I ran the A7R V for three nights on a ranch outside Fort Davis, Texas, about forty minutes from the McDonald Observatory, specifically to test whether a 61-megapixel body was worth the extra hassle for star stacking. Short answer: yes, but not for the reasons I expected going in. The resolution doesn't make the stars themselves sharper. It makes every small mistake in your capture technique more visible, and it changes the math on how long you can leave the shutter open before star trails show up in a 100% crop. What follows is the workflow I actually settled on, not the settings sheet I started the trip with.
What 61 Megapixels Actually Changes for Star Stacking
The A7R V's pixel pitch is roughly 3.76 microns, against about 5.9 microns on the 24MP A7 III and around 8.4 microns on the A7S III. Smaller photosites gather less light per pixel and show read noise sooner as you push ISO, which sounds like a problem for a low-light genre. In practice it matters less than people assume, because stacking does the noise reduction for you: you're averaging dozens of frames instead of leaning on one high-ISO exposure, so the per-pixel noise disadvantage gets diluted out across the stack.
What the extra resolution does change is how forgiving your exposure time can be. A star that still reads as a clean point at 100% on a 24MP file will show visible elongation on the A7R V after a noticeably shorter shutter speed, purely because you're looking at more magnification pixel for pixel. The old "rule of 500" (500 divided by focal length, giving you a max shutter speed in seconds) was calibrated for lower-resolution sensors and web-sized output. On this body, at full resolution, I treat it more like a rule of 250 to 300 if I actually care about pinpoint stars in a print. If you're going to downsample the final file anyway, the old rule is fine and you can ignore the rest of this paragraph.
Locking Down Camera Settings Before You Start the Sequence
Everything here gets set once, before the first frame, and then left alone for the rest of the session:
- Mode and white balance: full manual exposure, and a fixed Kelvin value (I use 3800K) instead of auto white balance. Auto WB will drift slightly frame to frame as clouds pass or the moon rises, and that shift becomes visible as color banding once you stack.
- ISO: ISO 3200 when I'm tracking, ISO 6400 when I'm not and need to keep shutter speed down. The A7R V holds up better at 6400 than its resolution would suggest, but I still prefer more subs at a lower ISO over fewer subs pushed hard.
- Raw format: lossless compressed. Uncompressed raw at 61MP eats card space and buffer depth for a quality difference I can't see in a stacked file, and it slows down how many frames you can queue during an interval sequence.
- Long exposure noise reduction: off. Sony bodies going back to the original A7S had a reputation for a noise reduction algorithm that quietly suppressed faint stars in long raw exposures, the "star eater" problem. The A7R V is much better behaved than those older bodies, but I still test it on location: shoot one 15-second sub with LENR on and one with it off, zoom into a dim patch of sky on the rear screen, and compare. Leaving it off also means the camera isn't burning an equal-length dark frame after every single sub, which matters when you're racing a cloud bank.
- Focus: manual focus only. Press the magnifier button, zoom onto a bright star or a visible planet, turn the ring until the point is as small as it'll get, then lock it down with a strip of gaffer tape across the focus ring and barrel so it can't creep as the lens cools.
- Drive mode: the camera's built-in interval shooting function, set for 80 to 100 frames with the interval set about a second longer than the exposure itself, so the buffer has a moment to clear between subs.
- Shutter type: mechanical, not fully silent electronic. I've seen faint banding show up in shadow areas on high-ISO electronic-shutter subs that wasn't there on mechanical-shutter frames from the same night, and it's not worth discovering after you've already stacked forty of them.
Card choice matters more than people expect at this resolution. A compressed raw file off the A7R V runs somewhere around 60 to 70MB, and an uncompressed one can hit 120MB. On a short interval, a slower SD card will start to lag behind the buffer partway through a long sequence. If you own a CFexpress Type A card, this is the session to use it. Also carry more batteries than you think you need. The NP-FZ100 is a solid battery, but it drains faster than the daytime rating suggests once temperatures drop into the 40s Fahrenheit or lower, and a dead battery mid-sequence means restarting the whole stack.
Practical Sub-Exposure Limits by Focal Length
These are the shutter speeds where I start to see visible star elongation at 100% on the A7R V's 61MP files, shooting untracked from a static tripod. Your mileage will vary slightly by declination (stars near the celestial equator trail faster than stars near the poles), but this is close enough to plan a session around.
| Focal length | Typical aperture | Max untracked sub-exposure | Notes |
|---|---|---|---|
| 14mm | f/1.8 | ~8 seconds | Widest margin, best for foreground-heavy nightscapes |
| 20mm | f/1.8 | ~6 seconds | Good middle ground for Milky Way core framing |
| 24mm | f/2.8 | ~4-5 seconds | Elongation becomes visible fast at 100% crop |
| 35mm | f/2.8 | ~2-3 seconds | Really needs a tracker at this resolution |
Tracked vs Untracked: Which Setup Actually Fits This Body
A star tracker changes the whole equation because it moves the mount to counter Earth's rotation, so instead of being limited by trailing you're only limited by how well you polar aligned. I use a Skywatcher Star Adventurer GTi, rated for roughly 5kg of payload. The A7R V body is about 723 grams, and even a heavier lens like the Sigma 14mm F1.4 DG DN at around 1,170 grams leaves plenty of headroom. With the Sony FE 14mm F1.8 GM instead, at about 460 grams, the whole rig is light enough that I don't worry about balance at all.
