[Mikesul]

Am I the only one that hates that aperture lever sticking out the backs fo non-KAF4 lenses?

That thing just scares me during lens changes.

You are not the only one, Norm. I have hated it for a long time. It is awkward to sit it on a table when changing lenses. We are now free of that. At least for the most recent lenses.

[Kunzite]

I'm quite content with the direction Pentax is going. Yes, it's too bad that they have removed forwards compatibility, but you still have the backwards compatibility that you find important, and if, as you say, camera bodies are disposable commodities, it really shouldn't be a big deal to just buy a newer camera body that can take advantage of the new lenses.

OTOH this is likely the last evolution of the K-mount.
They can change focus motors, update communication protocols to some extent, even adding in-lens stabilisation; but there should be no need for a KAF 5

[Wheatfield]

OTOH this is likely the last evolution of the K-mount.
They can change focus motors, update communication protocols to some extent, even adding in-lens stabilisation; but there should be no need for a KAF 5

The EOS mount is still going strong at 33 years of age, and my understanding is their new mount is more of a technical evolution than revolution with full compatibility for now with an adaptor. In all that time, I never heard of any big problems with the lens mount part of the EOS system. It seems pretty reliable. I'm sure it's been updated with every new lens release too.
I don't see why this emulation of the K mount won't carry the brand well into the future, presuming there is a future.

[Serkevan]

OTOH this is likely the last evolution of the K-mount.
They can change focus motors, update communication protocols to some extent, even adding in-lens stabilisation; but there should be no need for a KAF 5

Yeah, KAF4 lenses are basically the same technology base as the rest of manufacturers (although Fuji has mechanical aperture too). Nikon followed the same path with G lenses some time ago, Sony's E mount has been the same for about 10 years now, same for Oly and their 43/M43 mounts - only one iteration for them. Canon hasn't changed the mount since the first EF camera either I think.

[Mistral75]

OTOH this is likely the last evolution of the K-mount.
They can change focus motors, update communication protocols to some extent, even adding in-lens stabilisation; but there should be no need for a KAF 5

I'm not so sure. Faster and more precise autofocus could need much faster data transfer between lens and camera and therefore hardware changes. When going from the EF mount to the RF mount Canon used more pins in the new mount and drastically increased the data throughput to increase the performance available today and pave the way for tomorrow. They did it while maintaining backwards compatibility: R cameras can use the EF communication protocol, therefore adapting EF lenses on R cameras is possible (and will maintain the lenses' performance).

On a side note: Canon haven't disclosed the RF communication protocol. If some third parties are already offering autofocus RF lenses, it's because these lenses use the EF communication protocol, which said third third parties have reverse-engineered years ago.

[Serkevan]

The EOS mount is still going strong at 33 years of age, and my understanding is their new mount is more of a technical evolution than revolution with full compatibility for now with an adaptor. In all that time, I never heard of any big problems with the lens mount part of the EOS system. It seems pretty reliable. I'm sure it's been updated with every new lens release too.
I don't see why this emulation of the K mount won't carry the brand well into the future, presuming there is a future.

AFAIK the mirrorless mount is basically the EF mount with a couple extra pins for additional lens information. Something about having more bandwidth to make AF faster and more accurate (or so Canon claims - not sure how much faster RF lenses are compared to adapted EF).

---------- Post added 03-26-20 at 04:41 PM ----------

I'm not so sure. Faster and more precise autofocus could need much faster data transfer between lens and camera and therefore hardware changes. When going from the EF mount to the RF mount Canon used more pins in the new mount and drastically increased the data throughput to increase the performance available today and pave the way for tomorrow. They did it while maintaining backwards compatibility: R cameras can use the EF communication protocol, therefore adapting EF lenses on R cameras is possible (and will maintain the lenses' performance).

On a side note: Canon haven't disclosed the RF communication protocol. If some third parties are already offering autofocus RF lenses, it's because these lenses use the EF communication protocol, which said third third parties have reverse-engineered years ago.

I know the question is moot for Canon since you obviously cannot use RF lenses on EF bodies, but in the case of a hypothetical KAF5 it might be interesting: would there be any reason for such a "KAF5" lens to not work on a KAF4-ready DSLR? I would assume that it would just work without the extra data, but of course it would depend on implementation...

[Wheatfield]

AFAIK the mirrorless mount is basically the EF mount with a couple extra pins for additional lens information. Something about having more bandwidth to make AF faster and more accurate (or so Canon claims - not sure how much faster RF lenses are compared to adapted EF).

