Canon Full Frame Mirrorless is Definitely Coming, and The Wait Won't Be as Long as We Thought

[Talys]

I'm sure Canon execs have stayed up late at night trying to figure out whether to jump into mirrorless with one, or two feet. I love my OVF, but it's undeniable that mirrorless is the future. A dual release would be a good way to test what the market truly wants, but not sure how viable that really is. Although I will say I personally don't want a mirrorless body any smaller than the current 5D, so if they just remove the mirrorbox to squeeze in an EVF, leaving everything else the same, it would be perfect for me.

I don't think that would prove much. I think that there is more demand for a mirrorless release, simply because most Canon owners don't own a full frame mirrorless (ie Sony A7/A9). DSLRs are pretty mature. I mean, I know lots of guys with 5DMk3, because they don't see enough of an advantage in the Mk4 to be worth a $3000 spend.

But once someone has both a mirrorless and DSLR, which will they prefer? I think that's the real test. And how many people who buy a FF Mirrorless will go on the MILC treadmill for a while, like they did with DSLRs until those matured?

And of course, 10-15 years after a FF mirrorless launch, when someone is using a FF, will they still prefer DSLR for some (common) things.

 

[ahsanford]

I will say I personally don't want a mirrorless body any smaller than the current 5D, so if they just remove the mirrorbox to squeeze in an EVF, leaving everything else the same, it would be perfect for me.

I don't think that would prove much.

Upsides to this idea:

• The 'EF-X' lenses that nest inside of the EF mount could lead to some smaller aggregate lens + body sizes on wide FLs at modest apertures
• Still could be quicker fps-wise than it's SLR counterpart (no mirror box slowing things down)
• Manual lens use could be awesome (for all mirrorless shapes/sizes, in fairness)
• Great for controls and seamlessness with our current daily drivers (6D, 5D, 7D, etc.)
• No adaptors and no mandate to buy new lenses to enjoy the new platform; EF is good to go on day one
• Great for handling bigger lenses
• Potential to share 5D accessories, L-plates, batteries, etc.

Downsides to this idea:

• Not as small as it could be (for the small-size-insistent crowd)
• Full EF mount = can't adapt other lenses
• The loud braying of the internet, eager to bash anything that isn't simultaneously very small and very well spec'd. They'd have a field day with this.

I think that there is more demand for a mirrorless release, simply because most Canon owners don't own a full frame mirrorless (ie Sony A7/A9). DSLRs are pretty mature. I mean, I know lots of guys with 5DMk3, because they don't see enough of an advantage in the Mk4 to be worth a $3000 spend.

^^^ This is 100% me ^^^

Tell me I can (a) use manual focus large aperture glass handheld or (b) amplify light in a dark shooting environment, mirrorless absolutely would get my attention in a 5D form factor.

- A

 

[JohanCruyff]

It needs to be a professional body (5D equivalent ruggedness and features) and it needs to be EF mount. End of story. I could see it being a flagship 1D mirrorless, but that'll put it out of reach for most Canon users. Lower than where the 5D is and 1. current Canon professionals won't leave their 5D for it and 2. it won't be competitive with Sony.

Who said anything about pleasing professionals with this first offering? I see a next-to-zero chance of that.

I think the first FF offering will be a pricey toy for enthusiasts and affluent folks with disposable income -- a 6D2 feature set with an EVF (and possibly a better sensor). But as much as I want a mirrorless 5D, I'd be stunned if we got that right out of the gate.

- A

I disagree. The fact that Canon EoL (and CPS members) were surveyed about mirrorless and are now field testing, along with the report that Canon "wants to get it right," indicates a professional body. It would otherwise be an unusual interest in what professionals want for an enthusiast product. Canon also has a strong history of releasing professional bodies first and then trickling/watering down those features into more affordable options:

1D then 5D
5D then 6D
C300 then C100
etc.

