Would you buy an EVF only camera with an M mount?

[raizans]

Advertisement (gone after registration)

34 minutes ago, adan said:

Raizans:

If something is out of focus wide-open - it will not be as sharp as it could be at f/5.6, either.

Depth of field is a poor substitute for accurate focusing (wide-open) in the first place. And then stopping down.

Unless one only prints at "drugstore" sizes.

Yes, assuming that focus shift isn't an issue, focusing with the lens at full aperture ensures accurate focus if you then stop down to take the photo.

But is there any reason to do the opposite: focusing with the lens stopped down, then opening up to take the photo at full aperture? That's just asking for inaccurate focus. I don't get it.

[Advertisement]

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

[raizans]

46 minutes ago, lct said:

So this grandpa almost focus all his pics at working aperture thanks to image magnification and focus peeking if needed.

Yeah, I know! Focusing at working aperture works great on EVFs.

[pgk]

10 minutes ago, raizans said:

Yeah, I know! Focusing at working aperture works great on EVFs.

No as precise as a rangefinder though, especially with wide-angles😄. I use both RF and EVFs. Horses for courses.

[250swb]

The world today for cameras that need an EVF (or not if you only want to use the LCD). Sigma fp L body, Leica L mount to M mount adapter, and 61mp for a fraction of the price of an M11. No brainer if you are in the market.

[Tailwagger]

1 hour ago, IkarusJohn said:

 

1. There’s no doubt that the SL lenses outperform the M equivalents on the SL - even the 50 APO Summicron-M. 

2. Just because you get smeary corners with the 35 ZM and 21 SEM does not mean the SL was not designed for M lenses.  One swallow does not summer make.

3. Now, this discussion will actually get us nowhere in making absolute statements.  My experience with my M lenses does not mean that all wide M lenses are fine on the SL.  But that’s not what I’m saying …

1. To be clear, I'm not just talking about Leica L-glass.  I'm talking about a plastic Pano zoom at 1/4 the cost of the WATE.  Further, in the central frame the Zeiss is just as sharp, possibly sharper IMO, than the vaunted SL-35. But the corners? Awful. If the SL2 has the equivalent M micro-lensing, how can that be?

2. Well that's two swallows. And the same is true of the WATE (less so at 21mm, but at 16mm, back to ugh) and the Elmarit 28mm V3.  So that's four swigs at the bottle. I have no such complaint regarding how any of these lenses have performed on every M I've owned.  The 28 'lux, as you've noted, seems okay, not quite M10-R/M11*** good, but good enough not to be of serious concern.

3. The whole thing is simple physics. Lens which project at acute angles are going to suffer without introducing additional measures to compensate for the inevitable diffraction caused by the cover glass. The greater the angle of incidence, the worse things become. There's nothing mystical here, the M does a better job because it was specifically designed to do so. 

Now some may find a drop in corner performance acceptable and in many instances even unnoticeable given the photograph's composition. Shoot a landscape at infinity with sky top of frame and field of wheat in near foreground and you'll swear that SL2 kicks ass with a WATE or 21 SEM, ZD 35, what have you, as the central cross frames of the image will indeed be sharp. The corner grass will just seem a little more OoF, the sky has no detail with sharp edges anyway and your mountains and trees center frame will be sharp. Shoot close focus indoors with the lens extended and lower angle of incidence, and again, you may not see anything untoward. Shoot street, same deal. But shoot a concert scene with an offending wide, as I did in my first weeks with the SL2, from a balcony back of house, with performers bottom of frame and cathedral ceiling at top and be prepared to be quite disappointed.

I get that a bunch of M lenses work fine on the SL... I've shot with the 75 'lux, 135 APO on it and they're great. And I also get that you're happy with the 21/28 'luxes on the SL. I recently tested the 28 'lux on the SL2 and lo... I concur. But any number of lenses, past, present and presumably future aren't. I have zero interest in buying a system that may or may not fully support an M lens I currently or  one day decide to own.  Were I to opt for a CV 35 APO or a ZD 15, let alone a SEM 18, I don't want to even wonder about whether it will be compromised before I push 'Add to Cart'.  With an M, I don't, with an SL... different story.  We are talking about the most expensive commercially available FF camera systems and optics on the planet. As such, my expectations aren't about acceptable, nor am I prepared to shrug off such hassles, compromises or outright failures. It's one thing for folks to consider moving to the SL, taking their current M lens collection along while they build up a set of native lenses and accept some compromises or bumps in the road in the mean time. It's entirely another to suggest that an SL is a suitable substitute for a true RF-less M. We've had that argument on these pages for half a dozen years now and many of those who once militantly saw things the SL that way, Jono in particular... Jaap never of course 😉, are finally coming around. Assuming the kinks can be worked out, an M mount based EVF is inevitable. In the meantime, M on M, L on SL, AFAIC.

