M240 liveview and focus confirmation in optical viewfinder

[diogenis]

Advertisement (gone after registration)

Genuine improvement: some folks are older than us and can't handle well the rangefinder procedure. Also, just because one is young it doesn't mean when he grows older he wont face the same issues. So, for these cases and for whatever else I can't think of right now, this is a genuine imrpovement.

Focus confirmation is happening within the VF. You never chimp. Which means that you never lose your.... decisive moment

Obviously if this is very difficult to implement or, if Leica wants to take another route -after all you can do the same right now by using an optional EVF- then other things are more important.

Plus, this whole idea starts to remind me of the hybrid VF of Fuji, which ok, even if it's great I don't want to see it in an M: they don't want to clutter the VF for a good reason, they want to keep things simple so one can focus in composing rather than gizmos blinking in and out.

[mjh]

Call me a skeptic, but since the rangefinder is the same as the M9P, how does it magically communicate which part of the frame you want in focus to LV?

It probably didn’t. Who said it would? I suppose it did check for micro-contrast in the center of the frame. And just to re-iterate: reportedly this didn’t work very reliably, which doesn’t surprise me at all. And thus this (experimental) feature was scrapped.

 

Again: The feature the OP called for can be implemented and has been implemented, but it didn’t work so well as to make it into the final product. And that’s probably the end of this; Leica isn’t likely to revive this feature in a future firmware update.

[luigi bertolotti]

It probably didn’t. Who said it would? I suppose it did check for micro-contrast in the center of the frame. And just to re-iterate: reportedly this didn’t work very reliably, which doesn’t surprise me at all. And thus this (experimental) feature was scrapped.

 

Again: The feature the OP called for can be implemented and has been implemented, but it didn’t work so well as to make it into the final product. And that’s probably the end of this; Leica isn’t likely to revive this feature in a future firmware update.

 

Me too think so... even it could be made technically reliable, I think it could someway "disturb" the user looking at the RF,,, would the 2 infos really MATCH ? I am aligning something with the RF... the focus confirmation "blinks"... is it on the SAME detail of the image ? The area on which the focus analisys is performed is lens' dependent... the RF patch is not...(shadow of goggles... ) they could even engineer some "variable area of focus check" system, based on lens' recognition...(in theory with an alghoritm like "analyzed area is always a circle with radius=height of RF patch")...thus forther making complex its design... .

Yesterday, for the first time, I played a little with the XPro1 of a friend... I think that that could be the right future for Leica VF... switch to LV when you don't want RF... I found that switching is quick and easy on the Fuij... even if I think that in a possible implementation within the classical Leica RF, to "double check the focus" won't prove to be so useful, nor effective in practice... they are complementary system for different situations and lenses.

[digitalfx]

It probably didn’t. Who said it would?

 

Common sense. What would be the point of the feature if you didnt know "what" was in focus? This is precisely my point, if there is no communication what exactly is being confirmed?

 

 

Again: The feature the OP called for can be implemented and has been implemented, but it didn’t work so well as to make it into the final product. And that’s probably the end of this; Leica isn’t likely to revive this feature in a future firmware update.

 

I dont understand how you can make this claim if you have never seen this. Clearly the camera can't communicate with the rangefinder, so how could it be possible? It seems like you are saying it can be implemented but it cant so it won't.

[luigi bertolotti]

I dont understand how you can make this claim if you have never seen this. Clearly the camera can't communicate with the rangefinder, so how could it be possible? It seems like you are saying it can be implemented but it cant so it won't.

 

This doesn't matter at all in the system illustrated by mjh... there is no need of communication with the rangefinder mechanism in itself : he illustrates a system prototyped by Leica, in which a focus confirmation checked at sensor level in LV mode (probably through "usual" technologies used for AF systems) activates a signal visible into the Viewfinder (some blinking LED...other, in this thread, suggested the blinking of frames... same principle) : the RF mechanism has no role into this : simply, one has both the usual RF alignement AND the focus confirmation under his eye, in the Viewfinder : smart idea in theory, proved not so practical in Leica-made tests, according to mjh.... I think he has told a true story : we have had, from him, many proofs that he is personally rather close to Leica R&D dep.

[digitalfx]

This doesn't matter at all in the system illustrated by mjh... there is no need of communication with the rangefinder mechanism in itself : he illustrates a system prototyped by Leica, in which a focus confirmation checked at sensor level in LV mode (probably through "usual" technologies used for AF systems) activates a signal visible into the Viewfinder (some blinking LED...other, in this thread, suggested the blinking of frames... same principle) : the RF mechanism has no role into this : simply, one has both the usual RF alignement AND the focus confirmation under his eye, in the Viewfinder : smart idea in theory, proved not so practical in Leica-made tests, according to mjh.... I think he has told a true story : we have had, from him, many proofs that he is personally rather close to Leica R&D dep.

