Don’t expect any third-party autofocus lenses in the near future

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
I'm wondering how long it was after EF came out before Canon licensed it?
I don't think Canon ever licensed the EF mount.

All third-party EF lenses are reverse-engineered, but presumably Sigma, Tamron etc didn't infringe any patents in doing so.

Where Viltrox (and Samyang?) have come adrift this time, is that Canon apparently patented the RF protocols and electronics pertaining to AF.

Canon has not stopped Laowa or any of the other third parties who manufacture entirely manual lenses with no electronics.
 
  • Like
Reactions: 2 users
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,440
22,875
I don't think Canon ever licensed the EF mount.

All third-party EF lenses are reverse-engineered, but presumably Sigma, Tamron etc didn't infringe any patents in doing so.

Where Viltrox (and Samyang?) have come adrift this time, is that Canon apparently patented the RF protocols and electronics pertaining to AF.

Canon has not stopped Laowa or any of the other third parties who manufacture entirely manual lenses with no electronics.
It's been posted several times here but is perhaps being lost as there are so may duplicate threads with so many replies. Canon could have indeed patented the RF protocols but it is perfectly legal to reverse engineer them. However, if a competitor had obtained the code by deception, theft etc, then it would be illegal to use it. It has been reported several times but I cannot vouch for it that Canon has licensed to Sigma and Tamron but would not license to a non-Japanese company.
 
  • Like
Reactions: 1 user
Upvote 0

unfocused

Photos/Photo Book Reviews: www.thecuriouseye.com
Jul 20, 2010
7,184
5,484
70
Springfield, IL
www.thecuriouseye.com
It's been posted several times here but is perhaps being lost as there are so may duplicate threads with so many replies. Canon could have indeed patented the RF protocols but it is perfectly legal to reverse engineer them. However, if a competitor had obtained the code by deception, theft etc, then it would be illegal to use it...
You keep referencing "obtained the code by deception, theft etc." but that's not how patents work. Stealing something is a criminal offense, so obtaining corporate secrets through theft or deception would be a matter to take to criminal court. Protecting a patent is a property right. A patent case does not need to prove that the information was obtained illegally. It just has to prove that someone is using your patented design. How they came up with that design is irrelevant. Even if you come up with the design all on our own, without any knowledge of the patent, if it duplicates one that is patented then you are infringing on the patent.

I don't think Canon ever licensed the EF mount.

All third-party EF lenses are reverse-engineered, but presumably Sigma, Tamron etc didn't infringe any patents in doing so.

Where Viltrox (and Samyang?) have come adrift this time, is that Canon apparently patented the RF protocols and electronics pertaining to AF.

Canon has not stopped Laowa or any of the other third parties who manufacture entirely manual lenses with no electronics.

I agree with your first two statements. As for the third, we do not know where Viltrox came "adrift" and probably shouldn't make any assumptions. As far as I know there has been no court case filed, so we really don't know what Canon has objected to.

It has been reported several times but I cannot vouch for it that Canon has licensed to Sigma and Tamron but would not license to a non-Japanese company.
Where has that been reported? My understanding is that @entoman is correct that Canon never licensed the EF designs to anyone, but that Sigma and other third party manufacturers had designs that did not violate the patents. Most people assume they came up with those designs through reverse engineering, which certainly sounds plausible.
 
  • Like
Reactions: 1 users
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
I agree with your first two statements. As for the third, we do not know where Viltrox came "adrift" and probably shouldn't make any assumptions. As far as I know there has been no court case filed, so we really don't know what Canon has objected to.

From Canon Germany (Google Translated)​

“SHENZHEN JUEYING TECHNOLOGY CO.LTD, manufactures auto focus lenses for Canon RF mount under the brand name “Viltrox”. Canon believes that these products infringe their patent and design rights and has therefore requested the company to stop all activities infringing on Canon’s intellectual property rights.”