*** Actually, and I find this rather perplexing, the level of CA manufactured by the 28 'lux on the M11 is driving me absolutely nuts. I'd send the lens in to be checked, but I see nowhere near the CA on either the SL2 or M10. If I have a knock on the M11, it's that I'm seeing significantly more CA with all my M lenses across the board than I ever did previously.  Sure, there's more pixels so you'd expect you'd wind up with a few extra rows of purple pixels, but I'm seeing far, far more purpling than I'd ever have suspected was possible.  With most lenses not that bad and easily corrected, but the 28 'lux is particularly problematic to the point where it's now out of the bag for general use. Clearly, no matter which way you fall in this discussion, making a device that works optimally across a legacy of 70 years+ of optics isn't that easy. 

[Tailwagger]

3 hours ago, raizans said:

You should focus with the lens wide open if you're going to take a photo with the lens wide open. Stopping down to focus and then opening up to take a photo is an odd thing to do, and I can't think of any reason to do that.

Arg.  It's a test, not a methodology. Stop down. Focus. Now open the lens up. Is the point of focus you select when stopped down perfectly sharp? If not, you missed focus and were fooled into selecting the wrong plane of focus by the EVF.  See my evidence above. The EVF says that the window frames are in focus. The result shows that they clearly aren't. It all too easy to be fooled into thinking you're in when you're out. 

[Tailwagger]

Advertisement (gone after registration)

2 hours ago, raizans said:

Yeah, I know! Focusing at working aperture works great on EVFs.

What's the widest focal length you use?  This is utterly untrue shooting wides on the M. As for focus shift,  a few inches either way when shooting with a 21mm is a non-issue, being off by multiple feet is. 

[raizans]

This test is a demonstration of faulty technique, not faulty technology. You can focus accurately with an EVF if you know how to use it, the way that makes sense. If you have to purposefully use bad technique that nobody does in practice to “reveal” a shortcoming, that only goes to show that EVFs work well.

Edited March 14, 2022 by raizans

[IkarusJohn]

24 minutes ago, Tailwagger said:

1. To be clear, I'm not just talking about Leica L-glass.  I'm talking about a plastic Pano zoom at 1/4 the cost of the WATE.  Further, in the central frame the Zeiss is just as sharp, possibly sharper IMO, than the vaunted SL-35. But the corners? Awful. If the SL2 has the equivalent M micro-lensing, how can that be?

2. Well that's two swallows. And the same is true of the WATE (less so at 21mm, but at 16mm, back to ugh) and the Elmarit 28mm V3.  So that's four swigs at the bottle. I have no such complaint regarding how any of these lenses have performed on every M I've owned.  The 28 'lux, as you've noted, seems okay, not quite M10-R/M11*** good, but good enough not to be of serious concern.

3. The whole thing is simple physics. Lens which project at acute angles are going to suffer without introducing additional measures to compensate for the inevitable diffraction caused by the cover glass. The greater the angle of incidence, the worse things become. There's nothing mystical here, the M does a better job because it was specifically designed to do so. 

Now some may find a drop in corner performance acceptable and in many instances even unnoticeable given the photograph's composition. Shoot a landscape at infinity with sky top of frame and field of wheat in near foreground and you'll swear that SL2 kicks ass with a WATE or 21 SEM, ZD 35, what have you, as the central cross frames of the image will indeed be sharp. The corner grass will just seem a little more OoF, the sky has no detail with sharp edges anyway and your mountains and trees center frame will be sharp. Shoot close focus indoors with the lens extended and lower angle of incidence, and again, you may not see anything untoward. Shoot street, same deal. But shoot a concert scene with an offending wide, as I did in my first weeks with the SL2, from a balcony back of house, with performers bottom of frame and cathedral ceiling at top and be prepared to be quite disappointed.