Furthermore, I think Canon feels the heat on the FF mirrorless front. Sony went from having one professional FF mirrorless with the a9 to having two after the third iteration of the a7 line. If I were Canon I'd feel the need to compete with at least one of these with my first entry into this segment. A mirrorless 6d Mk II won't. Canon needs to flex some muscle and renew excitement among their user base. A prosumer offering won't.

TL;DR: Canon will announce the 5D V, their first full-frame mirrorless camera in Q1 2019 with availability summer 2019 You heard it here first.

I'm with ahsanford on this one.

I think that the long term goal (Tokyo 2020) is to have a professional (1D) product, and therefore, professionals and EoL are important to set Canon on the right path.

But the 2019 goal will be something closer to a enthusiast / portable professional, 5D price point product.

+1: 6D or 5D segment in 2019, and 1D segment in 2020.
Where do we have to sign?
___________________________

What can we suggest Canon about model's naming?
If M5 Mark II and M6 Mark II are going to be replaced by full frame versions (M5 Mark II and M6 Mark II), then the "pro" model will be called M1. Otherwise?

 

[djkraq]

What I want in a Canon FF Mirrorless:

A new DP RAW similar to Pixel Shift or similar
DPAF
8-10 fps RAW
Eye-AF would be a plus
Focus Peaking (for manual lenses)
Ergonomics like a standard 5D DSLR but with larger battery
A new mount (allowing for more AF Points to sensor)
An EF Mount Adapter (for our old lenses)
5 Axis IBIS

C-Log for Video
4k 30p 4:2:0
1080p up to 120fps
1080p 60fps 4:2:2 8-bit
1080p 30fps 4:2:2 10-bit

I actually like my cameras larger as they are more ergonomic to me and have a professional look to my clients. They tend to think that since my camera is larger, it will produce better photos (yeah, right lol). But in all seriousness, I would rather them keep it large, have better weather sealing and durability.

 

[Deleted member 379512]

What I want in a Canon FF Mirrorless:

A new DP RAW similar to Pixel Shift or similar
DPAF
8-10 fps RAW
Eye-AF would be a plus
Focus Peaking (for manual lenses)
Ergonomics like a standard 5D DSLR but with larger battery
A new mount (allowing for more AF Points to sensor)
An EF Mount Adapter (for our old lenses)
5 Axis IBIS

C-Log for Video
4k 30p 4:2:0
1080p up to 120fps
1080p 60fps 4:2:2 8-bit
1080p 30fps 4:2:2 10-bit

I actually like my cameras larger as they are more ergonomic to me and have a professional look to my clients. They tend to think that since my camera is larger, it will produce better photos (yeah, right lol). But in all seriousness, I would rather them keep it large, have better weather sealing and durability.

This sounds pretty much like a sony a7r3 in a 5d body - which is pretty much what I would love to see also, along with a tilt/flip screen. Another surprisingly useful feature would be usb charging, with 2x usb ports.

 

[sigh]

What I want in a Canon FF Mirrorless:

A new DP RAW similar to Pixel Shift or similar
DPAF
8-10 fps RAW
Eye-AF would be a plus
Focus Peaking (for manual lenses)
Ergonomics like a standard 5D DSLR but with larger battery
A new mount (allowing for more AF Points to sensor)
An EF Mount Adapter (for our old lenses)
5 Axis IBIS

C-Log for Video
4k 30p 4:2:0
1080p up to 120fps
1080p 60fps 4:2:2 8-bit
1080p 30fps 4:2:2 10-bit

I actually like my cameras larger as they are more ergonomic to me and have a professional look to my clients. They tend to think that since my camera is larger, it will produce better photos (yeah, right lol). But in all seriousness, I would rather them keep it large, have better weather sealing and durability.

This sounds pretty much like a sony a7r3 in a 5d body - which is pretty much what I would love to see also, along with a tilt/flip screen. Another surprisingly useful feature would be usb charging, with 2x usb ports.

I'd also be happy with that. I'm not too bothered about a full tilt/flip screen like the 80D or M50 if we get DPAF.