According to Canon Germany, Viltrox infringed Canon's "patent and design rights".
Canon doesn't prevent Laowa from producing RF mount manual lenses.
I think therefore that it's safe to assume that the patent infringement refers to the RF protocols and lens electronics pertaining to AF.
The RF protocols will also presumably include things pertaining to the "control ring", the electronic iris, function buttons on big whites etc.
 
  • Like
Reactions: 1 users
Upvote 0
Aug 27, 2019
667
1,414
It's been posted several times here but is perhaps being lost as there are so may duplicate threads with so many replies. Canon could have indeed patented the RF protocols but it is perfectly legal to reverse engineer them. However, if a competitor had obtained the code by deception, theft etc, then it would be illegal to use it. It has been reported several times but I cannot vouch for it that Canon has licensed to Sigma and Tamron but would not license to a non-Japanese company.
I would not be at all surprised if there was some patent infringement going on.

In my day job we work with the ONVIF protocol for all third party devices connecting to our VMS. Non native devices using ONVIF are limited both on our side (VMS) and on the third party side (Device) unless we have an agreement in place.

Sadly when dealing with some of the Chinese brands we often find where they have attempted to talk directly to our hardware protocol layer. This is a MAJOR no no so we will not certify the use of these devices.

The crazy things I have seen in 20 years of trying to certified anything from POE switches that cannot negotiate who is going to provide power in a switch to switch connection to malicious attempts to capture firmware data once connected to our network. Just last month we had a device that cooked off so violently we had to evacuate the lab and one of our 200K environmental chambers was pretty messed up.

Just my opinion but no amount of money saved is worth me risking using any of these non certified 4th party manufactures.
 
  • Like
Reactions: 1 users
Upvote 0

AlanF

Desperately seeking birds
CR Pro
Aug 16, 2012
12,440
22,875
You keep referencing "obtained the code by deception, theft etc." but that's not how patents work. Stealing something is a criminal offense, so obtaining corporate secrets through theft or deception would be a matter to take to criminal court. Protecting a patent is a property right. A patent case does not need to prove that the information was obtained illegally. It just has to prove that someone is using your patented design. How they came up with that design is irrelevant. Even if you come up with the design all on our own, without any knowledge of the patent, if it duplicates one that is patented then you are infringing on the patent.
It is far more complicated than you think. Read this from beginning to end, especially the end, https://en.wikipedia.org/wiki/Reverse_engineering where it is explained. Some court cases about reverse engineering are also summarised in wiki https://en.wikipedia.org/wiki/Atari_Games_Corp._v._Nintendo_of_America_Inc. https://en.wikipedia.org/wiki/Bowers_v._Baystate_Technologies,_Inc.
Reverse engineering per se is not illegal, but can be illegal if deception is involved, end user agreements are breached etc.

Where has that been reported? My understanding is that @entoman is correct that Canon never licensed the EF designs to anyone, but that Sigma and other third party manufacturers had designs that did not violate the patents. Most people assume they came up with those designs through reverse engineering, which certainly sounds plausible.
I wrote: "It has been reported several times but I cannot vouch for it that Canon has licensed to Sigma and Tamron but would not license to a non-Japanese company." The meaning should be clear: I have seen it reported many times but have never seen the evidence that supports the assertions. I don't keep records of gossip.
 
Upvote 0
Although I have a Canon R5 and am heavily-invested in Canon RF glass, Canon's approach to third-party lenses really does limit my glass options. In my own view, this is a strong reason for a person to not purchase a Canon camera, but to go with a manufacturer that has both a) excellent cameras and glass and b) accepts third-party lenses, some of which are outstanding in terms of optics and value.

Personally, I purchased an R5 as an upgrade from my two 6Ds after many years of use. I purchased it partially on the spec and partially on the advice of a friend who shoots birds. He recommended the Sigma Sport 150-600 for my bird photography and I was waiting for the Sigma Sport 150-600 DN to come out with the RF mount. Then I saw that Canon was blocking 3rd party lenses. If I had known that initially, I probably wouldn't have purchased the R5 even though I have 8 or 9 EF mount lenses - mostly L series.

