Sony, give us FE mount adapters with all focus modes. Like Sigma does.

venice

Senior Member
Messages
1,973
Reaction score
283
Location
US
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
 
Are you really really sure you are going to have full af function on your canon 100-400mm???
 
Who said canon lens can get full AF function in MC-11?
 
All native focus functions when using Sigma mount lenses... not Canon mount.

Unless i've missed some new verified info.
 
All native focus functions when using Sigma mount lenses... not Canon mount.

Unless i've missed some new verified info.
 
All native focus functions when using Sigma mount lenses... not Canon mount.

Unless i've missed some new verified info.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
 
Sigma mount lenses are compatible with Canon cameras, i assume, otherwise, why would Canon owners buy them.

And MC-ll is compatible with Canon mount sigma lenses.

It could well be that MC-11 is compatible with some Canon lenses - we won't know until they are shipped and someone tests them. As Rishi says in the DPreview article - a game changer.

I can see why Sigma claims no compatibility with Canon lenses - they don't want to spend time with customers on every little compatibility problem for Canon lenses for which they get no money. But they may be compatible- to some degree.

Just be patient and lets see what develops. The cool thing is that the MC-ll is firmware upgrade-able and will help put pressure to compress MB prices. $250 for the MC-ll sounds good to me. I don't mind buying more than one adapter if i know any new adapter will work on at least one desirable lens.

--
Phil B
Sony A7RII, Pentax K-3, Sony Nex 6, 5n, Epson 3880
 
Last edited:
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
The main difference is probably in how CDAF is combined with PDAF. To find the eye and follow it wherever it moves in the frame requires access to the image pixels, which cover the whole frame, which means CDAF. Sigma, at least, has made statements in the past about how the motors in their lenses are (or were, at least) designed for PDAF positioning, and weren't optimal for CDAF. It sounds like Sigma has figured out how to do fast CDAF with their adapter and newer lenses, while Sony is (I think) doing PDAF for fast AF with A-mount lenses, and relies on fast (native E-mount) CDAF for eye-tracking.

That said, it also seems feasible, but probably not simple, for Sony to implement hybrid AF through the adapters so they can use PDAF for initial positioning, and CDAF for fine-tuning and tracking, or, if there are enough OSPDAF points, use the image data to select the nearest PDAF point as the eye moves across the frame.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
I have the SAM 85 f2.8 lens with the LAEA3 adapter. In continuous AF and Wide mode, it will lock onto a focus area, and as i move the frame around the tiny little focus boxes will follow that focus area anywhere within the 399 focus points. Its a screw drive lens, or at least sounds like it to me, but it chugs boldly away. Works in either FF or crop mode - pretty cool for a $300 new lens. Even in center mode or flexible mode, it will continue to sample the focus area until i shoot.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
You forget the part where the A7R2 is being fed focus position prediction data by the native lens while the A-mount lens simply cannot, allowing the A7R2 to decide when to turn the lens around for the CDAF phase.

That's one of the MAJOR differences between EA1/EA3 emulation and native E mount lenses. Native lenses provide both current and predicted (at 1/60 second from the current time) motor positions, EA1/EA3 emulation only provides current. Native E-mount lenses also support absolute positioning (go to position X as opposed to move forward/back Y steps), A-mount lenses, as far as I can tell, don't. These are all things that help a CDAF implementation operate with far less hunting/overshoot.

If A-mount lenses were capable of precise and rapid CDAF like native lenses were, CDAF on EA1/EA3 adapters wouldn't suck so much and they wouldn't be using the "chunk chunk chunk" multi-stepping phase of the EA1/EA3 CDAF cycle.

I've also seen zero evidence of focus motor speed control in EA1/EA3 adapters, implying A-mount lacks it. If Sony doesn't have focus motor speed control in A-mount, then attempting native lens emulation with A-mount lenses would be similar to the Techart III - works with some lenses (with excessive hunting) and totally fails with anything with a fast AF motor.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
I have the SAM 85 f2.8 lens with the LAEA3 adapter. In continuous AF and Wide mode, it will lock onto a focus area, and as i move the frame around the tiny little focus boxes will follow that focus area anywhere within the 399 focus points. Its a screw drive lens, or at least sounds like it to me, but it chugs boldly away. Works in either FF or crop mode - pretty cool for a $300 new lens. Even in center mode or flexible mode, it will continue to sample the focus area until i shoot.
 
