Why the Leica M11's Sensor Changes Your Star Stacking Workflow

I didn't buy a Leica M11 to shoot the Milky Way. I bought it for street work and the odd portrait session, and for the first year it barely left the city at all. Then a friend dragged me out to a dark sky patch in the Pyrenees for a long weekend, and out of stubbornness (and because I hadn't packed anything else) I decided to see what the M11 could actually do pointed straight up. It turned out to be a strange, occasionally frustrating, occasionally excellent tool for star stacking, and enough people have asked me about it since that I think it's worth writing down properly instead of repeating myself in comment sections.

This isn't a camera anyone would design specifically for astrophotography. No autofocus, no in body stabilization, no built in intervalometer, and a rangefinder patch that becomes close to useless once the light drops below what a streetlamp gives you. And yet the sensor underneath all that manual everything body turns out to be genuinely good for stacking work, which is the part nobody warns you about when they're busy telling you to buy a dedicated astro camera instead.

The Sensor You're Actually Working With

The M11's 60 megapixel back illuminated sensor is the reason this whole exercise works at all. Leica's triple resolution DNG option lets you record the same shot as a full 60MP file, a 36MP file, or an 18MP file, and for star stacking that choice matters more than most people expect. At native ISO 64 the sensor behaves like a landscape sensor: low noise, huge dynamic range, not really built for high ISO work. But push it to ISO 3200 or 6400, which is where most of my star sequences end up, and it holds together better than I expected from a camera with no stacked sensor tricks or dedicated cooling. It's not a Sony a7S III. It doesn't need to be, because stacking dozens of frames is doing the noise reduction work that a single higher ISO body would otherwise have to do alone.

Where the resolution choice actually bites is file size and processing time. A night of shooting at full 60MP, say 150 to 300 frames for a single composite, produces a genuinely enormous folder, and stacking software chews through 60MP DNGs a lot slower than 18MP ones. I've settled on shooting wide establishing shots at 36MP and dropping to 18MP for anything I already know is purely a stacking base layer with no intention of ever printing it large. Full 60MP stays reserved for single frame nightscapes where I want every bit of foreground detail.

Focusing at Night Without Autofocus

This is the part that trips people up first. The rangefinder patch on the M11 is fantastic in daylight and passable at dusk, but by the time stars are actually visible it's fighting a losing battle: there's no contrast for the patch to lock onto, and even a bright star like Vega or Sirius barely registers as a usable double image target. I tried it the "purist" way on my first two nights out and wasted probably forty minutes of dark adapted eyesight fumbling with a focus patch I could barely see.

The fix is the Visoflex 2 electronic viewfinder, and I'd call it close to mandatory for this use case rather than optional. Switch to live view, zoom in digitally on a bright star, and manually rack focus until it collapses to the smallest possible point. It's slower than autofocus on any modern mirrorless body, no argument there, but it's accurate, and once you find infinity you don't need to touch it again for the rest of the night unless the temperature swings hard enough to shift the lens barrel, which does happen on M mount glass more than people admit.

Trusting the Infinity Stop

Most Leica M lenses have a hard mechanical stop at infinity, which sounds like it should solve the whole problem for you. In practice I've found it's close but not always exact, especially on older lenses or ones that have been dropped or serviced. Rack out to the hard stop, take a test frame at your working aperture, zoom into the EVF playback at 100 percent, and nudge back a hair if the stars look soft or slightly doubled. Do this once per lens per session and mark the ring's exact position with a strip of gaffer tape so a stray knee bump in the dark doesn't cost you the shot.

How Many Frames, and How Long Each One

Because the M11's pixels are packed tighter than on a 24 megapixel full frame body, star trailing shows up sooner than the old "500 rule" would suggest. I stopped trusting that rule on this camera after my first batch of test shots came back with visibly elongated stars at settings that would have been fine on my previous 24MP body. What I use instead is a more conservative set of limits scaled to whichever resolution mode I'm shooting in, worked out over a few nights of test frames rather than a formula I trust blindly.

These numbers come from my own field tests, not a formula copied off a spec sheet, so treat them as a starting point rather than gospel. Your own tolerance for what counts as "trailing" at 100 percent crop might be tighter or looser than mine. What the table does show clearly is the practical tradeoff: shoot the full 60MP file and you're locked into shorter exposures, which means more frames, longer stacking sessions, and a bigger burden on whatever you're using to sort the keepers from the misses afterward.

Picking Glass From the M-Mount Line

Fast primes are where the M mount catalog genuinely earns its keep for this kind of work. The Noctilux-M 50mm f/0.95 is absurd for star fields wide open, coma smears the corners into little seagull shapes, and it's borderline unusable above 60MP where every flaw gets resolved in painful detail. Stop it down to f/1.4 or f/2 and it settles into something usable, though at that point you've given up most of the reason to own a 0.95 lens in the first place.

