There was a time when a camera lens was simply a set of shaped pieces of glass in a tube, with a mount and an aperture. But as cameras have embraced electronics ever more, technology has found its way past the lens mount to the extent that all features of a modern lens are electronically controllable. Can they be used outside the confines of the camera they were designed for? If the user is [Jana Marie] then certainly, because she’s created a nifty USB adapter and mount for Canon lenses for use with her custom streaming camera.
The hardware is a 3D printed lens mount with a PCB that mates with the pins on the lens. An STM32 does the hard work and talks to the outside world through a USB interface, however it’s in the software that the real effort lies. The Canon lens protocol has been extended since the 1980s, and the commands for different generations of lenses can be convoluted. All the information is in a GitHub repository, so the curious hacker can roll their own.
There are a wealth of camera projects to be found for those that don’t mind tearing apart some of their more valuable possessions, and this isn’t the first we’ve seen involving the hacking of the Canon protocol.
I wish this was for Sony E-mount lenses, not for Canon. Or for both Sony, Canon and Nikon, so we could make an open source cross-platform lens adapter…
Sony has never been hacker friendly. They are in my no-buy list for that.
if you buy a BRAVIA TV there is a long list of all the open source tools used in the firmware
my 37″ runs busybox
asking sony “tech support” about anything resulted in a stern letter from the “legal department”
I guess “open source” is relative
They all use proprietary protocols, third party manufacturers like Sigma and Tamron had to reverse-engineer protocols to make their lenses for Canon, Nikon, Olympus, Sony, …
Sony E mount is an open standard by the way. That’s why there are so many different 3rd party lenses for it
No, it’s not. There is an open-source implementation, but protocol had to be reverse-engineered, and. The effort is incomplete.
Sony also uses their own proprietary USB connector for remote control, that hides additional pins. It has some serial communication included, but that is also proprietary, and hidden behind NDAs and licenses…
I highly doubt the last part, that will be more like a licensing and NDA thing.
Not sure about recent changes but Canon didn’t want to license EF protocol to anyone. Earlier Sigma lenses had a lots of problems because Sigma reverse engineered protocol incompletely and some lenses had to be re-chipped when Canon released newer bodies that used commands Sigma still didn’t implement. They did better job later, and also relased USB docking for Sigma lenses that can be used to update firmware in lenses.
Neat, I happen to be a canon camera guy (For the simple reason Dad was so when I was getting started I used his, then borrowed his lenses) so for me this could actually be quite useful now. Having a few duplicate ‘kit’ lens and probably some other spares now that should now be easier to use for other purposes and far better than the cheap alternatives.
Very nice and get’s me wondering if worthwhile to marry with the auto-focus of a webcam also. I last left off looking at the C920 project using Saulius Lukse’s adapter and a quick look at his site looks like there has been some significant advancements… granted not open source as much.
Looks like an really awesome project and alone with the Canon lenses is handy. I envision a blog or site somewhere to at least side by side compare the mechanics and electronics in a table maybe… with the protocol info being super sweet?
Furthermore, the last word “protocol” has me thinking about the latest and greatest capabilities of open source protocol decoders/analyzers where I guess sigrok is the best? Anyone aware of a better open source way to decode the protocol than step by step procedure and correlation with the signals?
where is Nikon F mount and Z mount. Will it possible?