Sigma: Our RF lenses have been a great success

[David_B]

the physical mount itself is patented.

Depending on what you mean by "the physical mount itself is patent", I can't see how. They could patent the specific action of how the lens and camera fit together, but not just the physical design of the mount in the camera. Patents are for describing about how a combination of two or more things work together, so the mount plate in the camera wouldn't be paentable but maybe the action of the lens clicking into place is, along with the comms between the lens and camera.

 

[Bob Howland]

Maybe the sticking point between Canon and Sigma was whether Canon would fully disclose the communications protocol to Sigma and any other licensee. I understand that Sigma had problems reverse engineering the EF protocol and may have wanted to avoid a costly repetition.

 

[koenkooi]

Or if Canon brought out a successor to the M6-II (with at least 30MP) ... yeah, crickets.

While that would be a massive step in the right direction, it would likely be very close to the size of the R8, which I already have and love.
But I agree that a high performance,
compact crop body is missing from the Canon line up.

 

[Richard CN]

Depending on what you mean by "the physical mount itself is patent", I can't see how. They could patent the specific action of how the lens and camera fit together, but not just the physical design of the mount in the camera. Patents are for describing about how a combination of two or more things work together, so the mount plate in the camera wouldn't be paentable but maybe the action of the lens clicking into place is, along with the comms between the lens and camera.

a mount has two aspects that do connect. the RF mount has the camera side mount, and also the lens side mount. and how they work together was something that was discussed in patent applications. specifically about load bearing if i recall. if i recall correctly that is why Canon decided the uppermost lug is larger than the other two.

so many patents I've read over the last 6+ years but I'm pretty confident that Canon did apply for something of the sort.

 

[Stig Nygaard]

That was a reply to the statement;

Viltrox did make RF AF lenses: here is a review of one of them etc.

Viltrox RF 85mm F1.8 STM Review - DustinAbbott.net

Photographer Dustin Abbott shares a thorough review of the Viltrox RF 85mm F1.8 STM - a budget autofocus prime lens for Canon RF mirrorless

dustinabbott.net

Hi Alan

EDIT: I just see I'm replying to a message I thought was a reply to me (the CR notification was saying that - probably because I was quoted). But at closer look, you were probably replying to someone else. Now I have written it, I keep it here. But I know I probably misunderstood your post ...


I don't know what your point is. It sounds like you totally misunderstood my point.

I know the story about Viltrox and their lenses (and I actually have that Viltrox 85mm AF lens!)

My point is that some like to think that Canon has a principal decision not to allow fullframe AF lenses, but to allow AF crop-lenses.
I see no reason to conclude that in general, even though it just happened to be that the first licensed lenses are all crop-lenses.

The story with Viltrox was before Canon partly opened up their mount. Later Canon have stated that they wanted the mount (protocol) to be more mature before they wanted to allow others to make lenses for it. They have then stated that they now look at it on a case-by-case basis, when giving licensing to third-party lens makers. But at no time have they mentioned there should be a principal distinction between fullframe and crop.

So the first licensed AF-lenses that are hitting the market (6 from Sigma, 1 from Tamron) are all RF-S lenses. That has made many people conclude as a fact that Canon allows lens-makers to make crop-lenses, but not to make fullframe lenses. My objection is to those that believe or threat that as a fact. Especially for two reasons:
1) If I was a third-party lens-maker I would also start with the crop-lenses because that is clearly where there was the largest gaps in Canon's lineup
2) What advantage would it have for Canon to take a principal decision not to allow fullframe lenses, when they already license on a case-by-case basis?

Again, we don't know. I don't know! I'm just objecting to those that say it as a fact that Canon doesn't allow fullframe lenses being made, because they don't know either!

 

[P-visie]

It’s a compelling narrative, especially for places that conflate ‘full frame mirrorless’ with ‘the whole market’.
You can generally spot those places by the amount of glee accompanying the use of words like ‘cripple’.

And the narrative creates a lot of clicks and views (see this thread) and heated emotions.

 

[P-visie]

Hi Alan

EDIT: I just see I'm replying to a message I thought was a reply to me (the CR notification was saying that - probably because I was quoted). But at closer look, you were probably replying to someone else. Now I have written it, I keep it here. But I know I probably misunderstood your post ...


