well or better than higher ISO. All LENR was turned off. The sky is from 5 frames stacked in median mode, the foreground is a separate frame with tracking turned off. Similar processing was done in Capture One (for the IQ3100) and Lightroom for the Sony. Both setups have problems with coma and chromatic aberration that I have not addressed.

I've also included a screen shot from C1 showing a single frame from both cameras with not edits applied.

Notice the region above Orion where a dim portion of the Milky Way is visible. The IQ3100 captures a distinct field of tiny stars where the Sony does not. I've suspected the star eater phenomenon in other instances but this seems to show a clearly better result for the IQ3100.

Some other observations.

For a given exposure both cameras seem to record stars at about the same pixel dimension, such as 7 pixels wide for a reasonably bright star. The result is that stars actually print larger from the Sony for the same print size.

Most stars are recorded as pure white. Longer exposures simply record the star as a larger white spec. Of course, the stars don't really vary in apparent size nearly as much as they do in brightness but in photos brighter stars are shown as bigger rather than brighter.

The Sony seemed to do a better job of differentiating star colors but I believe a lot of that is false color from demosaicing small bright objects. Also Lightroom seems to be more prone to false colors around stars that C1.

Tracking and using a higher resolution camera seems to make the challenges worse. For instance it's common for the IQ3100 images to have such small star points that they're lost in the scene where if I use the Sony without tracking I get larger, more visible stars.

I don't know the solutions or best practices. I'm just sharing some observations in the hope of generating some discussion of other approaches.

 

[Paul2660]

The Fuji GFX albeit only 50MP may be a better solution as it's much lighter and you can adapt various 35mm lenses to it. The nikon ones except the 14-24 IMO have way too small a IC so you would crop 1:1 most times. However the Sigma Art fast lenses have a lot of coma towards the edges so the 1:1 crop might be OK.

I don't carry a tracking base, for me it's too much weight and time to setup, I know it works for others.

Way off topic, but the Pentax K1 still is my choice here as you can track with it internally due to the sensor stabilization and up to 5 minutes, which for me is way too much. I mostly track to 1.5 minutes to 2.5 minutes. Only real downsides are:

1. Noise, as Pentax did not fix the white dot issue on the 36MP chip like Nikon did, you have to pretty much plan on a dark frame each time.
2. Use of wider lenses (needed in my part of the the US) will not track well towards the edges and you will get a bit of trailing. Thus best to fit the M way in the center of the frame.

The Nikon D810A also gets rave reviews and if the price point was a bit lower I might try one. Very low noise, built in intervalometer (more for star trail work) and excellent higher ISO's up to 12K.

Both the Pentax and Nikon are 1/3 approx the resolution of the IQ3100, but for night work I don't find I need the extra resolution.

My experience with the A7RII is that it also gets massive amounts of noise worse than the K1 on the longer exposures especially when it's been on for a while. It has IBIS but no internal tracking app (last time I checked). C1 does the best on the longer exposures as LR just makes the noise worse. C1's single pixel noise reduction is a huge advantage for night work.

Paul Caldwell

 

[Don Libby]

I've shot tracking off a telescope last year as an experiment to see if it would handle the weight of the XF and found it would; this was prior to upgrading from the 180 to a IQ1-100 so the images were crappy. Since then I've used an IOptron "SkyTracker" that is much smaller and can handle the weight of the XF and a 28mm. Having switched to the IQ1-100 late last year I wasn't able to really get out as much as I'd have liked however I did manage a couple shots on tripod. I tried both the 28 and 35LS and found the 35 slightly better as it is just a little faster. Shortly afterwards I picked up a 45 f/2.8 and got even better results.

The upside to shooting off a Celestron 6" telescope is that it will handle the weight of any of my lenses and track for as long as I want. The downside is that it takes up just about all the bed of my truck to pack it in anywhere. The "SkyTracker" is limited in weight however it fits nicely in a Domke F-5XB bag.

I'll be shooting in the Valley of the Gods and Monument Valley shortly and plan on using the 45mm both on tripod and with the aid of the "SkyTracker". I also just picked up the Fuji GFX 50S and a 63 f/2.8 that I'll be trying out as well.


