[New test photos released] Leica Summicron 35/2 Eight Element copy made in China

davidde1000 said:
There are numerous "focus shift" grumblings with fast 35mm focal length rangefinder lenses, but I really think that these types of lenses require the users to know how to use them -- and manufacturers have not generally been good throughout history about explaining how to use these lenses. When lenses like the V1 35/2 Summicron and the 35/1.8 W-Nikkor came out, they represented some of the most (or "the most") advanced 35mm RF lenses made to that point. They were extraordinarily expensive. Leica and Nikon already had very competent, slightly slower 35s in the form of the Summaron 35/2.8 and W-Nikkor 35/2.5. The fast 35s shifted, the slower 35s generally did not.
Click to expand...

I just tested my LLL again, also against my Canon 35/2 LTM, which was initially released in the early '60s and a few years after the Cron v1. I didn't see focus shift with the Canon. If it shifts, it's within DoF.

davidde1000 said:
Today, many people who are new or relatively new to RF photography get initially flabbergasted by the fact that their Summiluxes and pre-ASPH Summicrons "shift." If you are coming from a film or digital SLR, this comes as a shock as the vast majority of modern SLR lenses have no noticeable shift (and your mirrorless camera can now automatically compensate for any shift in the newest lenses anyway). While many people did not notice this too much during the film era, digital mirrorless is unforgiving for any lack of precise focus on the sensor plane. So yes, your V1-V4 Summicrons and older Summiluxes will look out of focus when correctly focused with the rangefinder at f/5.6 on digital (and also on film) because that is how the lens behaves. What is really great about digital is that with liveview, you can actually focus absolutely spot on at f/4-8 with fast 35s, whereas during the film era, you may have had to guess or manually compensate in some way. I do not really understand why one would use a digital M without liveview -- that's the whole point! 100% accurate focus every single time without worrying about the primitive native focusing system.
Click to expand...

At least for me, it's not a shock to see focus shift, especially from an older design. It's the reason I tested the LLL to better understand how it behaves and therefore better understand how such characteristics may affect my images. Unfortunately I don't have firsthand experience with the Leica original to know whether or not it behaves similarly. Hence my uncertainty about my LLL copy. You're saying the 35 Cron v1 shifts, but others have said they don't see this. I still believe it somewhat depends on how images are viewed. An 8x10" print (whether analog or digital origin) will not show focus shift as much as a much larger image (whether analog or digital) unless relative viewing distance is maintained. Of course the temptation with large display images is to view them closely to pick out the details (i.e. pixel peeping digital images at 100%).

When discussing focus shift of the LLL (my copy) it should be understood that this behavior is hand in hand with its very specific field curvature characteristics.

Some lenses have fairly flat planes of focus. If they exhibit focus shift, the entire plane shifts consistently behind the intended point of focus when stopped down. Achieving good across-frame sharpness is a 'simple' matter of refocusing at the taking aperture, or with a rangefinder, intentionally front-focusing the RF patch.

The LLL (or at least my copy, but I think it's a characteristic of all, given it's documented in the spec sheet supplied with the lens by LLL) exhibits wavy field curvature in the shape of a 'w.'

Wide open, the curvature is a small 'w' and the plane of focus appears fairly flat. Stopped down, it becomes a big 'W.' 🙂 Its amplitude increases with the center 'peak' of the W pushing behind the intended point of focus and outside DoF (my copy). The rest of the W zigzags through the indented plane of focus and on average results in good enough focus across most of the image.

This behavior is bad for critical central sharpness at f/4-5.6 on a rangefinder, but if important subject content isn't in the very center of the image, the majority of the image is in focus at that distance. Meaning, this lens is potentially quite good for focus & recompose.

The downside of using live view or a mirrorless camera with this kind of field curvature and focus shift behavior is that when proper focus in the very center of the image is set, (pulling the peak of the W towards the camera), the entire W of field curvature, especially the bottom two valleys that comprise the mid-frame zone, pull forward enough that some or most of it is out the intended plane of focus and out of DoF. This is fine if you only want critical sharpness in the center of the image, such a person at a nearer distance against a far background, which is exactly the scenario that gave me fits when rangefinder focusing. But it's not good if you want good across-frame sharpness, such a group photo of a row of people (or brick walls 🙂 ).