I don't know what your point is. It sounds like you totally misunderstood my point.

I know the story about Viltrox and their lenses (and I actually have that Viltrox 85mm AF lens!)

My point is that some like to think that Canon has a principal decision not to allow fullframe AF lenses, but to allow AF crop-lenses.
I see no reason to conclude that in general, even though it just happened to be that the first licensed lenses are all crop-lenses.

The story with Viltrox was before Canon partly opened up their mount. Later Canon have stated that they wanted the mount (protocol) to be more mature before they wanted to allow others to make lenses for it. They have then stated that they now look at it on a case-by-case basis, when giving licensing to third-party lens makers. But at no time have they mentioned there should be a principal distinction between fullframe and crop.

So the first licensed AF-lenses that are hitting the market (6 from Sigma, 1 from Tamron) are all RF-S lenses. That has made many people conclude as a fact that Canon allows lens-makers to make crop-lenses, but not to make fullframe lenses. My objection is to those that believe or threat that as a fact. Especially for two reasons:
1) If I was a third-party lens-maker I would also start with the crop-lenses because that is clearly where there was the largest gaps in Canon's lineup
2) What advantage would it have for Canon to take a principal decision not to allow fullframe lenses, when they already license on a case-by-case basis?

Again, we don't know. I don't know! I'm just objecting to those that say it as a fact that Canon doesn't allow fullframe lenses being made, because they don't know either!

Go Tokura is quoted on third party RF lenses on Petapixel: “A common question regarding third-party optics is why Canon has been so unwilling to allow companies, even reputable ones like Sigma and Tamron, to make optics for its RF mount. When asked, Tokura provided an unexpectedly honest answer: it’s just business.
“So basically, the reason why we have not been collaborating or partnering with a third party is because of our business strategy. Our business strategy says that we are not going to partner with a third party. And the reason why we came up with that strategy is because of the business that we are doing. So our business is that if we have some technology, we try to lock them up inside of our company and try to make a profit out of that. That’s the basis of our business,” Tokura says. “

See: https://petapixel.com/2024/03/06/canon-is-actively-working-to-bring-third-party-lenses-to-rf-mount/

Canon wanted to flesh out the RF lens line up before allowing third party lenses. This, just like preventing other companies from using your IP, is sound business practice.

Unfortunately, lack of knowledge does not prevent a lot of people of making bold sweeping statements which generates of a lot of clicks and views (and income for the persons making these sweeping statements). Canon’s lack of (clear) communication does not help either.

 

[AlanF]

Hi Alan

EDIT: I just see I'm replying to a message I thought was a reply to me (the CR notification was saying that - probably because I was quoted). But at closer look, you were probably replying to someone else. Now I have written it, I keep it here. But I know I probably misunderstood your post ...

You did! @P-visie 's post makes it clear. Canon does it best to lock up its technology.

 

[neuroanatomist]

“So our business is that if we have some technology, we try to lock them up inside of our company and try to make a profit out of that. That’s the basis of our business,” Tokura says.

Still baffles me why people can’t grasp this simple concept.

 

[AlanF]

Depending on what you mean by "the physical mount itself is patent", I can't see how. They could patent the specific action of how the lens and camera fit together, but not just the physical design of the mount in the camera. Patents are for describing about how a combination of two or more things work together, so the mount plate in the camera wouldn't be paentable but maybe the action of the lens clicking into place is, along with the comms between the lens and camera.

You have it 100% the wrong way around. They can patent the physical design of the camera mount, preventing anyone else making an identical mount and sticking it on their own camera frame. But they can't patent the complementary mount that fits into it. In other words, they can't "patent the specific action of how the lens and camera fit together".

IP law is that you can patent a mount (or any legal object) but you cannot patent attachments to that mount. So, any third party manufacturer can legally make a lens mount that fits on to that mount. The same is true for vacuum cleaners or any other device.
IP law allows anyone to reverse engineer a protocol legally. However, if the reverse engineering is actually based on illegally seeing a patented or secret protocol, it is illegal (as alluded to by @koenkooi that Viltrox may have done).
The consequence is that Viltrox and everyone else can legally make manual AF lenses for Canon RF lenses and no way can Canon stop that by law. For AF lenses, they need to design their own communications protocols without seeing Canon's.