View attachment 126388View attachment 126389View attachment 126390View attachment 126391View attachment 126392

The last two images were captured using the IQ1-100 on tripod with no tracking


Don

 

[Craig Stocks]

What settings do you find work best with your various cameras? Last year I liked ISO 400 with 4 minute tracked exposures (IQ3100) and I tended to use ISO 1600 or 3200 with 15 second exposures with the Sony (not tracked). After doing more experimenting I'm leaning more toward stacked 30 second exposures at ISO 800 with both cameras. With both cameras ISO 800 with exposure boosted in post seems to be the sweet spot for noise though there is very little difference between boosting exposure 3 stops in post versus using ISO 6400 or any combination in between.

I've been using the Sky-Watcher Star Adventurer tracking mount (https://www.bhphotovideo.com/c/prod...r_s20520_star_adventurer_motorized_mount.html) for it's higher weight rating (11 pounds). It also has a good alignment scope built in. I also use the equatorial wedge base (https://www.bhphotovideo.com/c/prod...20530_star_adventurer_latitude_eq.html?sts=pi).

When possible I polar align based off of the north star using the offset position from the Polar Scope Align Pro app which gives you the exact sight picture for alignment. Of course that assumes you can see the north star. When I can't I align by dead reckoning using my phone's compass and the latitude scale on the base and it seems to work reasonably well for modest exposure times, at least it's quite a bit better than not tracking at all.

 

[voidshatter]

All LENR was turned off.

Thanks for the a-b comparison to show the star-eater effect of the Sony - great example!

How did you turn off the LENR of the XF+ system? Did you use DT's software?

 

[Craig Stocks]

How did you turn off the LENR of the XF+ system? Did you use DT's software?

I used the Time Lapse tool with dark frame set to "Suppress." It still did a dark frame on the first exposure but none after that.

I also looked back at the Sony frame with tracking turned off (for the foreground) and apparently there is enough smearing during a 30 second exposure for all of the stars to register since it didn't seem to have eaten any. It's another example of how a long-ish untracked exposure (like 15 seconds) solves a number of problems; you don't have to blend exposures and the stars appear more prominent since they're blurred. The tradeoff though is noise and "fuzzy" stars.

 

[voidshatter]

I used the Time Lapse tool with dark frame set to "Suppress." It still did a dark frame on the first exposure but none after that.

Thanks for confirming this. Taking a laptop as well as a tracking mount may not be a viable option for me, but may be easy for others.

"Layer-mask" work for trees (or other complicated foreground) may not be an easy job (and I can see that you didn't have patience with your A7R-II image). That's another reason why I'm reluctant to track the stars. Instead I prefer f/1.4 lenses of 35mm format so I can shorten the exposure time.

 

[dougpeterson]

Great report.

The IQ3 100mp sure does shine. Pun intended.

Note that someone mentioned you can adapt 35mm lenses to the Fuji... of course you can also do this with an IQ3 100mp since you can use the electronic shutter and a tech camera to use Canon/Nikon/etc lenses. Coverage will rarely take up the full 100mp sensor, but you'll get more from any given lens than you will on any other camera available. You can then use the DT Time Lapse Suite to control a time lapse capture with the electronic shutter on the tech camera.

The skies the limit. Again... pun intended.

 

[Craig Stocks]

I'd really like to find the magic balance of exposure and star size. Stars are pinpoints that vary in brightness. The camera tends to record all stars as white spots that vary in size, not brightness. The right size stars from the right distance create an illusion of varying brightness, but change the print size or viewing distance and the illusion falls apart. Big and little white spots are not the same thing as bright and dim points of light but it seems to be the only mechanism we have for prints on paper. There is probably a formula for optimum star size (in inches or mm) based on viewing distance. Too large and they look fake, too small and they get lost or look like noise.

For instance, I photographed a scene with Venus in a very prominent position with the Sony. Most stars were recorded around 6 to 7 pixels across but Venus was a blurry 40 to 70 pixels across depending on where you think it stops. So in a 36 inch print it would be between .25 and .5 inches across. Yes, it appeared bright in the sky, but certainly not large, and definitely not that large relative to other stars. In that case I reduced the size of Venus so that it was only 2 or 3 times the size of the other bright stars and left a bit of glow around it.

Changing star size is rather easy. Longer exposures record the stars as larger, not brighter. In post processing you can adjust the sizes using Photoshop's minimum and maximum filters (which can also make them more round and more clearly defined if you want).

If you haven't seen it this is a good discussion of how NASA and JPL scientists use Photoshop to process their images. https://blogs.adobe.com/conversations/2015/09/how-photoshop-helps-nasa-reveal-the-unseeable.html. It appears their goal is to accurately portray what the subject would look like if you could see it and not just what the camera (telescope) captured.