On the video side, I doubt we'll get C-Log first time round (maybe as a paid upgrade?), or any form of 10-bit video, but I would like there to be no crop in 4k. I used the 5D4 over a long weekend (I currently have a 5D3 and A7R3) and the crop in 4k was not to my liking. That's just my opinion, I have no idea what pros, EoL or Canon's marketing says. I didn't have too many concerns about the MJPEG codec other than space requirements and that, even on my beefy machine, I had to transcode the footage to edit it smoothly.

 

[3kramd5]

A new mount (allowing for more AF Points to sensor).

What do you mean by this? You don’t need additional electrical contacts when you increase the AF sensor count (which with DPAF is pretty much a linear relationship with resolution). The logic is done in the camera, the mount only has to send power and instructions to the lens accordingly.

 

[Duct_Taper]

My L lenses work just fine on my M5. Not as snappy as the STM glass but accurate and decently fast.

Your USM lenses on the adaptor are slower than STM on the native EF-M mount?!

Unless you are shooting some famously slow USM L lenses, like the 85 f/1.2L II or the 180L Macro, I would call that a blistering indictment for adapting lenses. I say that because modern/contemporary USM should mop the floor with STM in virtually any comparison w.r.t. stills focusing speed.

Other folks who adapt on EF-M today -- have you had the same experience?

- A

I can't speak for adapted EF-M performance but based on EF-S experience you have to be a bit careful with generalizations w.r.t STM focusing speed... it's very lens dependent. In my experience the 24, 40, and 50 are all a bit 'leisurely' but the 18-135 and the 55-250 are very fast. I'd actually say my 55-250 STM is as fast or faster than my 70-210 f/3.5-4.5 USM, maybe even my 100 f/2 USM.

 

[Talys]

My L lenses work just fine on my M5. Not as snappy as the STM glass but accurate and decently fast.

Your USM lenses on the adaptor are slower than STM on the native EF-M mount?!

Unless you are shooting some famously slow USM L lenses, like the 85 f/1.2L II or the 180L Macro, I would call that a blistering indictment for adapting lenses. I say that because modern/contemporary USM should mop the floor with STM in virtually any comparison w.r.t. stills focusing speed.

Other folks who adapt on EF-M today -- have you had the same experience?

- A

I can't speak for adapted EF-M performance but based on EF-S experience you have to be a bit careful with generalizations w.r.t STM focusing speed... it's very lens dependent. In my experience the 24, 40, and 50 are all a bit 'leisurely' but the 18-135 and the 55-250 are very fast. I'd actually say my 55-250 STM is as fast or faster than my 70-210 f/3.5-4.5 USM, maybe even my 100 f/2 USM.

The fastest autofocusing lens I have ever seen is the focus by wire EFS 18-135 USM (nano), which is to say, it's so fast that I don't even know that it's autofocused; plus, it's totally quiet.

However, I still wouldn't choose it over ring USM for nearly all purposes.

 

[ahsanford]

I can't speak for adapted EF-M performance but based on EF-S experience you have to be a bit careful with generalizations w.r.t STM focusing speed... it's very lens dependent. In my experience the 24, 40, and 50 are all a bit 'leisurely' but the 18-135 and the 55-250 are very fast. I'd actually say my 55-250 STM is as fast or faster than my 70-210 f/3.5-4.5 USM, maybe even my 100 f/2 USM.

Appreciate that STM has improved, sure. Modern STM being on the level with 20+ year old non-L USM is plausible, I guess.

But slclick said his L lenses were slower than STM. That's a different animal, right?

slclick, can you tell us which L lenses you've seen slower-than-STM performance with on your M5? If it's a modern ring USM setup that we normally don't associate with having slow focus, either the new STM is blisteringly fast or the adaptor is slowing down the AF.

