SNIPPED
Not quite. You can do a "clean room" copy of code, but not for a Patent. If Canon has patented the protocol rather than simply copyrighted the code, then it does not matter whether or not they used a "clean room." I guess that Canon would have a mix of code and patents.
Yes, they have patents, but I haven't seen any that actual cover the communication of the RF protocol itself. The reason this is a trade secret rather than patented is likely because (1) Canon does not want to disclose the exact operation of the protocol and (2) it is probably not patentable. There are some patents related to specific aspects of the communication, such as the detection of a new-protocol lens via communication content (on the mount side) and some aspects of IS coordination. There are also some patents (Canon appears to reference design patents) covering physical construction (though Canon has not had a problem with any MF lens, and the basic fitting into the mount is not covered by any Canon patent claim I've seen). No patent covers the basics of the RF protocol commands (on purpose, because they don't want those public) and there's surely a way to accept the basic commands for changing focus distance without violating any Canon patent I have seen.
A good point is that you can't enforce patent-type IP until the patent has been issued with the allowed claims. So if they are using patents rather than copyright, the patents must exist.
A patent on the communication protocol would not necessarily have to give the specific implementation away. A good patent would be more generic and broadly applied, but this is a special case as Canon is probably not worried about Sony or Nikon copying their protocol.
Design patents are extremely weak and only go to the aesthetic value and are easy to get around unless you are trying to make it look like it is another company's product.
I have NOT looked at Canon's patents. But there could be something in how the lens locks onto the mount.
The other area is in the optics that might be specific to the optical design for an RF camera. Granted, if it is just the optical physics of adapting, say Sony mount to the RF mount with a slightly different flange distance, it might not hold up in court even if Canon has a patent. But would the 3rd party risk it?
My main point is that Canon has many possible layers of defense if they want to make it unpalatable to 3rd parties and not all the defenses may be obvious yet.
Patents in the US and most other places are specifically excluded from anti-trust. In effect, the whole purpose of a patent is to give a monopoly for a set time period.
Well, they're somewhat separate areas, but yes, patents are a monopoly on a specific invention. US anti-trust law has different targets than inventions generally, though.
Regardless, it would be highly risky for anyone to develop an unauthorized 3rd party lens unless they stick strictly to the EF protocols.
Maybe, depending on exactly how RF is implemented. On the other hand, from reading the EF docs I have seen, there is no reason you could not make a lens that is internally EF but uses the RF flange distance. I thought that the Samyang and Viltrox lenses might have been doing this internally, but I have no evidence of that being quite the case.
The EF mount was introduced in 1987 with very primitive technology. I would expect that anyone doing an "unfriendly" RF mount would have to hold closely to EF protocols.
Even then, Canon might be able to figure out how to make non-Canon lenses. We don't know if there is not some hidden code that will spring or be enabled by future firmware or camera models. Apple sprung this type of code on 3rd parties that made unauthorized Lightning cables, and they stopped working.
They definitely
could do this technologically. They never did on DSLRs or mirrorless cameras before, though. So it seems unlikely that they would begin now. All the lenses that use the EF protocol accurately (and the issues with a few early Sigma lenses have been documented by the modchip guys; they weren't blocked, but they got a few details wrong) have continued to work.
We are seeing reports that the Sigma EF150-600 is having erratic focusing issues with servo AF. Specifically, Duade Paton has reported the problem (
) and got the following response from Sigma:
"Sigma have responded that the Sigma 150-600C is not compatible with the Eye Tracking Servo mode of the Canon mirrorless bodies. You will need to use single shot AF to get accurate autofocus. The AF will still work in tracking mode but you will get inconsistent focus as I have demonstrated."
Whether by design or omission, there will be problems with at least some 3rd Party EF mount lenses even is "properly adapted." Canon will probably feel obligated to develop firmware to fix Canon EF lenses, but no obligation for 3rd parties.
We also don't know if there was some "future-proofing" in the RF interface that may only be deployed on new bodies that may cause problems for 3rd party lenses.
Once again, even if Canon has diabolically put specific code in that causes Sigma EF lenses not to work (say the AF to work unreliably) with their RF cameras, it is probably not an Anti-Trust matter. It may or may not be a good business decision.
Though they have not put in such code, yes, if they did it, I don't think there would be any anti-trust lawsuit. It would be similar to what they have already done with batteries. I don't think it would make business sense for lenses. It isn't like they didn't know about third-party EF lenses in the age of upgradeable firmware either. They even mentioned them on the webpage I linked to earlier.
We are over 3 decades of technological advancement from the EF mount. Canon was much more hamstrung even with their newer EF lenses that supported firmware based on their legacy lenses and bodies. As I mentioned before, there may be features built into their roadmap that could screw up 3rd party lenses (maybe totally by accident, as they will verify Canon lenses.
Even though I have no plan right now to buy a 3rd party lens. I think it would be good for the RF ecosystem to have a 3rd party licensing program. I would prefer it to be a formal program to assure interoperability. They could, like Sony has done, hold back some features from the 3rd party, but at least consumers would know what works are what does not when they buy a lens.
I believe that Canon at least has to think about their strategy to nix 3rd party lenses completely in the wake of all the bad press they are getting. They have gotten pretty much uniformly bad press from every reviewer, even ones that are highly favorable to Canon. While I own about $20K worth of RF bodies and lenses, I would not recommend to a family member or friend to buy into the RF system, I would recommend Sony (I think Nikon is riskier than Canon).
You have to be as dominant as Microsoft was in PCs in the late 1980s for it to possibly be an Anti-trust matter, and even Microsoft got away with "Windows isn't done until Lotus won't run" (and worse) behavior.