The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

S2-70mm Brick Wall Test

dfarkas

Workshop Member
The well known Leica expert, Ken R. says;

"Leica's trademarks tell you nothing about the lens' design, its age or its history. To refer to a 90mm Summicron means nothing more than to refer to a Leica 90mm f/2 lens. Leica has made at least three totally different 90mm f/2 optical designs housed inside many cosmetically different barrels over the past fifty years, all called Summicron. "

Does that mean all S lenses are 2.5? (wow, even a 350mm?
Actaully ZEISS names like Planar and Sonnar did mean design types.

If Guy is right then Leica has found a way to standardize on both an aperture and design. Can't see it in the 35 to 350mm rage. There have GOT to be design differences.

Victor
Ken is right - the naming of Leica lenses only tells the max aperture, not the design of the lens, like it does in Zeiss, where a Planar could be f/2 or f/2.8 (for instance).

In fact, take two lenses, the 50 Summilux-M ASPH and the 75 APO-Summicron ASPH. One is a Lux and the other is a Cron, yet they are essentially the exact SAME design (the 75 has one less element). Yet, a 75 Lux and 50 Lux are vastly different. And, the current 50 Cron and the 75 Cron vastly different.

With the S lenses, the various lenses are individually unique optical designs. Here is an excerpt from my S2 review, information which was gathered in an interview with Peter Karbe, the head of optics design for Leica:

Many of the designs for the S lenses originated with R designs. The 35mm f/2.5 Summarit-S ASPH is based on the 19mm Elmarit-R (a lens that I loved on my DMR), but includes an asphere in the front optical group to reduce aberrations. The 120mm f/2.5 APO Macro-Summarit-S used the incredible 100mm APO-Macro-Elmarit-R as a foundation, but incorporated a floating element to optimize performance at all distances from close-focus to infinity (and increased maximum aperture by 1/2 a stop to boot). The 180mm f/3.5 APO-Tele-Elmar-S owes its heritage to the legendary 180mm APO-Summicron-R. Contrary to the other three lenses, the 70mm f/2.5 Summarit-S ASPH is an entirely new design, but has a similar signature Karbe modified double gauss design like the 50 Summilux-M ASPH.

I actually forgot to add at the time of writing that the 70mm Summarit-S also uses a floating element. As Jack pointed out a few posts back, lenses are optimized for a certain distance. By using a floating element, this single distance can be extended over more of the focus range, improving performance from minimum focus distance to infinity.

David
 

thomas

New member
By using a floating element, this single distance can be extended over more of the focus range, improving performance from minimum focus distance to infinity.
very interessting! Now more than ever I am curious to see that nice theory translated to a real world image... there is no wide shot on the internet that shows satisfactory results at the edges (stopped down).
Mr. Karbe also said the S lenses are sharp to the edges wide open.
I really don't care whether a fast lens wide open is sharp at the edges or not (actually the falloff at large apertures can be very nice)... but that's what Leica promised (amongst other things to justify the steep prices).
 

Guy Mancuso

Administrator, Instructor
Based on the design but not exactly like it. As soon as you throw another element in there it is a different design. End of the day they are a Peter Karbe design. They call them Summarits.
 
Sorry to start a thread and disappear - it has been very busy and long day with only brief moments to look at the thread via iPhone.

I agree with you Guy, it does look like barrel distortion. I have concentrated on the upper portion of the frame and noticed it at time of capture. However, looking at the image again it is obvious the bottom portion of the frame show a similar effect.

I wished I had printed some sort of resolution targets to tape to the wall in the corners to make it easier to judge sharpness versus the subtle textures and variations of the bricks. However, I didn't and probably never will.

Thomas, I would love to see examples of what you consider sharp corners. I admit that my experience in this area may be more limited than yours so I would truly like to be educated. I am not trying to be confrontational here - I genuinely want to learn new things everyday.
 
S

Shelby Lewis

Guest
I think the corners are pretty danged nice... a slight lessening of contrast and a bit of falloff give the perception of less sharpness, but overall these are really nice. I do find the barrel a bit surprising, especially on a normal lens.

Overall, really nice though.

Guy, that 80D really continues to impress me. I'm presently picking up an RZ system for the "look" of the glass in tandem with a 9 micron sensor, but man do those D lenses on the 645 system impress! Such contrast and sharpness.
 

Jack

Sr. Administrator
Staff member
Guy,
WTF? your upper corners are black LOL
Actually, if you look closer on a good monitor, they are not black -- in fact you can see shadow detail pretty well!

Guy, that was my 80D on your P40+ right?
 

Guy Mancuso

Administrator, Instructor
That was yours and Jack is right you can actually see the roof inside through the window . Maybe what 9 stops at least:D:D:D

Gotta love MF.