Tracked, I can run subs at 60 to 120 seconds instead of single-digit seconds, which means far fewer frames for the same total integration time, and a much lower chance that a passing satellite or aircraft ruins a meaningful fraction of your stack. The tradeoff is setup time. Polar alignment through a tracker's scope or app takes ten to fifteen minutes I don't always have, especially if I'm still scouting a composition. My actual habit: untracked for wide nightscapes where a foreground element is doing a lot of the compositional work, tracked for tighter shots of the Milky Way core where the sky itself is the subject and I'm planning to blend in a separately exposed foreground anyway.
Stacking Software, Compared for This Specific Workflow
| Software | Platform | Cost | Where it fits |
|---|---|---|---|
| Sequator | Windows only | Free | Fastest path from untracked nightscape subs to a clean sky, has decent automatic foreground masking |
| Starry Landscape Stacker | Mac only | Paid, one-time | Better manual foreground masking control than Sequator, worth it for tricky treelines or ridgelines |
| DeepSkyStacker | Windows only | Free | Built for tracked deep-sky stacking with dark/flat/bias frame support, overkill for a simple nightscape |
| PixInsight | Windows, Mac, Linux | Paid subscription | Steep learning curve, but the only one here with serious gradient removal and calibration tools for tracked sessions |
Building the Stack: From Capture to a Culled Sequence
A typical untracked session for me looks like 60 subs at 8 seconds, which is about 8 minutes of total light gathered across roughly 15 minutes of real time once you account for buffer clearing between frames. A tracked session might be 40 subs at 90 seconds, closer to an hour on the mount for one composition. Either way, once I've tried two or three framings and a couple of exposure tests across a night, I'm coming home with 300 to 500 raw files, each 60 to 120MB, and most of them look nearly identical to each other at a glance.
That's where the culling step actually saves time rather than just organizing files for the sake of it. I run the folder through imagic before touching any stacking software. Its burst and duplicate clustering groups the near-identical consecutive subs from each sequence automatically, so I'm not scrolling frame by frame trying to work out where one composition ends and the next test begins. Its local sharpness and focus scoring is more useful than I expected for this genre too: on the second night, dew crept onto the front element about twenty minutes into a sequence without me noticing from the rear screen, and the scoring flagged the soft back third of that stack before I fed all of it into DeepSkyStacker and wasted twenty minutes processing frames I'd have to throw out anyway. All of it runs locally, which matters more at a dark sky site than it does at home. There's no cell signal forty minutes outside Fort Davis, and I'm not waiting on an upload to review a night's shooting before I break down the tripod.
Once the sequence is culled, I run the light frames through the stacker of choice from the table above, and separately expose and process a single well-focused foreground frame if I'm shooting a nightscape rather than a tight sky-only crop. I blend the two manually, sky stack for noise and detail, foreground frame for a sharp, clean base layer, rather than trying to force a stacking tool's automatic foreground masking to handle a complicated treeline.
Field Notes and Mistakes Worth Avoiding
Dew is the single biggest killer of an otherwise clean sequence, and it happens faster in humid conditions than most people expect, even in a place as dry as West Texas can be by day. A rechargeable USB dew heater strap around the lens barrel, run off a small power bank, solved this for me after losing half a night's frames to a fogged front element early on. In a pinch, a chemical hand warmer taped loosely around the lens barrel with a rubber band works almost as well for a couple of hours.
Don't panic over a stray colored pixel in a single frame review on the rear screen. Hot pixels show up occasionally at high ISO on long exposures and can look like a star that shouldn't be there. Most stacking software (Sequator and DeepSkyStacker both do this by default) uses a sigma-clipping or similar rejection method that throws out pixel values that don't agree across the stack, so a hot pixel present in one frame out of sixty just vanishes in the final result. It's only a problem if you're judging quality off a single sub instead of the finished stack.
Once I'm back with a signal and going through the trip's full catalog, including the daytime scouting shots and the wider portfolio images from the same location, I lean on imagic's apply_my_style preset to keep color grading consistent across everything that isn't a stacked night sky, since the astro files get their own separate treatment in a dedicated stacking tool anyway. For more on culling large sessions in general, see how AI photo culling works, and if you're deciding whether a one-time-purchase tool fits your workflow, the pricing page lays out what's included.
Frequently Asked Questions
Does the A7R V's pixel density actually hurt image quality for astro work?
Not in a stacked result. Smaller photosites do show more read noise at a given ISO than a lower-resolution sensor would, but stacking multiple subs averages that noise out across the frame count, so the disadvantage mostly disappears by the time you have a finished image. Where the pixel density does bite is exposure time: you'll hit visible star trailing sooner than the old exposure-time rules of thumb suggest, simply because you're viewing more magnification at 100%.
How many subs do I actually need for a clean Milky Way stack?
For an untracked nightscape, I get usable results starting around 30 to 40 subs, with real improvement up to about 60. Past that point the gains flatten out and you're mostly burning battery and card space. Tracked sessions need fewer frames for the same noise reduction because each sub is already gathering far more light, so 20 to 30 subs at 60 to 120 seconds each is usually plenty.
Is Sony's long exposure noise reduction safe to leave on for star stacking?
I leave it off. The A7R V doesn't show the aggressive star suppression that older Sony bodies were known for, but leaving LENR on means the camera shoots a matching dark frame after every light frame, doubling your effective exposure time and cutting your usable shooting window roughly in half on a night where clouds might roll in. It's easier to shoot your own dark frames separately at the end of the session if you want them.
Can I get away with a lighter tripod instead of a full star tracker?
Yes, for untracked stacking a sturdy travel tripod is enough, since you're not trying to counter Earth's rotation, just hold the camera still for a few seconds per frame. A tracker only becomes necessary once you want single exposures longer than the limits in the table above, or when you're shooting tighter focal lengths where trailing shows up within a couple of seconds.