- A

 

[ahsanford]

The fastest autofocusing lens I have ever seen is the focus by wire EFS 18-135 USM (nano), which is to say, it's so fast that I don't even know that it's autofocused; plus, it's totally quiet.

However, I still wouldn't choose it over ring USM for nearly all purposes.

I prefer my ring USM glass, but yes, the two nano lenses' AF seems impressive.

From LensTip (one of the few groups that looks at AF speed) on the other Nano USM lens:

"What’s more the autofocus performance is sensationally fast. Running through the whole distance scale and confirming the focus at the shorter end of the focal lengths spectrum takes 0.1-0.2 of a second; for the longer focal lengths the process is by 0.1-0.2 of a second longer.

To be honest I am a bit surprised that, for the first time, such technology appears in completely amateur constructions. The working culture of the autofocus, its noiselessness, 100% accuracy, and superior speed put to shame even some professional lenses. A round of applause for Canon!"

And a gold-standard-quick ring USM zoom?

"On the EOS 1Ds Mark III, even in the full working range, the running though the scale, no matter from what end, takes about 0.3-0.4 of a second. Using the limited 2.5 meters to infinity range you can shorten that time by next 0.1 of a second. These are results worth of a reliable journalistic instrument."

So the Nano was:

70mm: 0.1 - 0.2s
300mm 0.2 - 0.4s

and the 70-200L was (at either end of the FL):

Not delimited: 0.3 - 0.4s
Delimited: 0.2 - 0.3s

Which is quite impressive for the Nano (considering the major cost delta). These comparisons are never perfect -- it appears the former was run on the 5D3 and the latter on the 1DS3, but still: not bad, Nano.

- A

 

[Loswr]

Also, don't forget the EF mount is 30 years old.

There are many reasons to replace the mount.

A more intelligent modern mount might include the ability to have a fast data link between the lens and the body. Imagine, for example, a situation where a specific lens as a coprocessor within it to help deal with specific calculations it needs (for example, for an enhanced IS system), or with a macro light embedded in the lens that, because of the improved communication system is able to communicate via ETTL with the camera.

Autofocus tilt-shift lenses and the ability for tilt-shift lenses and for lenses with apodization filters to communicate and record those settings into the EXIF data also become possible.

Also, more efficient power usage, better weatherproofing, all of these can be put into a new design.

Right now we're stuck with an interface designed at the time the Commodore 64 was the most popular home computer.

How do you know those things aren't possible with the current interface? Current lenses have processors, current lenses have firmware that can be updated via the camera body.

Are you suggesting the reason tilt-shift lenses don't have AF is that the lens interface is too slow? That the slow interface —and not the lack of lens encoders for things like TS settings and Macro Lite status— is the reason those things aren't recorded in the EXIF? I know people like to make stuff up to bash Canon, but it should at least seem credible.

Oh, and just to drive the point home, let's look at what Canon has to say on the matter:

[quote author=Canon]
Since the first EOS cameras and EF lenses in 1987, a number of new technologies have been introduced into Canon's EOS system. As new camera and/or lens features have been developed, this has added to the amount of items communicated between body and lens. Canon has been able to do this and maintain practically total compatibility going back to the earliest EOS bodies and lenses. Most importantly, the lens mount and the gold contacts have not changed one bit! Features introduced since 1987, which have altered the way data, is communicated include:

Predictive AI Servo AF (focus tracking on moving subjects)
Micro-USMs (extremely small Ultrasonic focusing motors)
Tilt-shift lenses with Automatic Diaphragm operation
Multiple-point AF systems, from 3 points up to 45
Image Stabilization
E-TTL flash (which relies on instant analysis of a pre-flash)
Wireless E-TTL flash

Furthermore, as new cameras have been developed, new and faster communication methods have been introduced to give us faster autofocus, more precise light metering, faster shooting speeds (up to 9 fps with autofocus on the EOS-1v, for instance), and of course the new features that digital SLRs bring to the table. Data communication has accordingly changed over time, and occasionally a new camera will be launched that modifies how data is transmitted between body and lens.
[/quote]

So again, please explain how the '30 year old communication' is a limiting factor of the EF mount...

