Leica CL Image Quality--CL vs SL vs M10

[scott kirkpatrick]

Advertisement (gone after registration)

For panoramas I certainly agree that edge sharpness matters.  But would you be making a panorama with the lens wide open?  I don't think I ever have, but you may be different.  Keep in mind, I'm not arguing about the relevance of corner sharpness in general, just about its relevance with a large aperture.  Generally the narrow depth of field is going to do more to damage corner sharpness than the lens quality ever would--even with a 28mm f/2.8  I don't know.  I looked through my own images (the challenge I gave to LCT), and after about five minutes of work (not exhaustive, obviously) I could only find two classes of my own photographs where I was shooting wide open and would care about corner performance.  The first is astrophotography, of which I do a lot.  Obviously, with an astrophoto everything is at infinity, so you effectively have a flat subject.  It's also, essentially, a landscape image where lots of fine detail distributed across the frame is desirable.  And the light is very low, of course, so you would want to shoot wide open if possible  Plus, you minimize diffraction spikes if you don't have the aperture diaphragm in the way.  So for this use case, definitely it matters.  The second use case I could find in my own pictures was pictures I have taken of my daughter's theater performances.  When shooting from the audience, again there is effectively a flat stage or very nearly so.  Depending on the scene, edge and corner performance could be relevant.  Most of the pictures it wouldn't have mattered, but there were definitely a few where it did.

 

That's it.  I couldn't find anything else.  I'm not saying there aren't other situations, just that I can't think of them.

 

I'll offer you the same challenge as I did LCT.  Spend five minutes going through your own pictures and see if you can find one that was shot wide open (even a panorama would be fine) where edge/corner sharpness mattered.  At least with my photographs it's hard.  Everyone has their own style, though, so your needs may be different from mine.

I love night sky photographs but seldom take them.  I did get a red moon a few years back with the R 280/4 on (I think) an Olympus E-M1, after waiting most of the night for the moon to enter the penumbra and the clouds and skyline to cooperate.  That's the opposite limit from your wide angle whole MilkyWay examples, but it had to be taken wide open, and the exposure was constantly changing.  I didn't worry about corner sharpness.  I have a lovely example, pinned up in my office, that I didn't take, of the Milky Way over the ALMA observatory in Chile, shot from the salar, with the lights of a truck or possibly an antenna-mover, making a streak on the road up to the high site.

 

I recently shot a panorama from the roof of our building using an M 18/3.8 lens, wide open as it is intended to be used.  It's in the CL image thread somewhere and in my CL samples album on Flickr.  Here's another candid, also within the first page that I looked at, where the subjects are all over the frame:

 

C1010472 by scott kirkpatrick, on Flickr

 

(misfocussed, of course, damned AF)

Edited February 15, 2018 by scott kirkpatrick

[jaapv]

I love night sky photographs but seldom take them.  I did get a red moon a few years back with the R 280/4 on (I think) an Olympus E-M1, after waiting most of the night for the moon to enter the penumbra and the clouds and skyline to cooperate.  That's the opposite limit from your wide angle whole MilkyWay examples, but it had to be taken wide open, and the exposure was constantly changing.  I didn't worry about corner sharpness.  I have a lovely example, pinned up in my office, that I didn't take, of the Milky Way over the ALMA observatory in Chile, shot from the salar, with the lights of a truck or possibly an antenna-mover, making a streak on the road up to the high site.

 

I recently shot a panorama from the roof of our building using an M 18/3.8 lens, wide open as it is intended to be used.  It's in the CL image thread somewhere and in my CL samples album on Flickr.  Here's another candid, also within the first page that I looked at, where the subjects are all over the frame:

 

C1010472 by scott kirkpatrick, on Flickr

 

(misfocussed, of course, damned AF)

 

 

I think panoramas should be taken out of the equation. In my book they are composed of multiple images and stitched, intended to be printed large. A single cropped wideangle shot will never cut it. It is a pity that Leica did not implement Panasonic's panorama mode on the CL. It saves ages on the computer.

 

(late edit: it is there, but hidden in the menu)

 

 

 

Anyway, here is the CL in a final version of a 12-shot panorama 15500x4000 pixels. 18-56 @ 18.