I get that a bunch of M lenses work fine on the SL... I've shot with the 75 'lux, 135 APO on it and they're great. And I also get that you're happy with the 21/28 'luxes on the SL. I recently tested the 28 'lux on the SL2 and lo... I concur. But any number of lenses, past, present and presumably future aren't. I have zero interest in buying a system that may or may not fully support an M lens I currently or  one day decide to own.  Were I to opt for a CV 35 APO or a ZD 15, let alone a SEM 18, I don't want to even wonder about whether it will be compromised before I push 'Add to Cart'.  With an M, I don't, with an SL... different story.  We are talking about the most expensive commercially available FF camera systems and optics on the planet. As such, my expectations aren't about acceptable, nor am I prepared to shrug off such hassles, compromises or outright failures. It's one thing for folks to consider moving to the SL, taking their current M lens collection along while they build up a set of native lenses and accept some compromises or bumps in the road in the mean time. It's entirely another to suggest that an SL is a suitable substitute for a true RF-less M. We've had that argument on these pages for half a dozen years now and many of those who once militantly saw things the SL that way, Jono in particular... Jaap never of course 😉, are finally coming around. Assuming the kinks can be worked out, an M mount based EVF is inevitable. In the meantime, M on M, L on SL, AFAIC.

*** Actually, and I find this rather perplexing, the level of CA manufactured by the 28 'lux on the M11 is driving me absolutely nuts. I'd send the lens in to be checked, but I see nowhere near the CA on either the SL2 or M10. If I have a knock on the M11, it's that I'm seeing significantly more CA with all my M lenses across the board than I ever did previously.  Sure, there's more pixels so you'd expect you'd wind up with a few extra rows of purple pixels, but I'm seeing far, far more purpling than I'd ever have suspected was possible.  With most lenses not that bad and easily corrected, but the 28 'lux is particularly problematic to the point where it's now out of the bag for general use. Clearly, no matter which way you fall in this discussion, making a device that works optimally across a legacy of 70 years+ of optics isn't that easy. 

You referred in your original post to the SL - ie, SL(601), yet your specific references are to the SL2.  I have no idea about this camera.  I was responding to the SL, which as I recall is optimised for M lenses.  I’m sure I could search back through the forum to find the relevant quotes, but I’m happy that with the SL (not the SL2), the 21 Summilux, 28 Summaron, 28 Summilux, 35 Summilux pre-asph, 50 Summitar, 50 Summilux ASPH, 50 APO Summicron, 50 Noctilux and 75 Summilux, there is no visible softness in the corners (more than on an M).  Perhaps I’m lucky, or just wilfully blind.

I had the 3 SL zooms, 75 Summicron & 50 Summilux.  I decided that I like the M system more, and I could get the images I like with the 24-90 zoom and my adapted M lenses.  The Noctilux is way better on the SL than the M (easier to focus accurately and almost no purple fringing); Leica recommend using the 75 Noctilux and 90 Summilux on the SL, if you wish to shoot wide open.

Are you able to point to the problems you encountered on the SL(601) - I’m not talking about pixel peeping.  Undoubtedly, viewed at 200%, it is possible to establish that for a given focal length, SL primes are better.  That’s not the issue.  I agree that maintaining compatibility is a challenge.  The SL(601) is the best universal platform for Leica lenses as far as I’m concerned.  The Elmarit-R 180/2.8 is fantastic on the SL.  I wasn’t taken with the improvements in the SL2, as I’m not taken with the M11.  Were they both bridges too far?

[lct]

I have no experience with the SL but i seem to recall that Sean Reids used to test M lenses on it a couple years ago. His conclusion IIRC was that off-axis with challenging WA and UWA lenses the M240 did outperform any other camera that can mount RF lenses, including the SL.

[Tailwagger]

1 minute ago, IkarusJohn said:

Are you able to point to the problems you encountered on the SL(601) ...  Were they both bridges too far?

Nope, at the time I decided to add the M10 along side the 240 and passed on the SL.  Later, I began find reasons to acquire a FF AF camera and went SL2, so I can't speak to the SL... apologies if I use the term SL interchangeable for the two. 

As for bridge too far, who knows, but then I'm not the one to ask as I rarely know where I'm going wind up anyway 😉. Overall, I find the M11 to be the best M yet. Difficult to quantify, but I find the files are more malleable and to my eye wind up after processing with a greater sense of depth and smoothness. The added DR is welcome as is the improved metering when shooting in more challenging circumstance as I so often do. I also find the results with my Mandler glass far more appealing than I did with the 10-R.  So much so, that I no longer feel the need to re-acquire an M10 for that purpose. That said, there is nothing so dramatic in the M11 that I'd consider anyone uninterested in it to be foolish. In my case, I simply find it more satisfying to shoot with than the R, hence it stays and ultimately the R goes. While I wouldn't have said this when I started with the 240, I no longer see Leica as providing a style/level of IQ that cant be duplicated elsewhere. Given that, in some sense I suppose buying any new Leica could be reasonably seen as a bridge too far at this point. But how you get there is often just as meaningful as the result; I really can't see myself shooting regularly with anything other than a Leica which currently means the M11/SL2.

