Leica Q3 or Fujifilm GFX 100RF

[Le Chef]

Advertisement (gone after registration)

What do you mean by “chipped lens” and “chipped adaptor”? What do they do and what instructions are they giving the processor? Surely the eye detection is part of the processor programming. In which case what role does this “chipped lens” or “chipped adaptor” play?

Edited April 19 by Le Chef

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Smogg]

7 hours ago, Simone_DF said:

Just wanted to point out that even with the Nikon ZF, eye confirmation is supported only with chipped lenses or adapter, e.g. if you mount a M lens on a dumb adapter, it won’t work, but if you mount the same lens on a chipped adapter, then you get a green confirmation box when the eye of a subject is in focus. 

Maybe the chipped adapter only tells the camera the focal length? Leica has a code on the lens for that.

[Simone_DF]

18 hours ago, Le Chef said:

What do you mean by “chipped lens” and “chipped adaptor”? What do they do and what instructions are they giving the processor? Surely the eye detection is part of the processor programming. In which case what role does this “chipped lens” or “chipped adaptor” play?

A dandelion chip is an electronic device installed on the bayonet mount of a non-electronic lens or adapter, which enables electronic interoperation with certain cameras. Camera functions enabled by the chip may include exposure metering, aperture display and control, focus confirmation and fine-tuning, and Exif metadata recording, but they depend on the camera model and brand. These adapters have been around for 20+ years, I had one for my old Canon DSLR. 

The Nikon ZF works like this:

Non-chipped lens, like your average M mount lens: automatic eye recognition and focus magnification on the eye (focus box is always grey)

Chipped lens: automatic eye recognition and focus magnification on the eye + focus confirmation (box turns green when focus is achieved)

Non-chipped lens on chipped adapter: automatic eye recognition and focus magnification on the eye + focus confirmation (box turns green when focus is achieved)

Perhaps, as @Smogg points out, the 6bit coding on Leica lenses could be enough, as to me it sounds like the most important bit of information transmitted by the electronic contact to the camera is the focal length.

[Simone_DF]

The TTArtisan adapter from M mount to Nikon Z goes even one step further by adding "focus trap":

TLDR if you don't want to watch the video: the camera will automatically take a picture while you're focusing when it confirms focus via focus confirmation  

[Chris W]

On 4/15/2025 at 12:18 PM, keithlaban.co.uk said:

I'd miss being able to create images such as these:-

 

 

 

Fab photos. Makes me more liable to buy a Q3 43 one day.

[Smogg]

39 minutes ago, Simone_DF said:

The TTArtisan adapter from M mount to Nikon Z goes even one step further by adding "focus trap":

TLDR if you don't want to watch the video: the camera will automatically take a picture while you're focusing when it confirms focus via focus confirmation  

So, as far as I understand, there are no technical obstacles to eye/face detection in the EVF-M. The only question is whether Leica has the expertise to implement this in 2025 or whether we will have to wait a little longer.

[keithlaban.co.uk]

Advertisement (gone after registration)

9 minutes ago, Chris W said:

Fab photos. Makes me more liable to buy a Q3 43 one day.

Thank you, much appreciated.

[Le Chef]

Since it’s more likely than not that the M-EVF will be based on the M11, then the restrictions will be determined by the current sensor and processor.

As I understand it, and I may be wrong:

The sensor as used in M11 has the code for PDAF written into it, but it’s not been “switched on” to work with the processor. Leica presumably reasoned there was no need to do this as M lenses are all manual focus. And I’m not sure about the micro lenses needed to make PDAF functional.

The Maestro III processor can handle PDAF but is not that sophisticated by current standards and would have difficulty with face and eye detection.

It leaves a lot of questions to be answered about what realistically Leica would commit to in terms of developing this variant. Go to far and you shade the M12 whenever that comes out. Too little development and you end up with a “meh” response and poor sales.

[Simone_DF]

31 minutes ago, Le Chef said:

The sensor as used in M11 has the code for PDAF written into it, but it’s not been “switched on” to work with the processor. Leica presumably reasoned there was no need to do this as M lenses are all manual focus. And I’m not sure about the micro lenses needed to make PDAF functional.

This is fine. The SL3 has the same sensor, microlenses and PDAF. 

 

32 minutes ago, Le Chef said:

The Maestro III processor can handle PDAF but is not that sophisticated by current standards and would have difficulty with face and eye detection.