 

 

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!

[Advertisement]

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

[scott kirkpatrick]

  The second use case I could find in my own pictures was pictures I have taken of my daughter's theater performances.  When shooting from the audience, again there is effectively a flat stage or very nearly so.  Depending on the scene, edge and corner performance could be relevant.  Most of the pictures it wouldn't have mattered, but there were definitely a few where it did.

 

I've also met that use case.  For stills I usually try to sit in an aisle seat which means that I have a 30 to 45 degree angle into the set and have some clear space to shoot over.  High school drama does seem to occur mostly in a line across the front of stage (I suspect because they don't have much lighting flexibility), so anything with more than a single actor is going to need to be clear across the frame.  Here's an example using the SL and an R Summilux 80 wide open:

 

S1000554 by scott kirkpatrick, on Flickr

 

My son Tom, showing his Method chops.

Edited February 15, 2018 by scott kirkpatrick

[scott kirkpatrick]

I think panoramas should be taken out of the equation. In my book they are composed of multiple images and stitched, intended to be printed large. A single cropped wideangle shot will never cut it. It is a pity that Leica did not implement Panasonic's panorama mode on the CL. It saves ages on the computer.

 

Anyway, here is the CL in a final version of a 12-shot panorama 15500x4000 pixels. 18-56 @ 18.

 

 

Panoprint2.jpg

What a shame to have to view that at 480KB!  It's quite a lovely scene.  What size are you planning to print it?

 

I just put enough post processing time into the shots intended for a pano to ensure that I get skies of continuous intensity, render them, stitch them in PTGui, crop off the rough edges in Preview, and I am done.  That offers pretty good control, and takes 3-5 minutes, maybe once a month.

Edited February 15, 2018 by scott kirkpatrick

[jaapv]

About 2 meters

 

I use PS but it often messes up the perspective. that is most of the work.

[jmahto]

About 2 meters

 

I use PS but it often messes up the perspective. that is most of the work.

Lovely indeed. Have you tried LR's panorama merge? I am becoming a fan of it since it painlessly creates a merged DNG where I can still fix WB if needed. It is quite painless. Just select multiple pictures and from menu click merge. For complicated merges (needing manual intervention for aligning edges) you do need PS but only in rare case. 

[Jared]

Advertisement (gone after registration)

I love night sky photographs but seldom take them.  I did get a red moon a few years back with the R 280/4 on (I think) an Olympus E-M1, after waiting most of the night for the moon to enter the penumbra and the clouds and skyline to cooperate.  That's the opposite limit from your wide angle whole MilkyWay examples, but it had to be taken wide open, and the exposure was constantly changing.  I didn't worry about corner sharpness.  I have a lovely example, pinned up in my office, that I didn't take, of the Milky Way over the ALMA observatory in Chile, shot from the salar, with the lights of a truck or possibly an antenna-mover, making a streak on the road up to the high site.

 

I recently shot a panorama from the roof of our building using an M 18/3.8 lens, wide open as it is intended to be used.  It's in the CL image thread somewhere and in my CL samples album on Flickr.  Here's another candid, also within the first page that I looked at, where the subjects are all over the frame:

 

C1010472 by scott kirkpatrick, on Flickr

 

(misfocussed, of course, damned AF)

 

 

This is exactly what I mean, though!  Ignoring the mis-focus part... 

 

The subjects certainly aren't dead center, but they aren't very close to the edge, either.  Things like the menu, the water glasses, the air vents and ceiling lights.  They are outside the depth of field (or would be if the darned AF hadn't messed up).  If they were all sharp, it would simply district rather than making a better picture.  Having perfect corner sharpness wouldn't hurt this image since DOF is going to take care of things anyway, but it wouldn't help in any way, either.  For this image you need really solid performance out to, say, 9mm off axis (for APS-C format).  After that?  Wouldn't make a difference either way.

[scott kirkpatrick]

Well, OK.  I agree that I don't often have scenes with a flat field of interest, but i do try to get close and fill the frame.  With the typical dinner shot, I try to make the stuff on the tabletop add some rhythm to the scene.  The point of the picture was the smile.  She had refused to be photographed, but borrowed my camera and shot pictures of everyone else, then finally consented for a very brief pose.