 

[Talys]

The fastest autofocusing lens I have ever seen is the focus by wire EFS 18-135 USM (nano), which is to say, it's so fast that I don't even know that it's autofocused; plus, it's totally quiet.

However, I still wouldn't choose it over ring USM for nearly all purposes.

I prefer my ring USM glass, but yes, the two nano lenses' AF seems impressive.

From LensTip (one of the few groups that looks at AF speed) on the other Nano USM lens:

"What’s more the autofocus performance is sensationally fast. Running through the whole distance scale and confirming the focus at the shorter end of the focal lengths spectrum takes 0.1-0.2 of a second; for the longer focal lengths the process is by 0.1-0.2 of a second longer.

To be honest I am a bit surprised that, for the first time, such technology appears in completely amateur constructions. The working culture of the autofocus, its noiselessness, 100% accuracy, and superior speed put to shame even some professional lenses. A round of applause for Canon!"

And a gold-standard-quick ring USM zoom?

"On the EOS 1Ds Mark III, even in the full working range, the running though the scale, no matter from what end, takes about 0.3-0.4 of a second. Using the limited 2.5 meters to infinity range you can shorten that time by next 0.1 of a second. These are results worth of a reliable journalistic instrument."

So the Nano was:

70mm: 0.1 - 0.2s
300mm 0.2 - 0.4s

and the 70-200L was (at either end of the FL):

Not delimited: 0.3 - 0.4s
Delimited: 0.2 - 0.3s

Which is quite impressive for the Nano (considering the major cost delta). These comparisons are never perfect -- it appears the former was run on the 5D3 and the latter on the 1DS3, but still: not bad, Nano.

- A

I have owned both the EFS18-135 USM and the EF 70-300 USM II, and the 18-135 feels way faster (it is noticeably faster than a 24-105 or 24-70 mounted on an 80D).

There is another big difference: on the 70-300 (which I ultimately sold), autofocus was not as consistent, sometimes resulting in slight softness, while my 18-135 is very consistent. Of course, it just be my luck of the draw.

 

[ahsanford]

I have owned both the EFS18-135 USM and the EF 70-300 USM II, and the 18-135 feels way faster (it is noticeably faster than a 24-105 or 24-70 mounted on an 80D).

There is another big difference: on the 70-300 (which I ultimately sold), autofocus was not as consistent, sometimes resulting in slight softness, while my 18-135 is very consistent. Of course, it just be my luck of the draw.

Keep in mind that I'm just using these numbers to soften the blow when my EF 50mm f/1.4 IS (Ring) USM turns out to just be an EF 50mm f/1.4 (Nano) USM II. ;D

If it's quick, consistent and internal focusing, I suppose I could live with FBW.

But I'm OT here. Apologies.

- A

 

[slclick]

I can't speak for adapted EF-M performance but based on EF-S experience you have to be a bit careful with generalizations w.r.t STM focusing speed... it's very lens dependent. In my experience the 24, 40, and 50 are all a bit 'leisurely' but the 18-135 and the 55-250 are very fast. I'd actually say my 55-250 STM is as fast or faster than my 70-210 f/3.5-4.5 USM, maybe even my 100 f/2 USM.

Appreciate that STM has improved, sure. Modern STM being on the level with 20+ year old non-L USM is plausible, I guess.

But slclick said his L lenses were slower than STM. That's a different animal, right?

slclick, can you tell us which L lenses you've seen slower-than-STM performance with on your M5? If it's a modern ring USM setup that we normally don't associate with having slow focus, either the new STM is blisteringly fast or the adaptor is slowing down the AF.