The Voigtlander options, a cheaper way into wide fast M mount glass than buying Leica branded lenses for a genre where you're likely to leave the lens out in dew and cold overnight, have been my actual workhorses. The 21mm f/1.4 Ultron and the 15mm Super Wide-Heliar cover most of what I need for Milky Way arches and wider star fields, and both are sharp enough in the corners by f/2 that coma stops being the limiting factor. If I were buying into this system purely for astro work and nothing else, I'd start there rather than chasing the Leica branded glass.

Sorting Hundreds of Frames the Morning After

Nobody talks about this part enough. A single stacking session at 4 to 10 seconds per frame, run for 40 minutes to build up a composite, leaves you with 250 to 500 raw files, and a meaningful chunk of them are unusable: a plane crossed the frame, condensation crept onto the front element halfway through, the tripod got bumped, or focus drifted slightly as the temperature dropped. Culling that pile by eye, frame by frame, at one in the morning after a four hour shoot is exactly the kind of task I no longer do by hand.

I run the night's DNGs through imagic before they go anywhere near stacking software. It scores focus and sharpness locally on the machine, no upload involved, which matters when a single session's RAW folder can run into tens of gigabytes, and it clusters near duplicate frames from the same star field into groups so I'm not scrolling through 400 nearly identical exposures one at a time. If you haven't seen how that kind of automated culling actually works under the hood, here's a breakdown of the sharpness scoring approach. For a stacking workflow specifically, that scoring is what earns its keep: it flags the handful of frames where focus drifted or a gust nudged the tripod, so they never make it into the stack and drag the final composite soft.

From Stack to Finished Frame

Once the culled set is ready, the actual stacking happens with dedicated software, not imagic itself, and I'm not pretending otherwise. Sequator and Starry Landscape Stacker both handle the M11's DNGs without complaint once they're run through a DNG converter or exported as TIFFs first (Sequator in particular can be fussy about reading Leica's DNG metadata directly). DeepSkyStacker works too, but I've had more registration failures with it on wide field Milky Way shots than with the other two, possibly because of how much foreground ends up in frame compared to a pure deep sky target.

The editing pass afterward is where I've found the most personal value beyond culling. I trained an apply_my_style preset on a batch of nightscapes I'd already graded by hand, the shadow lift I like, how far I pull down highlights on the brighter stars, the specific color temperature I land on for Milky Way cores versus foreground rock. Running new stacked composites through that preset gets me most of the way to a finished look before I do any manual fine tuning, which matters more than it sounds like when you've got a backlog of five or six sessions' worth of stacks waiting to be edited. It's a one time purchase rather than another subscription to track, which is worth knowing before you commit to a workflow like this; the pricing page lays out the tiers if you want the specifics.

Cold Nights and Battery Reality

The M11's battery is rated generously for daytime shooting and rated nowhere near that once you're doing four to ten hours of exposures in freezing temperatures with the EVF running continuously for focus checks. I get through one battery in roughly 90 minutes to two hours of active stacking work in genuinely cold conditions, below freezing, which means carrying at least three spares for a full night session, more if you're also reviewing frames on the rear screen between sequences to check focus. USB-C charging is convenient for topping up in the car between locations, but it's slow enough that it won't rescue a dead session on its own.

Weather sealing is another area where I'd manage expectations. Leica calls the M11 splash resistant, and I believe that for light drizzle, but heavy dew on a clear cold night is a different, slower kind of moisture exposure that sealing gaskets don't really account for. I keep a small towel and a lens hood on every session specifically to buy time against dew forming on the front element, because a fogged lens thirty minutes into a stack means starting the whole sequence over.

Frequently Asked Questions

Do I need the Visoflex EVF to shoot star fields with the M11?

Not strictly, but I'd consider it close to essential rather than a nice to have. Focusing on stars through the optical rangefinder alone is genuinely difficult once ambient light drops, and the Visoflex's digital zoom during live view is the most reliable way I've found to confirm sharp focus on a point light source in the dark.

Which resolution mode should I actually shoot for a stacking session?

It depends on your exposure budget more than your final print size. Full 60MP forces shorter individual exposures before trailing appears, which means more frames and longer stacking sessions on the software side. I shoot 18MP for pure stacking bases and reserve 60MP for single nightscape frames where foreground detail matters more than exposure length.

Is the lack of in-body stabilization actually a problem for this kind of shooting?

Less than you'd think, since exposures short enough to avoid star trailing are also short enough that camera shake off a stable tripod isn't the limiting factor. IBIS would help more for handheld twilight shots before the stars come out than for the stacking sequence itself.

Can imagic replace dedicated stacking software like Sequator or DeepSkyStacker?

No, and it isn't trying to. imagic handles the culling stage, scoring sharpness and grouping duplicate frames locally before you hand a clean folder over to dedicated stacking software for the alignment and stacking math, which is a different job entirely.