Since it's a lens originally intended to be used with rangefinder focusing, I think the designers understood its specific focus shift consequences and calibrated the lens so that average sharpness across the frame remained good, at the expense of the very center.

Bottom line, IMO, based on how my copy performs, this is not a good lens to choose for group photos of people in a row, brick walls and other similar content where you want a relatively flat plane of focus.

But, this field curvature behavior also provides interesting opportunities. It's great if shooting down an alley, forest path, etc. Distant things in the center of the image appear in focus along with nearer objects in the midfield area. It's also good for placing a subject in the rule of thirds zone (focus & recompose).


davidde1000 said:
Although I have not handled any of these replicas, copying the original design will generally copy the original focus shift (like the UC-Hexanon's copy of the W-Nikkor copied almost precisely the shift of the original lens). The V1 35/2 Summicron was designed for pros who "needed" f/2 over f/2.8. The technology required to create an f/2 lens that focused accurately at that aperture required compromises. One of those compromises was that if accurate on the rangefinder somewhere at f/2-f/2.4-f/2.8, it was not going to be accurate at mid apertures. You would either use the Summicron wide open, and if you need to capture a larger scene, at f/8-11. I do not think that older Summicrons and Summiluxes were considered or are to be considered "general purpose" 35mm lenses. That is what the Summaron was for.
Click to expand...

I agree, this lens is looking more and more to me like one I'll use at f/2-2.8 and at f/8. This is also how I use other rangefinder lenses with wavy field curvature, such as the 21 & 50 Lux ASPHs. But with those lenses, the center point of focus does not shift like the LLL, rather, the midfield area is weak/nervous at around 3 stops from wide open until cleaned up by DoF at f/8. I find those lenses frustrating to use for focus & recompose when intending to put subjects in the rule of thirds zone.
 
rscheffler said:
I just tested my LLL again, also against my Canon 35/2 LTM, which was initially released in the early '60s and a few years after the Cron v1. I didn't see focus shift with the Canon. If it shifts, it's within DoF.
Click to expand...

Appreciate the response, and I think your insights on these issues are correct. And the center versus corners effects at mid-aperture makes sense. However, for me, an average amateur shooter, I find that it is far easier and more convenient to use "general purpose" 35s like the Summaron or the VC 35/2.5 because they lack any significant shifting that may come as a surprise later. Before I understood the characteristics of older fast 35s, I took so many pics on film with a IV Summicron and UC Hex that I "thought" were in precise focus, and looked from a distance to be fine, but then on high rez scans, I could see that mid aperture shots in the center were not quite nailed. There is probably a reason that Leica designed FLE lenses, and it was probably because so many people complained about the older fast lenses, even the ASPH 35/2, on digital.

The DOF "saving" images never quite provides me an adequate comfort level. Of course, for me, DOF only means that the plane of focus may be sufficient for human eyes at a certain resolution or enlargement size, but not that the image is actually in precise focus. I mean, try missing the focus a little bit on a digital mirrorless at f/8 and enlarge to max pixels on a computer and the missed focus is plain as day. For me, on rangefinders and not on liveview, the older fast 35s are f/2-2.4 or f/11 lenses, which is fine because there is tremendous value in f/2 lenses that perform well and create nice separation. And there may be no really great reason to shoot expensive 35/2s at mid-apertures as it does not create good separation while at the same time is generally not at the max resolution of the lens. For everyday use, I would rather use a slightly slower lens that provides better focus control throughout the range.
 
davidde1000 said:
Before I understood the characteristics of older fast 35s, I took so many pics on film with a IV Summicron and UC Hex that I "thought" were in precise focus, and looked from a distance to be fine, but then on high rez scans, I could see that mid aperture shots in the center were not quite nailed.
Click to expand...