- A

Perhaps I phrased that incorrectly. L lenses (24-70 2.8 Mk2, 135L, 16-35 f/4L) slower focusing on adapted M5 than on native FF but in no means hunting or inaccurate. Relative to the 50 1.8 STM and 40 2.8 adapted or the native EF-M, I'll have to get back to you on that. The 50 1.8 STM does very well on the M5.... the 40, not so well but still no slouch. The 50 is just a great pairing. I would say the 50 adapted is as fast as the EF-M 22. My findings on the 40 mirror Dustin's I have found. But it's only AF speed, everything else is fine. I will get around to testing the adapted L vs STM's on the M5 one of these days and tell you what I thought.

 

[ahsanford]

Perhaps I phrased that incorrectly. L lenses (24-70 2.8 Mk2, 135L, 16-35 f/4L) slower focusing on adapted M5 than on native FF but in no means hunting or inaccurate. Relative to the 50 1.8 STM and 40 2.8 adapted or the native EF-M, I'll have to get back to you on that. The 50 1.8 STM does very well on the M5.... the 40, not so well but still no slouch. The 50 is just a great pairing. I would say the 50 adapted is as fast as the EF-M 22. My findings on the 40 mirror Dustin's I have found. But it's only AF speed, everything else is fine. I will get around to testing the adapted L vs STM's on the M5 one of these days and tell you what I thought.

Thank you, this is what I want to drill down on. The thin mirrorless + adaptor camp would have us believe that there are no tradeoffs to AF performance on an adaptor, and if that hasn't been your experience, it might imply that the most discerning/picky photographer might need a true EF mount make their EF glass sing. (Could also be DPAF vs. the standard SLR AF, in fairness.)

Again: I'd like more on this from everyone, please. Those who have an M5 or M6, relatively modern FF SLR w/DPAF, some quick L lenses and an adaptor, please go to town and tell us what you find!

Ideally, we'd compare the focus speed of:

SLR standard AF setup
SLR with AF on LiveView
M5/M6 through an adaptor

- A

 

[Talys]

Perhaps I phrased that incorrectly. L lenses (24-70 2.8 Mk2, 135L, 16-35 f/4L) slower focusing on adapted M5 than on native FF but in no means hunting or inaccurate. Relative to the 50 1.8 STM and 40 2.8 adapted or the native EF-M, I'll have to get back to you on that. The 50 1.8 STM does very well on the M5.... the 40, not so well but still no slouch. The 50 is just a great pairing. I would say the 50 adapted is as fast as the EF-M 22. My findings on the 40 mirror Dustin's I have found. But it's only AF speed, everything else is fine. I will get around to testing the adapted L vs STM's on the M5 one of these days and tell you what I thought.

Thank you, this is what I want to drill down on. The thin mirrorless + adaptor camp would have us believe that there are no tradeoffs to AF performance on an adaptor, and if that hasn't been your experience, it might imply that the most discerning/picky photographer might need a true EF mount make their EF glass sing. (Could also be DPAF vs. the standard SLR AF, in fairness.)

Again: I'd like more on this from everyone, please. Those who have an M5 or M6, relatively modern FF SLR w/DPAF, some quick L lenses and an adaptor, please go to town and tell us what you find!

Ideally, we'd compare the focus speed of:

SLR standard AF setup
SLR with AF on LiveView
M5/M6 through an adaptor

- A

Or birder/action photographer. AF speed is really important if you're trying to catch something in motion, and even a small drop in autofocus performance can mean the difference between a keeper and a culled shot. There is no way someone for whom AF speed is important enough a reason to upgrade or choose a lens would then be satisfied with a system that downgraded AF speed

On the other hand, I would not put it past Canon to improve the electronics in an adapter mount aimed to professionals to provide 100% efficiency for EF lenses.

We should also consider that we may not be comparing apples to apples, because most people will comparing the AF speed through the lens on a 5D, for example, to DPAF through the sensor on an M5. I mean, how many photographers use live view on their DSLR for action shots?

I'd love to hear more about this, too!