I wanted the new Sigma Sport 150-600 DN series for my bird photography because it it much lighter, costs less than the EF version and according to reviews is sharper at the 600 end than the older Sport 150-600 EF HSM OS version. When you are hiking for nature shots, that extra weight counts.

So my question is as follows: Canon is NOT objecting to EF-RF converters made by 3rd parties, just check out B&H or Adorama and you'll see them available from multiple vendors. I would assume that means the actual RF mount can't be or wasn't patented, otherwise Canon would be objecting to those converters. Likewise, I believe that Sigma has stated that they will convert the Sport 150-600 DN between different mounts for a fee, Sony to Nikon etc.

So.... Why doesn't Sigma build the Sport 150-600 DN with the RF mount and EF focusing / stabilization protocols with the public statement that once Canon opens up the RF mount specs, they will either offer a firmware upgrade to full RF specs OR build a proper RF mount that can be installed for a reasonable fee?

Just a thought...
 
  • Like
Reactions: 1 users
Upvote 0
Aug 27, 2019
667
1,414
  • Like
Reactions: 2 users
Upvote 0

unfocused

Photos/Photo Book Reviews: www.thecuriouseye.com
Jul 20, 2010
7,184
5,484
70
Springfield, IL
www.thecuriouseye.com
It is far more complicated than you think. Read this from beginning to end, especially the end, https://en.wikipedia.org/wiki/Reverse_engineering where it is explained. Some court cases about reverse engineering are also summarised in wiki https://en.wikipedia.org/wiki/Atari_Games_Corp._v._Nintendo_of_America_Inc. https://en.wikipedia.org/wiki/Bowers_v._Baystate_Technologies,_Inc.
Reverse engineering per se is not illegal, but can be illegal if deception is involved, end user agreements are breached etc.
Thank you for the references. Informative. There is nothing in there that conflicts with what I wrote. In fact it confirms my points.
...I don't keep records of gossip.
But, you were willing to repeat gossip.
 
  • Like
Reactions: 1 user
Upvote 0

unfocused

Photos/Photo Book Reviews: www.thecuriouseye.com
Jul 20, 2010
7,184
5,484
70
Springfield, IL
www.thecuriouseye.com

From Canon Germany (Google Translated)​



According to Canon Germany, Viltrox infringed Canon's "patent and design rights".
Canon doesn't prevent Laowa from producing RF mount manual lenses.
I think therefore that it's safe to assume that the patent infringement refers to the RF protocols and lens electronics pertaining to AF.
The RF protocols will also presumably include things pertaining to the "control ring", the electronic iris, function buttons on big whites etc.
First, I was agreeing with your main points.

Second, if I read your reasoning correctly, you would include everything other than the physical mount in the category of RF protocols and lens electronics. So, sure, if you use a broad definition, that is probably correct. My point, which really shouldn't be a point of contention, is that we are not privy to the specifics of Canon's objections so it's safer not to assume what their objections entail. Not a big deal and certainly not worth debating.
 
Upvote 0
Aug 27, 2019
667
1,414
As I said on another forum the mental gymnastics required to make the leap from Canon pushing back at Viltrox and maybe Samyang to Canon WILL NEVER ALLOW 3RD PARTY glass on the RF mount is truly epic.

If only there was not people like Tony N or all the other countless channels and forums that rely on views and clicks to generate revenue we could be discussing the facts as they are known today:

Current set of facts I am aware of:
  • Canon has asked Viltox to stop selling things with the RF mount
  • There are no 3rd party options for RF from Sigma and Tamron
Rumors:
  • Samyang has also been asked to stop making autofocus RF glass.
So for those that wanted a Vitrox RF product best to get it ASAP and maybe the same for Samyang. Those that want 3rd party glass from Sigma and Tamron you will have to wait and see.