[Planetwide]

6 hours ago, Tailwagger said:

Arg.  It's a test, not a methodology. Stop down. Focus. Now open the lens up. Is the point of focus you select when stopped down perfectly sharp? If not, you missed focus and were fooled into selecting the wrong plane of focus by the EVF.  See my evidence above. The EVF says that the window frames are in focus. The result shows that they clearly aren't. It all too easy to be fooled into thinking you're in when you're out. 

Your methodology or test is completely flawed. By stopping down, you are deliberately making it impossible to accurately focus, and then blaming it on the EVF. I have honestly never heard of anyone doing this before now, and I am not sure why... It's simple, open it up and peaking will work the way it was designed. Open it up and zooming the EVF will work the way it was designed. Then stop down.

[adan]

24 minutes ago, Planetwide said:

Your methodology or test is completely flawed. By stopping down, you are deliberately making it impossible to accurately focus, and then blaming it on the EVF. I have honestly never heard of anyone doing this before now, and I am not sure why... It's simple, open it up and peaking will work the way it was designed. Open it up and zooming the EVF will work the way it was designed. Then stop down.

I suggest you go back and look at post #201. Tailwagger is saying exactly what you are saying - focus peaking doesn't work when stopped down.

His test was also to demonstrate exactly what you are saying - to another member who said that focusing when stopped down was easy.

[tashley]

2 hours ago, Planetwide said:

Your methodology or test is completely flawed. By stopping down, you are deliberately making it impossible to accurately focus, and then blaming it on the EVF. I have honestly never heard of anyone doing this before now, and I am not sure why... It's simple, open it up and peaking will work the way it was designed. Open it up and zooming the EVF will work the way it was designed. Then stop down.

You are 100% right of course but this does go to the heart of the problem: you can only focus accurately wide open but that has two consequences....

1) You can no longer shoot very spontaneously because you have top open up, focus, stop down. That's not a great recipe for classic Leica shooters

2) Focus shift on many lenses, particularly some of the older ones, means that you have to really know your lens to focus this way because with some lenses you need a bespoke approach. For example the classic pre-asp 35 Cron would really require to to focus wide open if you wanted to shoot wide open, but to focus at f2.8 and F4 if you wanted to shoot at those apertures. whereas by F5, depending on subject distance, you could revert to focusing wide open and then stopping down....

[pgk]

9 hours ago, Tailwagger said:

The whole thing is simple physics. Lens which project at acute angles are going to suffer without introducing additional measures to compensate for the inevitable diffraction caused by the cover glass. The greater the angle of incidence, the worse things become. There's nothing mystical here, the M does a better job because it was specifically designed to do so. 

And thereby lies the rub. Quite clearly any camera which uses micro lenses to compensate for acute angular illumination will be a compromise unless all the lenses operate at the same angle, which they don't. I simply do not understand this obsession with an EVF-M 'because it will be better with M glass'. Better, yes. Perfect, no. Using lenses designed to give a similar or the same angle of incidence (eg. SL leses on an SL) will give better results. Best to accept the M for what it is and not complicate it by producing a compromised system IMO. That will only lead to comparisons which may well show that digitally optimised systems can deliver extraordinarily good results which are probably better than those from compromised legacy systems. This crazy 60MP EVF obsession has very little to do with improving real world photography and a great deal to do with the obsession of specification.

[Tailwagger]

 

1 hour ago, pgk said:

 Best to accept the M for what it is and not complicate it by producing a compromised system

By this reasoning there never should have been a digital M in the first place. Should we just throw up our hands, toss every digital M every made into the ocean and go back to shooting film?  And while we're at it, round up all the SLs, Sonys, Nikons etc for dunking as well given that AFAIK they also employ micro-lensing.  No FF lens I'm aware of redirects each and every photon perfectly perpendicular to the frame. Every digital camera system made wrestles with the diffraction problem, the M just more so due to legacy.

Perfection is the province of gods, better is all we mortals have.  So I see things the other way around. If the M fails to move forward, there will be no M. That was proven nearly two decades ago with the intro of the M8 which, despite its undeniable charm, is a camera that few would ever tout as even remotely close to perfect. 

 

[pop]

17 minutes ago, Tailwagger said:

If the M fails to move forward, there will be no M

Possibly so, but which way is forward?

[lct]

With all due respect folks, it seems that some of you are underestimating image magnification as focus aid on electronic viewfinders. It works fine at all apertures up to f/11 in my experience, including on WA and UWA lenses if needed. You don't have to focus-recompose then so field curvature is not a problem any more this way. And or course you don't have to focus at full aperture a la grandpa . Just kidding. Here at f/11 @ 16mm (Sony A7r2 mod, Tri-Elmar 16-18-21/4, 16mm, full frame and 100% crop).