It seems quite commonplace to hear people say that they plan on using Canon glass with the MC-11 and a7r II; however, I have to point out that those that say so are wrong — Canon lenses won't work with the MC-11, or rather won't offer all the features of the MC-11. In a Sigma-made video, it details how in order to make a Sigma lens behave like a native FE lens in terms of AF performance, a custom lens firmware is stored on the adapter and overrides the built in firmware of the lens in use. So far only 15 Sigma lenses are supported (as in have their own custom firmwares available for the adapter) and they are in Sigma SA mount or Canon EF mount; however, this does not mean that Canon EF lenses can be used with the adapter — only the 15 Sigma lenses, in either SA mount or EF mount, will be fully compatible with the MC-11 — as Canon will likely not allow custom firmware to override their lenses. For what is known right now, using an unsupported lens might just result in similar AF performance as using any other adapter, such as the Metabones or Commlite adapters, or might even be a bit worse. Something like the Canon 100-400 will not work well...
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
I have the SAM 85 f2.8 lens with the LAEA3 adapter. In continuous AF and Wide mode, it will lock onto a focus area, and as i move the frame around the tiny little focus boxes will follow that focus area anywhere within the 399 focus points. Its a screw drive lens, or at least sounds like it to me, but it chugs boldly away. Works in either FF or crop mode - pretty cool for a $300 new lens. Even in center mode or flexible mode, it will continue to sample the focus area until i shoot.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
You forget the part where the A7R2 is being fed focus position prediction data by the native lens while the A-mount lens simply cannot, allowing the A7R2 to decide when to turn the lens around for the CDAF phase.

That's one of the MAJOR differences between EA1/EA3 emulation and native E mount lenses. Native lenses provide both current and predicted (at 1/60 second from the current time) motor positions, EA1/EA3 emulation only provides current. Native E-mount lenses also support absolute positioning (go to position X as opposed to move forward/back Y steps), A-mount lenses, as far as I can tell, don't. These are all things that help a CDAF implementation operate with far less hunting/overshoot.

If A-mount lenses were capable of precise and rapid CDAF like native lenses were, CDAF on EA1/EA3 adapters wouldn't suck so much and they wouldn't be using the "chunk chunk chunk" multi-stepping phase of the EA1/EA3 CDAF cycle.

I've also seen zero evidence of focus motor speed control in EA1/EA3 adapters, implying A-mount lacks it. If Sony doesn't have focus motor speed control in A-mount, then attempting native lens emulation with A-mount lenses would be similar to the Techart III - works with some lenses (with excessive hunting) and totally fails with anything with a fast AF motor.
The A7r2 has only 25 CDAF points. It has 399 OSPDAF points. I expect Eye Detect and the other missing focus modes will work just fine with a lens like the 70-400G2.

Sony, give us the adapter we deserve or a firmware update.

-Bill
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
I have the SAM 85 f2.8 lens with the LAEA3 adapter. In continuous AF and Wide mode, it will lock onto a focus area, and as i move the frame around the tiny little focus boxes will follow that focus area anywhere within the 399 focus points. Its a screw drive lens, or at least sounds like it to me, but it chugs boldly away. Works in either FF or crop mode - pretty cool for a $300 new lens. Even in center mode or flexible mode, it will continue to sample the focus area until i shoot.
 
Apparently, Sigma has made a FE mount adapter for Sigma and Canon mount glass that allows the use of all the native focus modes, including Eye Detect.

So now, I am considering selling my Sony 70-400G2 and laea3 and getting a Sigma MC-11 and Canon 100-400 so I can have full function on my A7r2.

Sony, that is ridiculous.

Give us a new converter or, better yet, a firmware update that unlocks ALL of the powerful focus modes on the A7r2 for A mount lenses.

-Bill
A-mount is simply too old. Not going to happen. Sony was barely able to shoehorn electronic focusing into the mount in the 2000s and still doesn't have electronic aperture, while Canon has had electronic focus and aperture since 1987 and has been constantly improving the mount comm protocols internally since then.

