star photos, what camera to buy?

I am looking cautiously at mirror less full frame cameras. I entertain myself by taking photos at night, sunsets, stars, people in motion with 1 second exposures. Before I spend more than I can afford, I want to do a little homework, and get the right camera.

 

I will be taking night photos of stars. Sunset photos in remote national parks, and animals in Yellowstone park.

I have been looking at Sony mirror-less cameras. As a non expert, I could use some advice.

 

Can a mirror less camera see further than an DSLR with the same lens? I read somewhere that a mirror less with a 200 mm lens could see further than a DSLR camera using the same 200 mm lens. Any one know if this is true?

I read somewhere that a mirror less with a 200 mm lens could see further than a DSLR camera using the same 200 mm lens.

 

I can only imagine they meant a less-than-full-frame mirrorless, and that's what would make the difference, not whether the camera has a mirror or not. If the camera is less than full-frame, it records a smaller piece of the lens's image, and that makes the 200 a more powerful telephoto for that camera than it is for a full-frame camera. However, the lens is behaving exactly the same in each case; all that's different is that the smaller-framed camera doesn't record the edges of the view.

All lenses can see to infinity, or we would have no star photos! ;)

"Can a mirror less camera see further than an DSLR with the same lens?"

 

- Absolutely not. Complete BS.

 

"I read somewhere that a mirror less with a 200 mm lens could see further than a DSLR camera using the same 200 mm lens. Any one know if this is true?"

 

- Don't read anything else from that source. It's rubbish.

Edited July 16, 2018 by William Michael

Photos of stars don't always work on digital cameras Some cameras are really terrible. But some are actually pretty darned good. From memory, the Fuji cameras are really good for astrophotography. You still need to eliminate chromatic aberration in the RAW converter, though.

Can a mirror less camera see further than an DSLR with the same lens? I read somewhere that a mirror less with a 200 mm lens could see further than a DSLR camera using the same 200 mm lens. Any one know if this is true?

You must have misread something. - It is frequently uttered that "a smaller sensored camera provides more reach with the same lens". But you are planning to get a FF Sony? - If you aren't shooting Phase One or Betterlight (<- MF / LF stuff) right now, you won't benefit from a smaller sensor.

 

"More reach" doesn't necessarily translate to "seeing further" In an ideal world I could rig up my 13 x18cm with the 355mm portrait lens take a shot and another with the film holder replaced with a smaller camera. - assuming both cameras held film in focus and manufacturers used the same emulsion for 13x18 & 35mm film I would get identical 35mm negatives after processing the same way and cutting down the 13x18cm sheet. All the fog between me and the subject would be rendered the same.

 

With digital you can sometimes get small higher pixel density sensors. Assuming your "same" lens was incredibly good i.e. "ideal", this would mean that a 20MP MFT camera could indeed pull more details out of a 200mm shot than a FF body with lower pixel density.

 

If you compare FF to FF a DSLR might "see further" than mirrorless right now, since Canon pack more megapixels on their EOS 5 D SR's sensor than Sony into their A7R III's. - Warning: I am neither suggesting to buy that Canon nor even sure if the sentence before will be visible truth IRL; it is just theory.

 

IDK if Sony mirrorless will be right for you. - I never used one. - Either you'll like their UI or you won't. - Sensors are decent, AF is sometimes excellent.

If I wanted to buy something for "animals in Yellowstone", I'd ponder a Nikon D500 with 200-500mm ($3k5ish?). AFAIK it is the(!) budged birding camera right now?

I am not sure about everything else on your list.

I entertain myself by taking photos at night, sunsets, stars, people in motion with 1 second exposures.

If I wanted to do such, I'd grab my Leica Ms & a tripod.

For stars and landscapes you should maybe take a look at Pentax. - They make DSLRs with admittedly sluggish AF but IBIS and a special mode to use the IBIS mechanism to compensate the earth's movement in astro photography.

 

As a bottom line: I understand your urge to shoot people moving through your frame with a non-SLR. I do not know if you can watch them during your 1sec exposure in live view through your EVF. I hope some Sony shooter like @Ed_Ingold will jump in to clarify on this.

 

Getting into a camera system with IBIS and OIS lenses for fun at night is a smart move. It might permit leaving the darn tripod at home once in a while. - I can confirm: Cranking the ISO up to maximum doesn't always cut the cake for handholding unstabilized cameras.

 

Make up your mind what you'll be willing to carry(!) and able to spend.

To me the "FF with zooms stuff" seems bulky & borderline. I have a Canon with 70-200/2.8 and think twice before I'll pack it; it doesn't feel like a casual hiking camera to me. But I don't mind tossing a Leica kit, including 90 & 135/4, into my backpack to just look & see. A Sony body would be lighter than the Canon, but mirrorless brings no zoom bulk advantage and wide primes seem a wash on DSLR & Sony too?