I know this is not as sexy as
  • "Canon is Doomed"
  • "Canon must be stopped"
  • "Do not buy Canon"
  • "Canon stole my truck and dog but left the wife, damn them"

Cheers
 
Last edited:
  • Like
  • Haha
Reactions: 2 users
Upvote 0

entoman

wildlife photography
May 8, 2015
1,998
2,438
UK
Although I have a Canon R5 and am heavily-invested in Canon RF glass, Canon's approach to third-party lenses really does limit my glass options. In my own view, this is a strong reason for a person to not purchase a Canon camera, but to go with a manufacturer that has both a) excellent cameras and glass and b) accepts third-party lenses, some of which are outstanding in terms of optics and value.

Yes, if you *truly need* a particular lens that isn't available in RF mount, and isn't on the Canon roadmap, it makes sense to switch to Sony.

But, it's a very expensive move, as you'll have to buy a complete new set of lenses for the new body, not just that one specialised lens, so you really have to ask yourself how badly you want that lens.

Consider also that the lens you want is almost certainly available in EF mount, and that EF lenses function 100% perfectly on RF bodies. An EF-RF adaptor costs a hell of a lot less than a Sony body!

Also, switching brands means that you have to relearn all the camera controls, menu system, AF quirks and adapt to new ergonomics. Don't underestimate the difficulties - while some people can switch back and forth from one brand to another, a lot of people will find that it takes time to adapt, and that they'll miss a lot of shots due to lack of familiarity with a new camera.
 
  • Like
Reactions: 1 user
Upvote 0
Jul 21, 2010
31,202
13,071
Yes, if you *truly need* a particular lens that isn't available in RF mount, and isn't on the Canon roadmap, it makes sense to switch to Sony.

But, it's a very expensive move, as you'll have to buy a complete new set of lenses for the new body, not just that one specialised lens, so you really have to ask yourself how badly you want that lens.

Consider also that the lens you want is almost certainly available in EF mount, and that EF lenses function 100% perfectly on RF bodies. An EF-RF adaptor costs a hell of a lot less than a Sony body!

Also, switching brands means that you have to relearn all the camera controls, menu system, AF quirks and adapt to new ergonomics. Don't underestimate the difficulties - while some people can switch back and forth from one brand to another, a lot of people will find that it takes time to adapt, and that they'll miss a lot of shots due to lack of familiarity with a new camera.
Well the above is true and quite relevant to a discussion on a forum dedicated to camera gear, I don’t believe any of that really applies to the ‘typical consumer’.

I suspect for most people it’s, ‘I need/want a new camera, I have/had a Canon, it was ok, I’ll buy another Canon.’ You can swap in toaster, tv, car, and whatever brand you like. That’s typical consumer behavior. It’s why, unless something truly paradigm-shifting comes along, market share changes slowly at best.
 
  • Like
Reactions: 1 user
Upvote 0

jd7

CR Pro
Feb 3, 2013
1,064
418
It is far more complicated than you think. Read this from beginning to end, especially the end, https://en.wikipedia.org/wiki/Reverse_engineering where it is explained. Some court cases about reverse engineering are also summarised in wiki https://en.wikipedia.org/wiki/Atari_Games_Corp._v._Nintendo_of_America_Inc. https://en.wikipedia.org/wiki/Bowers_v._Baystate_Technologies,_Inc.
Reverse engineering per se is not illegal, but can be illegal if deception is involved, end user agreements are breached etc.
It is correct that, per se, reverse engineering does not expose you to legal liability. The point is, though, that relates specifically to the reverse engineering process, ie pulling something apart to learn how it works. To make a product based on what you have learned is a separate step, and if your product infringes another person's intellectual property, you have a problem even if there was nothing illegal about you reverse engineering the other product to learn how it works. If what you learned was simply something which was previously a trade secret, it is unlikely you will be infringing IP by using what you learned in your product (unless, as you have already noted, you have breached an end user agreement, been deceptive, etc). However, with other forms of IP (ie other than confidential information) such as patents, copyright, registered designs, integrated circuit layout rights, etc, the situation is different. So, for example, reverse engineering the RF mount is one thing, developing a lens which can use the RF mount is another. In particular, what we seem to be seeing is that developing a lens which makes use of the electronics in an RF mount must be difficult or perhaps even impossible to do without infringing Canon's IP (in particular, it seems, infringing one or more Canon patents).
 