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!

 

 

Edited March 14, 2022 by lct

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[pgk]

 

2 hours ago, Tailwagger said:

By this reasoning there never should have been a digital M in the first place. Should we just throw up our hands, toss every digital M every made into the ocean and go back to shooting film?  And while we're at it, round up all the SLs, Sonys, Nikons etc for dunking as well given that AFAIK they also employ micro-lensing.  No FF lens I'm aware of redirects each and every photon perfectly perpendicular to the frame. Every digital camera system made wrestles with the diffraction problem, the M just more so due to legacy.

Perfection is the province of gods, better is all we mortals have.  So I see things the other way around. If the M fails to move forward, there will be no M. That was proven nearly two decades ago with the intro of the M8 which, despite its undeniable charm, is a camera that few would ever tout as even remotely close to perfect.

The digital M was a move forward. An EVF M is a move sideways. It offers a compromise which is far less effective than an EVF camera without AF and much more. At the same time an EVF only M would do away with the very thing that makes the M unique and offers a very different way to take photographs.

2 hours ago, pop said:

Possibly so, but which way is forward?

IMO increased dynamic range and reduce noise are both photographically useful and improvable parameters. 

26 minutes ago, lct said:

With all due respect folks, it seems that some of you are underestimating image magnification as focus aid on electronic viewfinders.

One of my most used lenses on a Sony A7ii/R is the 20/1.8. Accurate MF is much slower if you want accuracy. On an RF it is much faster and more precise. 

[Tailwagger]

46 minutes ago, pop said:

Possibly so, but which way is forward?

Like it or not, largely which ever way time, technology and societal trends take us.  And, of course, forward doesn't necessarily mean those steeped in the joys of M past and present, will be accepting of the direction of that development, which to a very large degree, is outside the control of Leica. But OTOH, there's also no guarantee that those same folks might not be deliriously happy either. The singular certainty, in my view, is that electronics will continue to replace mechanics whenever and where ever possible.  Forward in that sense might not mean greater levels of satisfaction, but it does mean lowering the cost of production which, assuming the optical RF finder is the single most important factor to the old guard, is an imperative given its production is so costly.

But the micro-lensing, angle of incidence problem is completely unrelated to whether or not there's an optical RF, a hybrid RF, fully electronic RF or merely a conventional EVF.  If the conundrum to be solved is removal of the intervening layer between the lens and photo sites to maximize existing M optical performance, then the solution is, so to speak, 'simple'... get rid of the cover glass.  And in some sense, this is a solved problem; just use film.  The question then becomes what might digital film look like?

I'm reminded of James Burke's book/television series, Connections, the premise of which is that breakthrough advances rarely are attributable to a singular development, but rather result of a combinations of smaller technological advances which initially seem unrelated until some clever brain realizes that they can be combined to produce something revolutionary.  Now I'm no where near clever enough for such things, but if we pretend for a moment, a quick scan of the net reveals dozens of scholarly articles from researchers doing work in electrical conductivity related to nanotubes. One could imagine an ultra thin, tough as nails, flexible backbone of zillions of conducting nanotubes overlaid with some form of organic sensor deposition.  60Mpx? Pshaw, 60Gpx. Again, a quick scan of the web and one finds several interesting, but undoubtedly for most of us, inscrutable papers...

• "As new kinds of porous materials with high surface area, Metal–Organic Frameworks (MOFs), comprising coupling units (metal ions or metal-oxo clusters) coordinated by organic ligands, have received a lot of attention since being first defined in the 1990s by Yaghi and Li.¹ Their diverse structures and tunable properties (including pore size, metal center, functional linkers and post-synthetic modification) exhibit broad potential for different applications... But MOFs are preferably required in the form of surface layers/films for many other applications such as sensors, catalysis, electronic devices (including optoelectronic and electrochemical energy storage and conversion devices) and membranes."

Would this 'digital film' come in a retractable roll housed in a can containing all the necessary computing and power to work in any exiting film camera? Doubtful, though one could muse over the potential for a 2053 kickstarter project to adapt the technology to rekindle interest in the M3 at the century mark.  What I do think is very likely, however, is that at some point down the road, the diffraction issue induced by the cover glass requirement will be overcome. It's not hard to imagine that all sorts of other demands completely unrelated to photography will drive materials science to advance to a point where an ultra thin coverless (or impossibly thing spray on cover) sensor resilient enough to stand up to rigors of photography will be possible.