I did two tests here: infinity and about a meter away from a subject with out of focus background and foreground.
For each I took photos at all apertures, but picked the ones I tend to use the most for comparison (wide open, f2.8 and f5.6), selecting specific areas of each scenario to have a better idea on how the lens behaves.
As previously mentioned on the thread, I have been shooting with a Voigtlander 40mm f1.2 for a few years and it has been my go to lens, now challenged by the 35mm f1.5 - which caught my interest being so tiny and light and I have been enjoying using a lot. So I did also a comparison between the two.
I used a tripod for all shots, with a 2sec timer, trying to sequence as fast as possible to minimise lighting variation but still it's present. On the 1m wide open shots I did my best to get perfect focus using live view, to bypass any rangefinder calibration variables. The focus point was the TR on CHARTREUSE. For the wide open shots, the same applies - on the 35mm, it's the hard stop; on the 40mm my copy reaches infinity slightly before the hard stop (which doesn't affect actual use, to be quite frank, I've gotten used to it). Both lenses with spot clean B+W Clear filters (which I never remove really). Closeups are from the 200% view on Lightroom Classic.
First time doing this kind of testing so be gentle open to suggestions and if other aperture values would be interesting to see, let me know and I can post them here as well.
To be quite frank doing this comparison didn't bring me immediately to any decisive conclusions, the 40/1.2 even at 1.5 looks softer than the 35/1.5 (the difference between 1.2 and 1.4 is very subtle), but due to the different focal lengths it's not a fair and square comparison. Both lenses deliver an out of focus rendering that pleases me, and are sharp enough to my taste even wide open. In the end, between the two I feel it's more about which focal length and weight/ergonomy (even more than actual size) is preferred.
Honestly who cares about the mid-frame sharpness wide open at infinity? If you're shooting a landscape you're probably stopped down. If you're shooting wide open your subject is probably pretty centered, i.e. a portrait.
I've been shooting this lens normally on film and it's stunning. I've used Adox CMS20II and Adox CHS100II. When shooting the 20II on a tripod, I'm stopping down to 4.0-5.6. The Adox was incredibly detailed. I enlarged the film to 20" and there was no grain, all detail. When using the conventional 100II I shot between wide open apertures and stopped down, so normal stuff. In no situation did I have an off center subject at infinity distance that needed to be tack sharp wide open...because that's a weird stress test to go for.
marcelography wrote:
I did two tests here: infinity and about a meter away from a subject with out of focus background and foreground.
For each I took photos at all apertures, but picked the ones I tend to use the most for comparison (wide open, f2.8 and f5.6), selecting specific areas of each scenario to have a better idea on how the lens behaves.
As previously mentioned on the thread, I have been shooting with a Voigtlander 40mm f1.2 for a few years and it has been my go to lens, now challenged by the 35mm f1.5 - which caught my interest being so tiny and light and I have been enjoying using a lot. So I did also a comparison between the two.
I used a tripod for all shots, with a 2sec timer, trying to sequence as fast as possible to minimise lighting variation but still it's present. On the 1m wide open shots I did my best to get perfect focus using live view, to bypass any rangefinder calibration variables. The focus point was the TR on CHARTREUSE. For the wide open shots, the same applies - on the 35mm, it's the hard stop; on the 40mm my copy reaches infinity slightly before the hard stop (which doesn't affect actual use, to be quite frank, I've gotten used to it). Both lenses with spot clean B+W Clear filters (which I never remove really). Closeups are from the 200% view on Lightroom Classic.
First time doing this kind of testing so be gentle open to suggestions and if other aperture values would be interesting to see, let me know and I can post them here as well.
To be quite frank doing this comparison didn't bring me immediately to any decisive conclusions, the 40/1.2 even at 1.5 looks softer than the 35/1.5 (the difference between 1.2 and 1.4 is very subtle), but due to the different focal lengths it's not a fair and square comparison. Both lenses deliver an out of focus rendering that pleases me, and are sharp enough to my taste even wide open. In the end, between the two I feel it's more about which focal length and weight/ergonomy (even more than actual size) is preferred.
Thank you for this test. For me, I'm not a pro, plus my eyes might be too old to notice, but they both look very similar, even I would say that 35mm is kinda better looking (IQ wise). Performance, plus its size and ergonomics compared to 40mm, I would pick 35mm. That is why I already sold my 40mm 1.2 SE. It was sad to see it go, but its time for this little 35mm to connect with my A7C.
speedgraphic wrote:
Honestly who cares about the mid-frame sharpness wide open at infinity? If you're shooting a landscape you're probably stopped down. If you're shooting wide open your subject is probably pretty centered, i.e. a portrait.
I've been shooting this lens normally on film and it's stunning. I've used Adox CMS20II and Adox CHS100II. When shooting the 20II on a tripod, I'm stopping down to 4.0-5.6. The Adox was incredibly detailed. I enlarged the film to 20" and there was no grain, all detail. When using the conventional 100II I shot between wide open apertures and stopped down, so normal stuff. In no situation did I have an off center subject at infinity distance that needed to be tack sharp wide open...because that's a weird stress test to go for. ...Show more →
Was only testing to verify there was no midframe weakness stopped down at infinity. Lenses can be weak in the midframe stopped down, and I wasn’t interested in owning another one like that. My first copy of the CV 50 f/1 had focus shift at infinity and the midframe at infinity was a joke stopped down.