I understand what you're getting at. The above snippet IMO actually confirms that those lenses performed within their design specifications/intent. Film/analog printed to a certain size and viewed at 'normal' viewing distances would mask focus shift. As would viewing digital images at smaller sizes.

On digital this would be akin to shooting (or downscaling) at 6-12MP vs. 24, 36 or 60, which as the resolution increases, will better resolve any given lens's weaknesses.

I have a lot of film images that if I critically reexamined the way I technically inspect digital images now, would certainly be technical disappointments. But I need to remind myself that it's a separate consideration from the value I perceive about the actual image content. Perhaps it's now too easy to get hung up on technical performance (or maybe it's always been that way 🙂 ).
 
The last few posts above have been illuminating thank you, particularly with regard to preferred apertures when using a lens with focus shift, and the pixel peeping vs the bigger picture.
 
Was shooting some backlit quince blossoms and got some funky flare at certain angles. Sun in the frame didn't always provoke flare.

 


Just to be clear, I think this flare is interesting and could provide some creative opportunities. It's similar to what can be provoked by the Canon 35/2 LTM (and perhaps other similar lenses of that era) and is a reason I got the Canon.
 
Haha, yeah... sun is in the frame.

Here are some less severe examples:





And with a branch in the way, no flare:



With the sun out of the frame, the flare changes:





IMO this lens flares far more interestingly than the 50 Cron Rigid because it's not just veiling flare.

Here's the 50 Cron Rigid:



And a more subdued example:

 
What f-stop was used for these flair shots with the LLL replica. A few months ago I posted a flair shot when the replica was used wide open (it was considerable) but stopped down to f2.8 made a significant difference in reducing it. I reposted this flair shot a few pages back.

Dave (D&A)
 
An early shot with my new LLL 8 Elements. Haven't had a chance to do any scientific testing of the focus shift/depth of field, etc., but I'm liking the shots so far. Here's a shot from today of a pair of Transformer sculptures in Georgetown, which the neighbors are suing to have removed from the tony neighborhood near the Exorcist steps.

Transformers by Brett Davis, on Flickr
 
D&A said:
What f-stop was used for these flair shots with the LLL replica. A few months ago I posted a flair shot when the replica was used wide open (it was considerable) but stopped down to f2.8 made a significant difference in reducing it. I reposted this flair shot a few pages back.

Dave (D&A)
Click to expand...

Those images were all wide open. I haven't yet tried provoking flare stopped down. This flare experiment happened spontaneously while photographing the blossoms and my intent was to work wide open to realize the lens's signature rendering.
 
Very quick comparison between an original mint 8e, and the LLL. Seems to me the LLL lets about 1/3rd of a stop more light in.

Shot at f2 indoors, and at f5.6 outdoors, the LLL seems sharper/has larger depth of field? I focused on the same thing (the little yellow excavator)

https://www.moerman.org/lll/
 
Test photos @ f8. With the snow & rain still waiting for some light with strong shadows outside. Very satisfied with the microcontrast. M4 / 35mm replica / FP4+ in Pyrocat/ prints on Ilford Coldtone FB /Formulary 130. I'll be happy with this one for my daily 35mm user. I'll keep the much smaller Elmar 35 f3.5 mounted on the CL

51169949670_2bb6b945c6.jpg


51169095943_a33a08e30e.jpg


51168184847_059fe80018.jpg

Flickr
 
You must log in or register to reply here.

Similar threads

J
  • Article Article
Celebrating the Leica’s 100th Birthday: Here’s how to do it yourself! What an EVF Leica M should look like and why it may not be made now.
2
Replies
21
Views
1K
boojum
boojum
A
A day out with the SL2-S
Replies
7
Views
428
Archiver
A
TenEleven
Lomography x Zenit Mercury-2 Lens 35mm f2
Replies
3
Views
131
TenEleven
TenEleven
Photar
Who has a Leica MP Hammertone?
Replies
1
Views
179
Photar
Photar
J
  • Featured
  • Article Article
What exactly is a point & shoot camera? Glad you asked!The Kodak of 1888 may have been the first, and even Leica got into the act!
Replies
16
Views
434
Jason Schneider
J
Back
Top Bottom