 

I dont think you understand what im asking, and I have no doubts that MJH is not making this up. But he stated that he never saw this, so my question is if you never saw this how are you so sure it can be done?

 

For the OP to get what he is asking for, the missing link is defining what part of the scene is in focus. How do you do that?

He wants visual confirmation in the rangefinder. This implies he is looking thru the viewfinder. If he is looking thru the viewfinder how does he tell the camera what part of the frame he wants in focus. It cant happen with telepathy.

 

If this cant be done, then the feature has absolutely no value.

[luigi bertolotti]

Advertisement (gone after registration)

I dont think you understand what im asking, and I have no doubts that MJH is not making this up. But he stated that he never saw this, so my question is if you never saw this how are you so sure it can be done?

 

Well... I'd say it can surely be done... and neither so difficult... basically, is a "normal" AF system (not different from the ones implemented in much more little and cheap cameras) in which the "focus OK" signal lights a LED instead of actuating the micromotor of a lens' focus mechanism... even simpler, at the end.

 

For the OP to get what he is asking for, the missing link is defining what part of the scene is in focus. How do you do that?

He wants visual confirmation in the rangefinder. This implies he is looking thru the viewfinder. If he is looking thru the viewfinder how does he tell the camera what part of the frame he wants in focus. It cant happen with telepathy.

 

If this cant be done, then the feature has absolutely no value.

 

That's the problem, I agree : by logic, the Leica-prototyped system would "check the focus" of a certain area in the central area of the sensor (round ?) = central area of the image.. roughly= area of the RF path (which you see in the VF).... but the uncertaineness about those "=" has proved the system valueless...

[digitalfx]

Well... I'd say it can surely be done... and neither so difficult... basically, is a "normal" AF system (not different from the ones implemented in much more little and cheap cameras) in which the "focus OK" signal lights a LED instead of actuating the micromotor of a lens' focus mechanism... even simpler, at the end.

 

So are you suggesting that a "normal" AF system can be added to a Leica rangefinder with a FW update only? Im sorry to be so blunt, but thats just absurd. Please read the first post more carefully. We all agree it can be done, but not with fw update alone. The OP asked if it could be done with a FW update. I still say this is impossible.

 

That's the problem, I agree : by logic, the Leica-prototyped system would "check the focus" of a certain area in the central area of the sensor (round ?) = central area of the image.. roughly= area of the RF path (which you see in the VF).... but the uncertaineness about those "=" has proved the system valueless...

 

This is silly, what is the point of this? Again, this feature is not what the OP asked for. Just because the original prototype had some unusable feature implemented does not mean what the OP asked for is possible. Clearly this proves my point that hardware is missing.

[luigi bertolotti]

So are you suggesting that a "normal" AF system can be added to a Leica with a FW update? .

 

Uh... obviously NO... ... sorry, I was thinking of the system described by mjh, in itself (it probably did include other hardware, you have to process the data for focus check and I doubt it can be made within the current CPU), and I simply forgot that the OP referred only to implementations made by firmware update...

[mjh]

Common sense. What would be the point of the feature if you didnt know "what" was in focus? This is precisely my point, if there is no communication what exactly is being confirmed?

As I have said: the camera would probably measure micro-contrast at the center of the frame. That would be the most logical choice. The rangefinder patch would give you a rough idea of where contrast is measured.

[pop]

To summarize:

 

Given the current hardware of the M (Typ 240) we can presume that it would be possible to implement some kind of focus confirmation.

 

The system could work by analyzing the image received by the sensor (in live view) and would flash some of the LEDs visible in the optical finder (such as the frame line illumination) when the contrast was high in one specific part of the image.

 

However, as there is no way for the camera to tell to the user which part of the image or the scene actually was in focus, most members here feel that the feature would be of limited use.

 

We may presume that Leica would feel the same and, hence, would not implement that particular feature in software.

[algrove]

OP --Got tired of reading all the replies so here's my take.

 

To me you are asking the frame lines to use the sensor instead of the RF patch to focus because to get anything flashing red or confirming where focus is achieved needs a sensor open to the image in order to determine said object is in fact in focus.

 

I have a 5D3 which gives me a focus conformation light and also an RX1 which does same, but both are using the sensor to determine that point of focus and in doing so need a focus point determined by you to tell the camera where in the overall image you want something in focus.

 

Trust this makes sense.

[digitalfx]

OP --Got tired of reading all the replies so here's my take.

 

To me you are asking the frame lines to use the sensor instead of the RF patch to focus because to get anything flashing red or confirming where focus is achieved needs a sensor open to the image in order to determine said object is in fact in focus.