  • Like
Reactions: 1 users
Upvote 0
I will not go into any details as I am not sure what is part of the public record but I work for MSI (Motorola Solutions) and we won a big case against Hytera in 2020 - https://www.motorolasolutions.com/n...t-theft-and-copyright-infringement-lawsu.html . Our patents are our life blood and the company is very serious about protecting them.
First of all, I'm not a lawyer. That said, I think you are comparing Apples to Oranges in this case. Yes, Motorola enforces their patents. And good for them. Either Canon did not patent the EF protocols ( according to one poster here ) OR Canon did not enforce their EF patents for about 35 years ( from 1987 onwards ) OR Sigma legitimately reverse engineered the EF mount communications protocols. So my understanding of patent law is that you have to enforce your patents or essentially, you lose them. They may have to enforce the current RF physical mount patent, as used on the EF to RF adapters, assuming they have one and send a Cease and Desist letters to the various makers of these adapters to retain that patent but it would appear to me that the EF focus / stabilization protocols could be used based on Canon's previous behaviors. Like I said previously, make a Sigma Sport 150-600 DN with Canon RF physical mount using Sigma's EF communications protocols with the promise that once they work things out with Canon, that mount can be upgraded either through firmware or by sending the lens back to Sigma to replace the original RF style mount with a new, true RF Mount.

I think most Canon R series users could live with this solution.
 
Upvote 0

unfocused

Photos/Photo Book Reviews: www.thecuriouseye.com
Jul 20, 2010
7,184
5,484
70
Springfield, IL
www.thecuriouseye.com
It is correct that, per se, reverse engineering does not expose you to legal liability. The point is, though, that relates specifically to the reverse engineering process, ie pulling something apart to learn how it works. To make a product based on what you have learned is a separate step, and if your product infringes another person's intellectual property, you have a problem even if there was nothing illegal about you reverse engineering the other product to learn how it works. If what you learned was simply something which was previously a trade secret, it is unlikely you will be infringing IP by using what you learned in your product (unless, as you have already noted, you have breached an end user agreement, been deceptive, etc). However, with other forms of IP (ie other than confidential information) such as patents, copyright, registered designs, integrated circuit layout rights, etc, the situation is different. So, for example, reverse engineering the RF mount is one thing, developing a lens which can use the RF mount is another. In particular, what we seem to be seeing is that developing a lens which makes use of the electronics in an RF mount must be difficult or perhaps even impossible to do without infringing Canon's IP (in particular, it seems, infringing one or more Canon patents).
Thanks for this thoughtful post.

I think a lot of times on this forum people use "reverse engineering" without understanding what it is and that it isn't some magical way around patent infringement.

As I understand it, and as you state in your post, reverse engineering simply means deconstructing something to figure out how it works. As photographers, most of us have reverse engineered images that we like, figuring out the lighting, focal length, point of view, processing, etc.

Roger's team at Lens Rentals often do tear downs of cameras to see how they work and how they are constructed. They are essentially reverse engineering the bodies, but they are not violating any patent because they aren't making new cameras from what they learned. As you point out, it's how you use that information that matters.

What I have tried to explain, and I hope you would agree, is that for patents it really doesn't matter how you get there. One could have a "Eureka" moment in the middle of the night, get up and design a new mirrorless lens system without any awareness or improper copying of Canon's system and still be in violation of patents if the end result is too similar to what has already been patented.
 
  • Like
Reactions: 1 user
Upvote 0
Aug 27, 2019
667
1,414
That said, I think you are comparing Apples to Oranges in this case.
Not at all... patent infringement is patent infringement. There is either only apples or there are only oranges.
Canon did not patent the EF protocols ( according to one poster here ) OR Canon did not enforce their EF patents for about 35 years ( from 1987 onwards ) OR Sigma legitimately reverse engineered the EF mount communications protocols.
Canon has one of the largest patent portfolio's in the world and are also one of the most successful litigators of patent infringement. If Sigma infringed on there IP they would have been sued. That said Sigma is a true 3rd party brand and have a reputation to uphold. So there is no way they would release a product that Canon can buy in any store, take apart and find firmware\hardware that is a direct copy of there design. Pro Tip: Reverse Engineering is not cloning.