 

[slclick]

Perhaps I phrased that incorrectly. L lenses (24-70 2.8 Mk2, 135L, 16-35 f/4L) slower focusing on adapted M5 than on native FF but in no means hunting or inaccurate. Relative to the 50 1.8 STM and 40 2.8 adapted or the native EF-M, I'll have to get back to you on that. The 50 1.8 STM does very well on the M5.... the 40, not so well but still no slouch. The 50 is just a great pairing. I would say the 50 adapted is as fast as the EF-M 22. My findings on the 40 mirror Dustin's I have found. But it's only AF speed, everything else is fine. I will get around to testing the adapted L vs STM's on the M5 one of these days and tell you what I thought.

Thank you, this is what I want to drill down on. The thin mirrorless + adaptor camp would have us believe that there are no tradeoffs to AF performance on an adaptor, and if that hasn't been your experience, it might imply that the most discerning/picky photographer might need a true EF mount make their EF glass sing. (Could also be DPAF vs. the standard SLR AF, in fairness.)

Again: I'd like more on this from everyone, please. Those who have an M5 or M6, relatively modern FF SLR w/DPAF, some quick L lenses and an adaptor, please go to town and tell us what you find!

Ideally, we'd compare the focus speed of:

SLR standard AF setup
SLR with AF on LiveView
M5/M6 through an adaptor

- A

I would never be surprised that anytime (everytime?) you add an adapter, an extender, bellows, extension tubes, spacers or an XYT4-X Pro Optical Modulator to a lens and extend the distance between rear element and sensor, you would lose something. And if anyone thinks this isn't true then they don't have a dog in their precious flange distance fight.

 

[AvTvM]

Thank you, this is what I want to drill down on. The thin mirrorless + adaptor camp would have us believe that there are no tradeoffs to AF performance on an adaptor, and if that hasn't been your experience, it might imply that the most discerning/picky photographer might need a true EF mount make their EF glass sing. (Could also be DPAF vs. the standard SLR AF, in fairness.)

you must be joking. A Canon lens mount adapter [like EF/EF-M] is a distance tube with wiring thru of the electrical contacts.

Any "negative impact" on AF performance comes from the fact that when (old) EF lenses are used, they are designed for DSLR / Phase-AF operation, not for mirrorless.

The only vaild comparison for EF glass is their AF performance *IN LIVE VIEW MODE* on DP-AF DSLRs .. and even that comparison is skewed, because the (better) DLSRs have bigger batteries and way more juice to power the AF drive in lenses, than the whimpy LP-E17 in an EOS M5/M6 offers.

There is no technical reason to assume AF performance of EF glass mounted on a decent Canon FF MILC with "slim mount" by means of an original-Canon adapter (not some cheap China cr+p) would not exactly match that lens' AF performance in live-view mode on a DP-AF sensored DSLR.

Older EF glass AF [not Nano USM, not STM, not chipped/firmwared for DP-AF sensors] will not necessarily "sing" in mirrorslapper Live view mode or on a mirrorless body ... totally irrespective of whether mirrorless cam has "native EF" mount or a new, short FFD "EF-X" mount ...

IF there is an AF speed/performance hit, it will NOT stem from use of a Canon lens mount adapter, but from those "legacy" EF lenses' inability [including L lenses preceding the very latest versions] to fully utilize the advanced options of DP-AF in DSLR Live View mode or of mirrorless DP-AF ... as simple as that.

 

[ahsanford]

Again, glad we're talking.

I've repeatedly heard one group of users state 'adding extra bits can't be helping' or 'Sony's adaptors don't work that well', etc. and imply that adaptors have a performance cost.

And then we have others saying and adaptor is just an electrical passthrough and should have no performance hit at all.

And AvTvM may very well be right that differences in focusing with adaptors may be more about the core AF technology of the body and not the lens.

So let's sort that out. Prove it once and for all. Run the same damn EF lens the three different ways I mentioned above -- SLR with the traditional AF setup, SLR with DPAF in liveview, and the adapted M5/M6. This should be a pretty straightforward exercise.

- A