 

[Richard CN]

And the narrative creates a lot of clicks and views (see this thread) and heated emotions.

clicks and heated emotions are like popcorn to us authors of said posts

 

[Richard CN]

You have it 100% the wrong way around. They can patent the physical design of the camera mount, preventing anyone else making an identical mount and sticking it on their own camera frame. But they can't patent the complementary mount that fits into it. In other words, they can't "patent the specific action of how the lens and camera fit together".

wouldn't the mechanical design of both the lens and camera side of the mount be applicable?

I'm 99% sure I saw that. but i really don't want to go looking through canonnews archives to find it

 

[AlanF]

wouldn't the mechanical design of both the lens and camera side of the mount be applicable?

I'm 99% sure I saw that. but i really don't want to go looking through canonnews archives to find it

Richard, I've dug up two patent articles;
https://www.vennershipley.com/insights-events/design-protection-of-components/

https://www.vennershipley.com/insights-events/design-protection-of-spare-parts/

The first excludes the lens mount combination because the joint between them is invisible in use. The second is more powerful and general about things that plug into each other.

"Must-fit exclusion
Another provision that affects the protection of spare parts is the “must-fit” exclusion. Article 8(2) of the CDR recites “a Community design shall not subsist in features of appearance of a product which must necessarily be reproduced in their exact form and dimensions in order to permit the product in which the design is incorporated or to which it is applied to be mechanically connected to or placed in, around or against another product so that either product may perform its function”. This exclusion is common to all of the registered and unregistered design regimes we have discussed so far.

In essence, the exclusion prohibits the protection of features of an article that are required to interconnect with another article. A common example is that of a plug and socket, wherein the three (or two, depending on where you are reading this from!) prongs of a power plug and corresponding socket holes are necessary to enable an interconnection and thus would be excluded from design protection. The exclusion only applies to the features that fall within the must-fit provision, and therefore a design may nevertheless be valid if it includes other features that contribute to the novelty and individual character of the article."

In other words, Canon can patent the lens design but not the bit at the end that fits into the camera because that feature is excluded because it is a "Must fit".

 

[SvO2]

What advantage would it have for Canon to take a principal decision not to allow fullframe lenses, when they already license on a case-by-case basis?

“So basically, the reason why we have not been collaborating or partnering with a third party is because of our business strategy. Our business strategy says that we are not going to partner with a third party. And the reason why we came up with that strategy is because of the business that we are doing. So our business is that if we have some technology, we try to lock them up inside of our company and try to make a profit out of that. That’s the basis of our business,” Tokura says. “

That seems to pretty directly answer why Canon's not currently allowing full frame lenses. Only way I see it changing honestly is if Sony starts pulling ahead and availability of 3rd party lenses is perceived to be part of that. While that seems to be the case for some, it doesn't seem to be happening in large enough numbers to matter to Canon.

wouldn't the mechanical design of both the lens and camera side of the mount be applicable?

I'm 99% sure I saw that. but i really don't want to go looking through canonnews archives to find it

Based on Canon seemingly never going after EF lens makers and similarly seemingly not having a problem with manual focus RF lenses, I suspect there probably is some exception to that end. Difference with the RF mount is it was launched in an era where adding encryption and DRM became easy, and bypassing DRM generally is generally not legal in most jurisdictions. Now I suspect that's not the *entire* story - it generally would make me wonder why at least the the lesser known 3rd parties don't just use EF protocol communications over the RF contacts (essentially emulating the EF-RF adapters).

 

[AlanF]

That seems to pretty directly answer why Canon's not currently allowing full frame lenses. Only way I see it changing honestly is if Sony starts pulling ahead and availability of 3rd party lenses is perceived to be part of that. While that seems to be the case for some, it doesn't seem to be happening in large enough numbers to matter to Canon.


Based on Canon seemingly never going after EF lens makers and similarly seemingly not having a problem with manual focus RF lenses, I suspect there probably is some exception to that end. Difference with the RF mount is it was launched in an era where adding encryption and DRM became easy, and bypassing DRM generally is generally not legal in most jurisdictions. Now I suspect that's not the *entire* story - it generally would make me wonder why at least the the lesser known 3rd parties don't just use EF protocol communications over the RF contacts (essentially emulating the EF-RF adapters).