Sony A-mount's protocol is too simplistic for Sony to extend it without breaking compatibility with all existing bodies, since there is no command/response mechanism. Lenses read out their ROM/status data without any command. Canon lenses, on the other hand, have a command/response messaging scheme for everything, which allows Canon to easily add new commands without breaking compatibility of the old ones.

Of specific note - Recent Canon lenses themselves have been designed to be fully capable of OSPDAF in addition to legacy PDAF - https://www.usa.canon.com/CUSA/assets/app/images/cameras/eos/DAF/compatible_lens_chart.pdf (note, Canon seems to have slacked on updating this list so lenses newer than 2014 aren't on the list despite being compatible according to anyone who has used them). All of Sony's A-mount lenses were all designed solely to operate with legacy PDAF.

Beyond this - Sigma is only offering full native capability with a limited subset of their own lenses, not all EF-mount lenses.

I personally believe it should be possible to adapt anything on Canon's DPAF compat list to operate like a native lens, and Techart came close to this - if Techart fixed their focus motor speed control support the adapter would've crushed everything else on the market back in September. Fortunately for Metabones and Sigma, Techart don't support their products more than a month after release.
My A7r2 can tell my A mount lens where to focus.

It can also find an eye, within an image, and determine how far away the eye is.

So, I think it can find an eye, figure out how far away the eye is, then tell the A mount lens to focus there. I don't really see why it is any different in principle from any other focus mode unless there is some critical piece of feedback missing from the lens.

If that feedback is missing then why can the A7r2 tell the A mount lens where to focus at all? What is fundamentally different about focusing on an eye vs focusing on a patch in the center of the screen other than the camera processing power needed to identify and follow an eye?

I don't see it. Sounds like Sony has made a choice to hold back features to me. It wouldn't be the first time a camera company has done that.

-Bill
You forget the part where the A7R2 is being fed focus position prediction data by the native lens while the A-mount lens simply cannot, allowing the A7R2 to decide when to turn the lens around for the CDAF phase.

That's one of the MAJOR differences between EA1/EA3 emulation and native E mount lenses. Native lenses provide both current and predicted (at 1/60 second from the current time) motor positions, EA1/EA3 emulation only provides current. Native E-mount lenses also support absolute positioning (go to position X as opposed to move forward/back Y steps), A-mount lenses, as far as I can tell, don't. These are all things that help a CDAF implementation operate with far less hunting/overshoot.

If A-mount lenses were capable of precise and rapid CDAF like native lenses were, CDAF on EA1/EA3 adapters wouldn't suck so much and they wouldn't be using the "chunk chunk chunk" multi-stepping phase of the EA1/EA3 CDAF cycle.

I've also seen zero evidence of focus motor speed control in EA1/EA3 adapters, implying A-mount lacks it. If Sony doesn't have focus motor speed control in A-mount, then attempting native lens emulation with A-mount lenses would be similar to the Techart III - works with some lenses (with excessive hunting) and totally fails with anything with a fast AF motor.
The A7r2 has only 25 CDAF points. It has 399 OSPDAF points. I expect Eye Detect and the other missing focus modes will work just fine with a lens like the 70-400G2.

Sony, give us the adapter we deserve or a firmware update.

-Bill
These are not equivalent. The 399 points are fixed points on the sensor, and PDAF can only happen for areas of the image that correspond to these points. The 25 points are how many areas the AF logic will evaluate when doing CDAF. Those 25 points can come from anywhere on the sensor and are not limited to specific physical locations.
 
I want cupcakes too:)

Sigma's converter gives full function with (some) Sigma lenses. Not Canon lenses. It makes sense, so that Sigma will probably sell a few lenses more.

It does not make sense for Sony to make an adapter that would make people buy Canon lenses, rather than Sony lenses. For those lenses that are missing from the Sony line, there are already adapters that work fine.
 
I want cupcakes too:)

Sigma's converter gives full function with (some) Sigma lenses. Not Canon lenses. It makes sense, so that Sigma will probably sell a few lenses more.

It does not make sense for Sony to make an adapter that would make people buy Canon lenses, rather than Sony lenses. For those lenses that are missing from the Sony line, there are already adapters that work fine.
 

Keyboard shortcuts

Back
Top