 

It is not my intention to write a Leica add here. I only mentioned that kit to say: Some subjects don't demand much of a camera and while chasing them you might appreciate not carrying what you don't need... Manual focus seems fine to me, for pictures of the sky.

 

Do you need high resolution or insane ISO settings for anything? - If not: Sit down, make an Excel sheet with the weights of your complete dream kit (after lottery wins & bank robberies...) in various systems. Include MFT, & Fuji, a 4th tab for a mix with the birding Nikon and come to your conclusion.

 

Whatever you'll end buying will be a compromise. There is no perfect camera for everything and when you'll start mixing systems you'll loose the convenience of just grabbing the photo bag some day.

"Reach" or magnification is not desirable for starry landscapes, rather light gathering power and minimum thermal noise. A wide or super-wide lens will produce a better effect. Planets and deep space objects need higher magnification and tracking. Most, however, are surprisingly large. The Andromeda galaxy subtends an angle greater than the full moon, which fills a full frame at about 16 power, and atmospheric disturbances affect the resolution. In general, the larger the sensor cell size the better the light gathering power and the lower the noise. In lieu of a tracking mount, the longest exposure for acceptably short start trails is roughly 300/F seconds for an FX camera. A 50 mm (equiv) lens may barely cover some of the larger constellations.

 

Some cameras are better than others, and some very good cameras like a Sony A7Rii have "star-eating" noise reduction for exposures longer than 3 seconds. The newer Sony A9 and A7Riii are much better in this regard. I shot this photo last night at a resort deep in the Cascade Mountains of Washington.

 

Sony A7Riii + Loxia 25/2.4, f/2.4, 10 sec at ISO 800

Magnificent picture Ed! I envy you such great 'seeing' as you obviously have in the area. I'd have to travel at least 100 miles to get such dark skies.

 

Tell me, are the apparently multi-coloured stars an artefact of the Bayer matrix in the camera, or really that colour? If the colour is artificial, would it be more natural looking to selectively desaturate the image?

 

The few times I've attempted star pictures, most of the stars I captured were rendered white. I've always wondered why a supposed point source should affect all 4, or at least 3, of the RGGB pixel cluster, and not just a single photosite.

I gotta make ma a barn door......

Sattelite upper left?

The star colors appear to be natural - hot blue, yellow and cool red. Some of the blue is probably noise, blue after correcting the normal dirty brown of the night sky. There's a hint of hydrogen alpha, but the orange at the bottom is scatter from lights in the condo, just outside the field of view. I think the line is a satellite, because an airplane would show pulses, which I don't detect in the master. A good color/resolution test would be Albeiro, an orange-blue optical double in Cygnus.

 

I think Sony has turned down their star-killer noise reduction, because I've never seen this kind of star density with my A7Rii, even on a dark night.

 

Stars brighter than a certain threshold will spread by diffraction, residual aberrations in the lens and focus issues. The Loxia has significant coma, and cat-eyes off center. I have a Batis 18/2.8 which is much cleaner. The example I uploaded is downsampled in LightRoom, which probably affects the geometry, making star images larger and rounder.

 

The "best" camera for deep space objects, aside from professional CCD's, may be a Nikon D810 modified to pass hydrogen alpha light. However, I think I can have fun with the Sony A87Riii, once I get far enough from Chicago lights.

Why do stars appear white in photos? On reflection, I think this phenomena must be due to over exposure. Stars are point sources, and the nearest are actually very bright. Deep space objects need longer exposures, which will naturally overexpose stars. Techniques of stacking and masking can largely overcome this disparity for graphic purposes. There are others with far more experience than me in these matters.

 

The photo above was exposed for 10 seconds at ISO 800. I did this primarily to avoid star trails, but star colors are clearly retained. The milky way was barely visible in the RAW image, but responded nicely to LR "Auto" toning, also to tone mapping in AuroraHD (shown). I also tried a program called "StarryLandscapeStacker", which does an excellent job of cancelling thermal noise (I stacked 5 exposures), and can selectively retain detail in the foreground. The ultimate exposure was boosted by only one stop. There are many good ways to skin this cat. One technique familiar to photographers, unsharp masking, was successfully used (developed?) by an Australian astronomer years ago, to show detail in planetary nebulae.

 

I have read that the physical diameter of the objective is a better measure of light gathering than the relative aperture, or f/stop. You can see far more through an 8" reflector telescope at f/8 or smaller than through a 50/1.4 lens. Light from deep space objects is often measured in photons, not f/stops.

Edited July 24, 2018 by Ed_Ingold