That was my thought as well. The nice thing about the EF mount lenses is the upgrades built into every new lens. I doubt if the difference in AF speed will be all that appreciable. My understanding is that EF lenses are really quick to focus.
Unless they throttle them on the RF mount, which would make EF owners stay away in droves, I am sure.

---------- Post added Mar 26th, 2020 at 05:50 PM ----------

I know the question is moot for Canon since you obviously cannot use RF lenses on EF bodies, but in the case of a hypothetical KAF5 it might be interesting: would there be any reason for such a "KAF5" lens to not work on a KAF4-ready DSLR? I would assume that it would just work without the extra data, but of course it would depend on implementation...

The KAF5 will be the short flange distance mirrorless mount.

[Serkevan]

That was my thought as well. The nice thing about the EF mount lenses is the upgrades built into every new lens. I doubt if the difference in AF speed will be all that appreciable. My understanding is that EF lenses are really quick to focus.
Unless they throttle them on the RF mount, which would make EF owners stay away in droves, I am sure.

They are - when the girlfriend was dancing (I am unfortunately physically incapable, I freeze up ) during a wedding, I was using her 70-200/4 L tele zoom to take pictures*. On a 6D Mk1, so 8 year old camera with 20 year old lens... every bit as quick as my D FA 28-105, although in low light it *did* hesitate and slightly misfocused quite often...

About RF, I meant that, although EF lenses are equally good on either EF or RF bodies, in theory RF lenses should be even better due to the extra data transfer. I haven't had any experience however.

*The K-1 was in the room, we decided to bring out only one camera that evening

[MikeyBugs95]

AFAIK the mirrorless mount is basically the EF mount with a couple extra pins for additional lens information. Something about having more bandwidth to make AF faster and more accurate (or so Canon claims - not sure how much faster RF lenses are compared to adapted EF).

---------- Post added 03-26-20 at 04:41 PM ----------



I know the question is moot for Canon since you obviously cannot use RF lenses on EF bodies, but in the case of a hypothetical KAF5 it might be interesting: would there be any reason for such a "KAF5" lens to not work on a KAF4-ready DSLR? I would assume that it would just work without the extra data, but of course it would depend on implementation...

I think that if there was a need for a "KAF5" mount with increased data transfer rates, and it were to be a DSLR mount, I think an added pin would be best. Then cameras would be backwards compatible with KAF4 lenses and back and older lenses forwards compatible with the KAF5 mount.

[Serkevan]

I think that if there was a need for a "KAF5" mount with increased data transfer rates, and it were to be a DSLR mount, I think an added pin would be best. Then cameras would be backwards compatible with KAF4 lenses and back and older lenses forwards compatible with the KAF5 mount.

That's what I thought as well. On the other hand, the K-mount (or any long-standing standard in anything, really) is a blessing and a curse in that regard, since some improvements can only be bandaids instead of complete overhauls. Okay, it's mostly a curse for engineers who have/had to spend longer adding stuff to the mount, for us it's been largely a blessing, with that treasure trove of K-mount classics

[thibs]

What do you mean using the words „it all“?

The Lens. For me, strictly personnal.
I won't have a focus by wire crap, ever.

Only idiots never change their mind, I know

[zzeitg]

I expected confirmation bias.

So you got what you expected. Everything's all right.

[Kunzite]

I'm not so sure. Faster and more precise autofocus could need much faster data transfer between lens and camera and therefore hardware changes. When going from the EF mount to the RF mount Canon used more pins in the new mount and drastically increased the data throughput to increase the performance available today and pave the way for tomorrow. They did it while maintaining backwards compatibility: R cameras can use the EF communication protocol, therefore adapting EF lenses on R cameras is possible (and will maintain the lenses' performance).

On a side note: Canon haven't disclosed the RF communication protocol. If some third parties are already offering autofocus RF lenses, it's because these lenses use the EF communication protocol, which said third third parties have reverse-engineered years ago.

I said "there should be no need for a KAF 5" so where exactly did you get the impression I'm absolutely sure?
I'm not completely discounting the possibility of another mount "version". I only say it is not needed to offer all functionalities a lens will likely have, or at least what the competition has (including function buttons).
Not all "hardware changes" have to be breaking changes, and your EOS example proves that. Even adding a new communication pin might not require breaking compatibility (and a new mount "version").

[Kunzite]

The KAF5 will be the short flange distance mirrorless mount.

Nope; a short flange distance mirrorless mount, by definition, will not be K-mount.

[Mistral75]

(...)
Not all "hardware changes" have to be breaking changes, and your EOS example proves that. Even adding a new communication pin might not require breaking compatibility (and a new mount "version").

Adding a new pin or several would not necessarily require breaking compatibility, I'll second that, but it would definitely be a new mount version. Think of KAF2 for instance.