These 4th party Chinese manufacturers often just copy the thing firmware\hardware and all with no attempt to hide what they are doing. Check out the Shuanghuan SCEO or the CH Lithia and then have a look at a BMW X5 and a Audi R8.

So my understanding of patent law is that you have to enforce your patents or essentially, you lose them.
Not really that simple... But in this case it is completely irrelevant since Canon was still filing patents on the EF mount in 2020 - https://petapixel.com/2020/01/23/new-patents-show-canon-hasnt-given-up-on-the-ef-mount-just-yet/

Overall I would say your level of understanding on this topic is a little on the light side and you might want to do a little more research. Link to Canons patents - https://www.canonwatch.com/category/patents/canon-patents/

Have fun there is a lot to read.

Oh and here is another great read on Canon and there Patent policy - https://global.canon/en/intellectual-property/sword-shield/

Cheers
 
  • Like
Reactions: 1 user
Upvote 0

jd7

CR Pro
Feb 3, 2013
1,064
418
Thanks for this thoughtful post.

I think a lot of times on this forum people use "reverse engineering" without understanding what it is and that it isn't some magical way around patent infringement.

As I understand it, and as you state in your post, reverse engineering simply means deconstructing something to figure out how it works. As photographers, most of us have reverse engineered images that we like, figuring out the lighting, focal length, point of view, processing, etc.

Roger's team at Lens Rentals often do tear downs of cameras to see how they work and how they are constructed. They are essentially reverse engineering the bodies, but they are not violating any patent because they aren't making new cameras from what they learned. As you point out, it's how you use that information that matters.

What I have tried to explain, and I hope you would agree, is that for patents it really doesn't matter how you get there. One could have a "Eureka" moment in the middle of the night, get up and design a new mirrorless lens system without any awareness or improper copying of Canon's system and still be in violation of patents if the end result is too similar to what has already been patented.
Yes, I would agree with that regarding patents. The situation isn't necessarily the same for all types of IP though. For example, for copyright infringment, essentially there needs to be some relevant dealing with the copyright work. So, if I come up with something which is identical to your copyright work but I didn't copy your work (eg I didn't even know you had ever even created your copyright work so I couldn't have copied it), that isn't copyright infringement. Obviously, that would be an unusual situation though!
 
Last edited:
  • Like
Reactions: 1 user
Upvote 0
Jul 27, 2021
193
190
Yes, if you *truly need* a particular lens that isn't available in RF mount, and isn't on the Canon roadmap, it makes sense to switch to Sony.

But, it's a very expensive move, as you'll have to buy a complete new set of lenses for the new body, not just that one specialised lens, so you really have to ask yourself how badly you want that lens.

Consider also that the lens you want is almost certainly available in EF mount, and that EF lenses function 100% perfectly on RF bodies. An EF-RF adaptor costs a hell of a lot less than a Sony body!

Also, switching brands means that you have to relearn all the camera controls, menu system, AF quirks and adapt to new ergonomics. Don't underestimate the difficulties - while some people can switch back and forth from one brand to another, a lot of people will find that it takes time to adapt, and that they'll miss a lot of shots due to lack of familiarity with a new camera.
In some instances a lot of money can be saved depending on a particular shooter’s circumstances by switching brands. In particular comparing OEM 35mm f1.4 prices its £2099 for the Canon L II vs £1499 for the Sony GM.

Another example is super telephoto primes, if someone wants a top of the line 800mm the Canon options are £19’099 for the RF, £13’549 for the EF or they could buy the Nikon Z for £6’299. Its a stop slower but being less than half the price many will happily accept that trade off.

There are more examples of this but in both cases the option of using an RF adapter results in a user having to spend significantly more money for an older and heavier lens.

There are many other instances where a certain lens simply does not exist on EF at all so adapting isn’t always a solution.
 
Upvote 0