As I wrote immediately above, you can't in IP law patent both a socket and a plug. And DRM does not prevent you independently writing your own communication protocol between the two.

 

[SvO2]

DRM does not prevent you independently writing your own communication protocol between the two.

How do you figure? The lens manufacturers don't really have any control over what the Camera will accept as far as communications. If the camera requires signed and/or encrypted messages to actually do anything, you can't effectively bypass that without either breaking the DRM (difficult to impossible if well implemented, and illegal even if you do find a flaw in the scheme) or getting permission and the keys from Canon (which seems to be the rate limiting step)

 

[AlanF]

How do you figure? The lens manufacturers don't really have any control over what the Camera will accept as far as communications. If the camera requires signed and/or encrypted messages to actually do anything, you can't effectively bypass that without either breaking the DRM (difficult to impossible if well implemented, and illegal even if you do find a flaw in the scheme) or getting permission and the keys from Canon (which seems to be the rate limiting step)

It's standard knowledge about IP, which you can Google for articles if you wish. It might be difficult but it's perfectly legal to write your own communication protocols. You can't patent an idea, only a specific novel application developing from that idea. And patent law interpretations try to prevent monopolies.

 

[SvO2]

It's standard knowledge about IP, which you can Google for articles if you wish. It might be difficult but it's perfectly legal to write your own communication protocols. You can't patent an idea, only a specific novel application developing from that idea. And patent law interpretations try to prevent monopolies.

You can write your communication protocol, but how do you plan on getting the camera to communicate in your protocol if it's expecting cryptographically signed messages in its own protocol? Lens manufacturers have no control over the firmware running on a camera. You can't just "translate" to Canon's protocols and sign your messages without either having the keys or finding a flaw in the DRM implementation (which again, would be illegal to break)

 

[AlanF]

You can write your communication protocol, but how do you plan on getting the camera to communicate in your protocol if it's expecting cryptographically signed messages in its own protocol? Lens manufacturers have no control over the firmware running on a camera. You can't just "translate" to Canon's protocols and sign your messages without either having the keys or finding a flaw in the DRM implementation (which again, would be illegal to break)

That's not my problem. I am just trying to transmit the information on what you can patent and what you cannot. Yongnuo is selling RF lenses and hasn't been shut down by Canon so presumably they have cracked the problem somehow.

YONGNUO Lenses - Best Third Party Lens

YONGNUO provide lenses for Sony E, Canon EF/RF, Nikon F/Z and MFT mount camera. Free Shipping1 Year Warranty7 Days Return

www.th.hkyongnuo.com

 

[SvO2]

That's not my problem. I am just trying to transmit the information on what you can patent and what you cannot. Yongnuo is selling RF lenses and hasn't been shut down by Canon so presumably they have cracked the problem somehow.

YONGNUO Lenses - Best Third Party Lens

YONGNUO provide lenses for Sony E, Canon EF/RF, Nikon F/Z and MFT mount camera. Free Shipping1 Year Warranty7 Days Return

www.th.hkyongnuo.com

You literally stated the below...

DRM does not prevent you independently writing your own communication protocol between the two.

You can claim "not my problem", but my entire point is that one cannot legally break DRM. If there's no DRM, it's trivial to reverse engineer these things. A well implemented DRM system is impossible to bypass without quantum computers. Most are not perfectly implemented, but even if you do manage to bypass it, the act of bypassing DRM is illegal in the large majority of jurisdictions around the world. Depending on how you bypass it, the manufacturers can also often issue updates to render such methods useless (as we saw with early 3rd party lenses). That's not a patent protection, but a DMCA protection (in the US, its equivalents elsewhere)

Perhaps Yongnuo is doing my prior suggestion and using EF protocol communications (which effectively has no DRM) over the RF contacts. Or perhaps Canon's lawyers haven't reached out to them yet. Or they ignored Canon's lawyers and it'll take some time before we see an actual lawsuit. Samyang / Rokinon / Viltrox were selling lenses for a good while before Canon cracked down on them.