 

I have a 5D3 which gives me a focus conformation light and also an RX1 which does same, but both are using the sensor to determine that point of focus and in doing so need a focus point determined by you to tell the camera where in the overall image you want something in focus.

 

Trust this makes sense.

 

precisely what I have been trying to say. Thanks.

If the user isnt telling the camera what he/she wants in focus, there is no point in having this feature.

[mjh]

If the user isnt telling the camera what he/she wants in focus, there is no point in having this feature.

How are you telling the camera what you want to be in focus when using the rangefinder? A sensor-based focus confirmation could work in just the same way. I think you are envisioning something very complex, with user-selectable focus points or whatever, but the actual implementation would be much simpler, i.e. it would probably evaluate a fixed, centered spot.

[luigi bertolotti]

How are you telling the camera what you want to be in focus when using the rangefinder? A sensor-based focus confirmation could work in just the same way. I think you are envisioning something very complex, with user-selectable focus points or whatever, but the actual implementation would be much simpler, i.e. it would probably evaluate a fixed, centered spot.

 

Yes... roughly coincident to the RF patch... and if I have undestood well what Michael said about the Leica prototype, it was that "roughly" that made the system not so valuable in practical.

[digitalfx]

How are you telling the camera what you want to be in focus when using the rangefinder? A sensor-based focus confirmation could work in just the same way. I think you are envisioning something very complex, with user-selectable focus points or whatever, but the actual implementation would be much simpler, i.e. it would probably evaluate a fixed, centered spot.

 

I just dont see how this approach would add any value. What we currently have is far more useful. The current version of LV and the automatic focus aid are immensely more useful than what you are describing. Its no wonder Leica abandoned it.

 

I cant imagine this is remotely close to what the OP had in mind, but I could be wrong.

Keep in mind what you are describing means LV must be on. If LV must be on why not just use LV and focus peaking?

[sydkugelmass]

I just dont see how this approach would add any value. What we currently have is far more useful. The current version of LV and the automatic focus aid are immensely more useful than what you are describing. Its no wonder Leica abandoned it.

 

I cant imagine this is remotely close to what the OP had in mind, but I could be wrong.

Keep in mind what you are describing means LV must be on. If LV must be on why not just use LV and focus peaking?

LV plus hybrid finder a la X-Pro 1 could be a good choice... but future...

[lct]

Not ideal for waist level photography though.

[diogenis]

To summarize:

 

Given the current hardware of the M (Typ 240) we can presume that it would be possible to implement some kind of focus confirmation.

 

The system could work by analyzing the image received by the sensor (in live view) and would flash some of the LEDs visible in the optical finder (such as the frame line illumination) when the contrast was high in one specific part of the image.

 

However, as there is no way for the camera to tell to the user which part of the image or the scene actually was in focus, most members here feel that the feature would be of limited use.

 

We may presume that Leica would feel the same and, hence, would not implement that particular feature in software.

 

 

Erm.. There is a way for the camera to tell the user which part of the image is in Focus and this is called "focus peaking", where there is a red outline that informs the user that this part is in focus...

 

Maybe you wanted to say that there is no way for the user to tell the camera which part of the scene he wants to be in focus, but that can also be done by preselecting one or more focus points in the scene, so you can also do that.

 

That Leica has abandoned the idea because of some "problems" that had to do with the mechanism or precision or handling, is pure speculation and wishfull thinking. We actually don't know and propably never learn why they abandoned it.

 

After all there will soon be a new EVF coming with double the resolution of the old and as dpreview claims, it's so good that it feels like an optical EVF of a dSLR.

And in there you can have as much optical confirmation as you can handle.

 

However Leica should think again out of the box and create a specialized and FAST system(With some new hardware:)) that will not have to use sensors, and in the blinkof an eye will be able to give focus in a point ahead thru the OVF and as you turn the focus ring.

[Advertisement]

Hallo,

Look here: Leica M

[NJH]

Over in Fuji land the focus box confirmation in the OVF has caused people no end of problems. For starters it can give misleading focus at short distances even though the box moves to account for parallax errors (but only after the camera has focused so it needs a fair bit of experience to predict where it will land). I also found that occasionally it would light up green leading me to think that it had focused on the object in the box when in fact the camera had decided to focus on the background. In the EVF or a DSLR this would be obvious but in the Fuji OVF you never actually see the camera focus just a box lighting up green. Personally I greatly prefer rangefinder focusing and the view provided by the patch.

 

There are lots of interesting things on the Fuji cameras I would like to see in a Leica but this certainly isn't one of them. The one big thing the X-pro has which I wish my M8 had above all else is a simple optical thing: the selectable viewfinder magnification to provide comfortable viewing for wide angle and telephoto lenses.