When will we get the M10M, really?

[dkmoore]

Advertisement (gone after registration)

On 12/20/2019 at 9:29 AM, Herr Barnack said:

Sometimes an M camera's rangefinder mechanism and a particular lens need to be calibrated to each other. 

The Noctiluxes - 50mm 0.95 and 75mm 1.25 as well as the new 90mm Summilux come to mind in that regard...

Whenever the "what is your wish for the next M" posts occur I say the same thing. I'm not necessarily wishing for an OVF that never needs adjustment. Rather, I am looking for a tweak to the current system as I don't want to go full EVF in the M. All I want is the sensor to know when the rangefinder patches align and light up the frame lines red, green, blue, or white when that happens. Then you get best of both worlds.

[Advertisement]

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

[adan]

4 hours ago, dkmoore said:

Whenever the "what is your wish for the next M" posts occur I say the same thing. I'm not necessarily wishing for an OVF that never needs adjustment. Rather, I am looking for a tweak to the current system as I don't want to go full EVF in the M. All I want is the sensor to know when the rangefinder patches align and light up the frame lines red, green, blue, or white when that happens. Then you get best of both worlds.

How do you propose Leica do that?

The rangefinder patch images are completely independent of each other and not aligned at all - until they pass through the viewfinder (where the are combined) and to your eye - which is the "sensor" that determines they are aligned.

You can move your eye a mm or so "off-center" and the images will dis-align themselves - which shows how critical positioning is. Thus a "sensor" to do what your eye does also must be placed directly in the middle of the camera eyepiece to work. Which will mean that looking though the viewfinder, you won't see your subject - you'll just see the sensor.

Rough schematic of the M RF/VF - you see the subject (red line) by looking through the eyepiece lens (bottom left) and a split block of glass with a silver "semi-mirror" coating diagonally through it. You also see a secondary view of the subject from a different angle (yellow line) reflected from the semi-mirror surface.

The only place those two views exist together to be compared or aligned - by eye or by electronics - is the blue line. Which is mostly inside solid glass.

You can download this .jpg diagram. if you can draw on it exactly where you would place optical sensor(s), and any additional hardware needed, so that the camera could sense when the "correct" two images are aligned, and repost your suggested engineeriing, that will be fine.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Assuming you can do that, you can move on to the problem of deciding which objects in the combined image are supposed to be aligned. Because the RF patch contains not just a single point, but a fair area of the scene.

-------------

This is a stereo pair. Two overlapped pictures from slightly different left-right locations. Which is exactly how the Leica rangefinder views the world - through two different windows a few inches apart.

It simulates just the "rangefinder patch" of dual overlapped images in the Leica RF. Two images with varying alignment due to parallax - viewing the world from two different places.

In the left image, I have aligned the red and cyan images to be "focused" on the worn patch inside the bus on the wheel-well (background). In the right image, I have aligned the images to be "focused" on the chrome "["-shaped railing just inside the door.

You, as a photographer with a human brain, know whether you want the focus/alignment on the worn patch, or the railing - but how does your electronic servant know which alignment you want? It may light up if the railing is aligned - or the handicapped sign is aligned - or the "Please allow..." sign is aligned - or the worn patch is aligned - or the far window frames are aligned.  It can't read your mind.

Edited December 22, 2019 by adan

[dkmoore]

11 hours ago, adan said:

 

You, as a photographer with a human brain, know whether you want the focus/alignment on the worn patch, or the railing - but how does your electronic servant know which alignment you want? It may light up if the railing is aligned - or the handicapped sign is aligned - or the "Please allow..." sign is aligned - or the worn patch is aligned - or the far window frames are aligned.  It can't read your mind.

 

I'm not 100% sure but the Fuji X100 gets somewhat close. And, auto focus systems can't read our mind either but can get what we intend the focus to be on.

Maybe this works via the thumb control that is on the SL2 or S system somehow?

I just think it would be great to keep the OVF but have some sort of overlay that resolved the worry of rangefinder alignment. That said, I am far from an engineer, but I am sure there is a way to do it.

[Mark II]

11 hours ago, adan said:

How do you propose Leica do that?

Leica could put an on-sensor PDAF detector, just like most current mirrorless cameras, except that rather than triggering an AF mechanism it could just change the frame line LED colour when "sufficiently" in focus. From an engineering perspective this is much easier than trying to play tricks with the rangefinder, and the usual RF calibration / mechanical tolerance issues are bypassed.

The only downside is that the focus detection is made through the lens, and so its usefulness would mainly be when shooting wide open (which might be what you need to make the current rash of Noctilux lenses practical for focusing with an M camera).

[jjvornov]

8 hours ago, Mark II said:

Leica could put an on-sensor PDAF detector, just like most current mirrorless cameras, except that rather than triggering an AF mechanism it could just change the frame line LED colour when "sufficiently" in focus. From an engineering perspective this is much easier than trying to play tricks with the rangefinder, and the usual RF calibration / mechanical tolerance issues are bypassed.

The only downside is that the focus detection is made through the lens, and so its usefulness would mainly be when shooting wide open (which might be what you need to make the current rash of Noctilux lenses practical for focusing with an M camera).

Isn’t this more or less what focus peaking provides with live view or EVF? It would be cool if they could overlay it on an optical viewfinder for us RF users.
 

[Jeff S]

Leica has experimented with a hybrid M viewfinder, but so far resulting in unacceptable compromise... the worst of both, not the best.  And there supposedly is this patent...

https://www.google.com/amp/s/leicarumors.com/2015/10/14/leicas-patents-for-optoelectronic-rangefinder.aspx/amp/

Jeff

[otto.f]

Advertisement (gone after registration)

1 hour ago, Jeff S said:

Leica has experimented with a hybrid M viewfinder, but so far resulting in unacceptable compromise... the worst of both, not the best.  And there supposedly is this patent...

https://www.google.com/amp/s/leicarumors.com/2015/10/14/leicas-patents-for-optoelectronic-rangefinder.aspx/amp/

Jeff

Probably the reason why Erwin Puts left the scene 😊

Edited December 23, 2019 by otto.f

[Mark II]

2 hours ago, jjvornov said:

Isn’t this more or less what focus peaking provides with live view or EVF? It would be cool if they could overlay it on an optical viewfinder for us RF users.
 

Yes, It is similar but the PDAF will give you an indication of the distance and the degree of the error, so in theory it can be more precise. 
 

I have used a canon DSLR with adapted MF lenses, with an adapter that enabled the red blinking square and bleep when accurate focus was achieved. Much easier and quicker than Using a traditional focus screen.
 

The main difficulty with PDAF or contrast detect will be the power consumption, but for a focus assist you could probably use a relatively low read out speed (or partial readout) from the sensor.