 

I was also responding to Thighslapper's comment that real photographers put the main subject in the center and don't need no stinkin' edges.  William Eggleston mischievously claims that his rule of composition is the Confederate battle flag, but he doesn't always follow that.  And the first picture that struck my eye had a little cactus fruit that had caught my attention, barely inside the frame.

Edited February 16, 2018 by scott kirkpatrick

[Jared]

Fair enough.  And the smile is great.

[tom0511]

so which is more fun to shoot? CL or M?

[jaapv]

I'll argue against Jared a bit.  I like to see little extraneous details in the edges and corners of the frame, and sometimes consciously frame things that way.  The categories that this effects might be called environmental portraits (as opposed to headshots), street, and landscape.  I also fairly often find a scene that exceeds my frame and requires an ad hoc panorama.  In that case I don't know where the two or three frames will be merged and need details to be resolved everywhere.  The 28/2.8 (both v1, now sold, and v2) is a favorite on my M10.  If possible, I use it at f/5.6 where everything works perfectly, but with the v2 I can go down to f/2.8 without concern about edge sharpness.  This would be rare for midday landscapes, but travel often runs into shadowy or end of the day scenes. 

Th CL has a panorama mode hidden in the menu - It merges quite overlapping -and you are in APS-C crop, so no soft details on the seams.

[scott kirkpatrick]

so which is more fun to shoot? CL or M?

I like variety, so I use both.  But the CL is certainly quicker in use if I can rely on AF.  If not, e.g. prefocused shots in a crowded place, then the M10 is quicker.

[LocalHero1953]

The CL is more fun to shoot, until it misses focus, or until you miss a shot because it can't find focus. Then the M is more fun! 

[jaapv]

 

I like variety, so I use both. But the CL is certainly quicker in use if I can rely on AF. If not, e.g. prefocused shots in a crowded place, then the M10 is quicker.

Face detection?

 

 

The CL is more fun to shoot, until it misses focus, or until you miss a shot because it can't find focus. Then the M is more fun!

Until you miss focus - or the photographer cannot find focus

[LocalHero1953]

.........

Until you miss focus - or the photographer cannot find focus

Oddly, that does not irritate me as much as the camera messing it up. When I make a decision to take a shot, I don't like the camera deciding not to because the focus spot has not turned green. When it works, of course, it is less effort than the M.

My point is, that there's no simple answer to "which one is more fun to shoot?"

[Ronazle]

Many years ago I did a very careful and extensive set of comparisons which involved mostly older (i.e., SM)  lenses; the outcome showed very gross differences of many characteristics. When I compared just the late production lenses the differences were pretty much the same as your results - yes, there were discernable edge differences at substantial blow-up (an antique term, no doubt) and very discernable differences in the corners. Late production lenses are indeed quite good. Regards, Ron

[tom0511]

I even sold my M10 since the CL is faster and more reliable to focus, and more suitable for longer focal length.

However I sometimes miss the simplicity of the M - specially the direct user interface (f-stop , focus, klick-klack)

[onasj]

I just want to say thank you to Jared for the meticulously designed experiments, and for manifesting in such a fun way the nerdy delight I suspect many of us share regarding these amazing instruments.

[Jared]

No problem! Glad you enjoyed.

[Advertisement]

Hallo,

Look here: Leica L-Mount

[M11 for me]

That is a great work. It is extremely interesting to read. Just a few unsorted thoughts:

 

Here its nearly about the question if APS-C renders less or the same as FF. And how does it compare to Sony (against A7 III we saw here) APS-C or Fuji (I am missing that part). 

 

How does the CL compare to the TL II? Does it not have the identical sensor? If so then the M10 sensor is not technologically older than the CL sensor (as we could read).

 

What is at the end of the day the advantage of the FF sensor besides faster lenses (following the equivalence)? Dynamic range seems to be almost even which is amazing.

 

If the CL had the focussing button on the rear side then it would be the upcoming wedding camera? 

 

Still I wonder if the equivalence of the 2 sensors has been fully followed in the tests above. I think that for comparison reason this would be a point to discuss.