My 40D , you won't even let me breath on your 65.:ROTFL::ROTFL::ROTFL:
 

Guy Mancuso

Administrator, Instructor
The 80D is a very good lens and for what 1200 bucks hard to put your nose in the air on. Actually my 80d might even be better.:eek:
 

thomas

New member
Thomas, I would love to see examples of what you consider sharp corners. I admit that my experience in this area may be more limited than yours so I would truly like to be educated. I am not trying to be confrontational here - I genuinely want to learn new things everyday.
I've once decided not to post my photos in forums... and would like to leave it at that. When my DB is back from service :angry: sometime next week I'll take a comparable shot with the Contax 2.0/80 & P45 and going to post it.
However... what I am referring to is simply that parts of the scene being in the same image plane should look somehow coherent when the lens is stopped down.
 
Thomas, I only posted the wide open crops from my brick wall test. I did test across the full range of apertures and the results were as expected - no surprises.
 

thomas

New member
Thomas, I would love to see examples of what you consider sharp corners.
FWIW here's the Contax 2.0/80 & P45.
Distance to the building is around 70 meters or so.
Sharpening in Capture One set to "Presharpening 1" (the setting Guy is using regularly).
In terms of comparision to the S2/70mm this is quite useless, however you were asking for an example.

f5.6 sharpness falloff is clearly visible.
However the falloff is somehow "smoother" than that of the S2/70mm.
The edges of the S2/70mm look like there is some motion blur. Somehow "turbulent". As if the crop of the side and the crop of the center do not originate from the same capture (see attachment). Whereas the Contax 80mm at 5.6 is a bit weak at the edges but it's still a coherent look … Besides you can apply sharpness falloff and it looks quite good whereas the S2/70mm is so coarse at the edges that additional sharpening won't help at all.

Here's the scene:


Respectively this is the scene (I've stitched bottom and sky to keep the camera leveled for the main object in the scene…):


And here are the crops of the Contax 2.0/80 at f 5.6, f8 and f11 (from the leveled capture).
At f8 and f11 the center and the side show a homogeneous image… IMO.
I think you have to be logged in to see them in full size.
Screenshots from Capture One.

f5.6, center: http://forum.getdpi.com/gallery/files/2/6/7/8/01_56_center_original.jpg
f5.6, center/side: http://forum.getdpi.com/gallery/files/2/6/7/8/02_56_center_side_original.jpg
f5.6, corner: http://forum.getdpi.com/gallery/files/2/6/7/8/03_56_corner_original.jpg
f8, center: http://forum.getdpi.com/gallery/files/2/6/7/8/04_8_center_original.jpg
f8, center/side: http://forum.getdpi.com/gallery/files/2/6/7/8/05_80_center_side_original.jpg
f8, corner: http://forum.getdpi.com/gallery/files/2/6/7/8/06_80_corner_original.jpg
f11, center: http://forum.getdpi.com/gallery/files/2/6/7/8/07_11_center_original.jpg
f11, center/side http://forum.getdpi.com/gallery/files/2/6/7/8/08_11_center_side_original.jpg
f11, corner: http://forum.getdpi.com/gallery/files/2/6/7/8/09_11_corner_original.jpg
 
Last edited:

fotografz

Well-known member
Can you really check for Barrel Distortion using a hand made brick wall?

When I shot some real world stuff with the S2 and 70 I saw a bit of barrel, but nothing unexpected despite the statements from Leica ... wouldn't use it for copy work maybe, but it seemed fine ... I was more concerned about the lack of any special character compared to other fine offerings from other MFD makers. Setting aside the M line ... I think the later R lens line displayed more image character ... don't know why ... maybe it's just a small sensor thing or something. But technical perfection isn't everything.

It's too bad that Leica put so much emphasis on their own marketing hype concerning digital optical corrections. Here they have a fully integrated camera with full communications between all elements of the image making chain that totally lends itself to such digital corrections ... now they would have to reverse themselves to take advantage of it.

-Marc
 
...However the falloff is somehow "smoother" than that of the S2/70mm.
The edges of the S2/70mm look like there is some motion blur. Somehow "turbulent". As if the crop of the side and the crop of the center do not originate from the same capture (see attachment). Whereas the Contax 80mm at 5.6 is a bit weak at the edges but it's still a coherent look … Besides you can apply sharpness falloff and it looks quite good whereas the S2/70mm is so coarse at the edges that additional sharpening won't help at all...
Thomas, thank you for posting your examples. I am still trying to digest and comprehend what you are saying about the images. However, I cannot agree with the quoted portion of your post above regarding "motion blur" and "turbulent" corners of the S2/70mm. Of course I had the advantage of seeing the bricks in my test up close - they are not uniform. I was quite pleased with the corner performance (e.g., representation of actual scene) especially considering the lens was wide open. However, I will try to re-evaluate with an open mind to see if I overlooked something (e.g., I will go back and compare the actual bricks to the photos to see if I can see what you are talking about).