The SL2 and SL2-S have the Maestro III and no difficulty with face and eye detection. 

[steve edmunds]

On 4/18/2025 at 3:21 PM, Smogg said:

Today I photographed my children's games with the GFX100RF. I was pleased... But then I picked up the M11-P and continued with it. It's like night and day, you can't even compare the speed of work and the satisfaction from shooting. How much easier it is to work with a rangefinder.

your focus with the rangefinder is quicker than the fuji AF with your kids moving around , is that what your saying?

[Le Chef]

7 hours ago, Simone_DF said:

This is fine. The SL3 has the same sensor, microlenses and PDAF.

But the version in the M11 does not. Leica would need to have the chips “activated” for the M11-EVF to communicate phase shifts to the processor, as I understand it.

[Smogg]

5 hours ago, steve edmunds said:

your focus with the rangefinder is quicker than the fuji AF with your kids moving around , is that what your saying?

Yes, that's right. However, I never shoot wide open, the context of what's happening is important to me.

[SrMi]

3 hours ago, Le Chef said:

But the version in the M11 does not. Leica would need to have the chips “activated” for the M11-EVF to communicate phase shifts to the processor, as I understand it.

I doubt that. PDAF pixels are regular pixels with different masking. As the toppings are customer-specific, Leica could have chosen to implement toppings without any PDAF-specific bindings. For example, Hasselblad's PDAF spacing differs from Fuji's, despite using the same sensor.

Jim Kasson: The PDAF pixels are created during the “topping” phase of sensor fabrication, at the same time as the color filter array (CFA) is applied. 

[Simone_DF]

16 hours ago, Smogg said:

So, as far as I understand, there are no technical obstacles to eye/face detection in the EVF-M. The only question is whether Leica has the expertise to implement this in 2025 or whether we will have to wait a little longer.

They do have the expertise, but my biggest concern is that the current implementation of eye/face detection on the SL and Q is far from ideal. When there's one or two people in the frame, it's very good. When there's 3 or more people, it starts to jump from face to face / eye to eye and it becomes a nightmare. If the same happens on the M, especially if there's auto-magnification on the eye, it won't be very usable.  

[setuporg]

Here’s an excellent review of the GFX100RF but its alleged thought-conceiver!  (To return this thread to the original meaning from a deep dive into PDAF pixels…)

https://jonasraskphotography.com/2025/03/20/impossibly-possible-the-fujifilm-gfx100rf-review/

[LocalHero1953]

On 4/19/2025 at 9:15 PM, Simone_DF said:

Just wanted to point out that even with the Nikon ZF, eye confirmation is supported only with chipped lenses or adapter, e.g. if you mount a M lens on a dumb adapter, it won’t work, but if you mount the same lens on a chipped adapter, then you get a green confirmation box when the eye of a subject is in focus. 

I assume a chipped lens is just to identify the lens to the camera. Leica's 6-bit coding would do the same.

[SrMi]

2 minutes ago, LocalHero1953 said:

I assume a chipped lens is just to identify the lens to the camera. Leica's 6-bit coding would do the same.

You can set the focal length manually, which helps with IBIS but does not replace chipped lenses functionality. I assume that chipped lenses also pass on the focus distance info.

[Smogg]

3 minutes ago, SrMi said:

You can set the focal length manually, which helps with IBIS but does not replace chipped lenses functionality. I assume that chipped lenses also pass on the focus distance info.

If a chipped adapter can determine the focal length and distance of an unchipped lens, then the same can be done inside the camera.

[Advertisement]

Hallo,

Look here: Leica Q

[LocalHero1953]

20 minutes ago, SrMi said:

You can set the focal length manually, which helps with IBIS but does not replace chipped lenses functionality. I assume that chipped lenses also pass on the focus distance info.

I can't see how the focus distance would help with picking out a face to highlight, but then I don't know how such things could work anyway.

[hdmesa]

On 4/19/2025 at 1:10 PM, Smogg said:

I still hope that Leica will present something special for its centenary.

You can hope, but there is no way Leica is doing to somehow implement a feature they don't even have working on the SL3/SL3-S, in this case subject detection with a manual focus M lens. The best we can hope for is an option to display a virtual rangefinder patch in the EVF.

Edited April 21 by hdmesa