Show more

The adapter does indeed need to be inset by 0.8 mm. With one final print last night I was able to confirm that the registration is correct!

I've no idea what was happening during these frames, but they make for some interesting glitch art.

Just for fun, a quick recording of state of the art live view, circa 2008.

After a couple of near misses I think I might have the correct registration offset. The current math says I need to offset the adapter _inside_ the body of the camera.

Emory Dunn boosted

Pitch: self-driving cars but they're on rails and carry hundreds of people and have an operator and they're called "trains" instead.

I really wish I had a faster prototyping method than 3D printing.

It appears I might have a bit of light leak. Almost as if translucent white plastic isn't the ideal photographic tool.

Using my professional-grade Sensor Registration Marking System (painters tape) I have a new offset of -3.5 mm.
Now to wait 3 hours for a new print. mastodon.emory.coffee/media/RM

Emory Dunn boosted

Below is an example. It's a throw-away photo, but the rendering is what caught me. The focus is off and the subject is soft. But I loved the look - it has enough detail to get the point across, but the unimportant details aren't able to be rendered.

Now all of the sudden I'm smitten with the idea of playing with a '40's or '50's era lens.

I'm going to have to suffer through the "live" view on the back to figure out whether the adapter is too thick or thin.

The 65 mm lens arrived so I was able to properly test the adapter. It's safe to say I do not have the sensor registered to the ground glass correctly 🙁

I think I've settled on the design for the real adapter. The goal is for it to be easy to make, ideally in a single CNC run on the back side.

The next step is finding a new lens. The 85mm that came with it doesn’t have PC sync and my 150mm is too long.

Emory Dunn boosted

Programming is great because you can just take that huge messy chuck from the middle of your function and hide it away under a new name in a new function and feel good about how you "cleaned up" the original function by abstracting the internals.

This is exactly how I used to clean my room as a kid - I'd refactor all the junk on the floor to be under my bed. Boom - problem solved.

Success of the day: replaced some old janky code with some much simpler Hazel rules!
Now the whole system won’t be a Blackbox to anyone but me.

I'm not sure how much of a professional and DT community is here yet, but you never know.
There's a project I've been working on for a while, called CaptureBot, that I'm looking for beta testers for.
Check it out: emory.coffee/applications/capt

I've revied my big 3D printer to begin the process of turning this (fig. 1) into this (fig. 2) over the next four hours.

Show more
Mastodon

emory.coffee is one server in the network