Mark
 
Marc, there was definitely some irregularities in the mortar thickness in the brick rows which may be exaggerating the amount of barrel distortion, but there does seem to be some distortion present.

Technically, I think it is possible for Leica to include optical corrections and still remain true their marketing hype. In other words they could include optical corrections in the firmware (ala M8/M9) and still be able to claim (as marketing hype) that no special software is required for post process lens corrections. I know this is splitting hairs.

Mark
 

thomas

New member
I cannot agree with the quoted portion of your post above regarding "motion blur" and "turbulent" corners of the S2/70mm. Of course I had the advantage of seeing the bricks in my test up close - they are not uniform. I was quite pleased with the corner performance
Your brick wall shot looks fine. I am talking about Tim's image... see the attachment at the end of my post (resp. see the image on his site: http://tashley1.zenfolio.com/p627107524/h36cc304#h2eb5b8e2 ). From the very beginning I was talking about shots at wider distances and stopped down. Not about close shots wide open.
 

fotografz

Well-known member
Marc, there was definitely some irregularities in the mortar thickness in the brick rows which may be exaggerating the amount of barrel distortion, but there does seem to be some distortion present.

Technically, I think it is possible for Leica to include optical corrections and still remain true their marketing hype. In other words they could include optical corrections in the firmware (ala M8/M9) and still be able to claim (as marketing hype) that no special software is required for post process lens corrections. I know this is splitting hairs.

Mark
The difference is that you can turn it off in software if you want. If it was a selectable feature in camera that would be an alert that the lens isn't "perfect" ... so what's the difference?

I did notice some barrel with the S 70 but it was on the sides ... with next to none that I could detect on the horizontal long side edges. Which is why I questioned the brick wall as target.

The real measure will be the wide angles.

It is amazing to watch the Hassey DAC software corrections as you toggle back and forth on an image ... but there have been cases where the perfect correction can look ... uh, well, unnatural. It's odd. Doesn't happen to often, so I just leave the software DAC box checked.

-Marc
 

Geoff

Well-known member
FWIW, was in NYC yesterday and visited Foto Care and handled an S2. Pleasant surprise. The camera is remarkable in the hand, feeling great, smooth and refined. It is a very light camera for all that it does. I am glad that my camera gear is committed in other directions. Thoughts of selling everything and one camera....were wisely disregarded. It is a much finer product than initially expected. THeir decision to abandon the R range for this newer approach was quite hard to understand before, but makes more sense now having seen this.
 

Steve Hendrix

Well-known member
The difference is that you can turn it off in software if you want. If it was a selectable feature in camera that would be an alert that the lens isn't "perfect" ... so what's the difference?

I did notice some barrel with the S 70 but it was on the sides ... with next to none that I could detect on the horizontal long side edges. Which is why I questioned the brick wall as target.

The real measure will be the wide angles.

It is amazing to watch the Hassey DAC software corrections as you toggle back and forth on an image ... but there have been cases where the perfect correction can look ... uh, well, unnatural. It's odd. Doesn't happen to often, so I just leave the software DAC box checked.

-Marc
Yes, I remember looking at the lenses before and after and it was indeed amazing how you would be looking at an image and then apply the corrections and realize how much distortion you were not aware of. Of course, this was mostly on the wider lenses. On the 80, 120, etc, you could barely tell a difference.

Still, something about the tinkerer in me always wanted to play with that correction somehow (honestly!). Instead of completely distorted to zero distortion, I wanted to be able to introduce (or allow back in) some of that wompiness. Not in architectural shots, naturally, but other stuff, yeah. And I might not have even liked the result and just pushed it back to complete correction, but I still wanted to be able to. I know, ridiculous.


Steve Hendrix
 

Jack

Sr. Administrator
Staff member
It is amazing to watch the Hassey DAC software corrections as you toggle back and forth on an image ... but there have been cases where the perfect correction can look ... uh, well, unnatural. It's odd. Doesn't happen to often, so I just leave the software DAC box checked.
I've noticed the same behavior with the C1 lens corrections for my Phase 28mm lens. Clearly the corrections make the image more optically "perfect" corner-to-corner, but I find that most of the time I prefer the image without them so leave it off. Weird indeed...
 

thomas

New member
Weird indeed...
not so weird actually.
We are used to some kind of falloff.
For instance I almost always reduce LCC based light falloff a little bit as at 100% the corners look too bright (they are not... but they look like). Not only to me, also to other viewers.
 
Top