Also not everyone puts a portrait subject’s head in the middle of the frame, that’s a ridiculous assumption. Midframe weakness at portrait distance wide open is important to many people even if you personally don’t need it.
highdesertmesa wrote:
Also not everyone puts a portrait subject’s head in the middle of the frame, that’s a ridiculous assumption. Midframe weakness at portrait distance wide open is important to many people even if you personally don’t need it.
The midzone softness wide open can be seen very clearly near the end of Jack Takahashi's review (which has been linked to several times already in this thread): https://jacktaka.com/voigtlander-35mm-f15
Even for someone as undemanding as me (I've never checked any of my lenses for centering and am happy with most aberrations except corner smearing), this would be a dealbreaker.
bjhurley wrote:
The midzone softness wide open can be seen very clearly near the end of Jack Takahashi's review (which has been linked to several times already in this thread): https://jacktaka.com/voigtlander-35mm-f15
Yeah, his review I think was what started the whole 35/1.5 midframegate. As he mentions, he did have the bad luck to get a lemon copy of the lens with side by side performance variation. Even at the center at f1.5 my copy (bought used from BH in Jan 2023) shows considerably better performance.
So I think it's a bit unfair to say that the lens overall has a bad performance, only that the quality control is not so tight - which itself is something to pay attention to and an issue, obviously.
I'm 100% happy with mine, to be quite frank. It just works so nicely, handling and image quality wise. Versus the 40/1.2 the weight difference is very perceivable and it's more forgiving being a 35, so nailing focus wide open is a bit more comfortable. I love how the out of focus elements are rendered, pretty smooth, it kinda feels/looks like it's an even faster lens than f1.4 (like the 35/1.2 Nokton).
speedgraphic wrote:
Honestly who cares about the mid-frame sharpness wide open at infinity? If you're shooting a landscape you're probably stopped down. If you're shooting wide open your subject is probably pretty centered, i.e. a portrait.
I do care, and to me, midframe is often more important than center. Because I would (almost) never have the subject exactly centered in the frame.
I also shoot a lot at low light and use a large aperture only to let as much light as possible in; not necessarily to get shallow DOF. So I appreciate lenses that are reasonably sharp all over the frame at large apertures, even at infinity.
bjhurley wrote:
The midzone softness wide open can be seen very clearly near the end of Jack Takahashi's review (which has been linked to several times already in this thread): https://jacktaka.com/voigtlander-35mm-f15
Even for someone as undemanding as me (I've never checked any of my lenses for centering and am happy with most aberrations except corner smearing), this would be a dealbreaker.
Midframe looks great on my copy at typical portrait distances. Past 3.5m there is some off center weakness wide open, but that’s it.
I wish that once someone recognizes they tested and reviewed a bad copy they would remove the review.
highdesertmesa wrote:
Midframe looks great on my copy at typical portrait distances. Past 3.5m there is some off center weakness wide open, but that’s it.
I wish that once someone recognizes they tested and reviewed a bad copy they would remove the review.
That's exactly my thoughts. There are not many reviews of this lens and when I was looking for one, I hit that review straight away and couldn't believe that this brand new design lens would be such a disappointment, like reviewer suggested. So, now seeing guys here showing more samples and seeing that its not that bad is a huge up lift, if you know what I mean.
That review should be either taken off, or re-done with different copy.
I thought Fred's examples showed mid-zone softness as well, though.
It seems like Cosina should spend more effort on QA/QC if there's this much variation across copies. I have the VM 35/2 Ultron, which I really like, but it came with a lot of exposed grease inside (some reviewers have remarked on this; one of them said the grease actually got onto his camera's sensor glass), and it doesn't always focus smoothly despite all that grease.
bjhurley wrote:
I thought Fred's examples showed mid-zone softness as well, though.
It seems like Cosina should spend more effort on QA/QC if there's this much variation across copies. I have the VM 35/2 Ultron, which I really like, but it came with a lot of exposed grease inside (some reviewers have remarked on this; one of them said the grease actually got onto his camera's sensor glass), and it doesn't always focus smoothly despite all that grease.
For sure the blame is on the manufacturer here, and it's a shame that reviewers get bad copies, sometimes multiple bad copies.
I'm on my second try at the 28 Ultron II and am giving up (after having had a good copy last year that I sold). Both copies hit infinity on the rangefinder before the hard stop. I think I may be pretty much done with rangefinder cameras. It's so stressful to try to not only get a good copy of the lens optically but to also get one that performs correctly with the rangefinder.
highdesertmesa wrote:
For sure the blame is on the manufacturer here, and it's a shame that reviewers get bad copies, sometimes multiple bad copies.
I'm on my second try at the 28 Ultron II and am giving up (after having had a good copy last year that I sold). Both copies hit infinity on the rangefinder before the hard stop. I think I may be pretty much done with rangefinder cameras. It's so stressful to try to not only get a good copy of the lens optically but to also get one that performs correctly with the rangefinder.
I am fed up with RF’s more and more, aside from focus shift, price, lack of IBIS, calibration and being limited outside of 28-50mm without extra/care and gear. I absolutely love shooting them, film and digital, I understand how singular they are, but I just don’t think the juice is worth the squeeze for me at this point. My modded MILC’s will do for me.
shooting with rangefinders is a PITA sometimes, more modern cameras with AF that don't depend on calibrations and tight tolerances are way less troublesome to shoot with. I also shoot with Sony to balance that...
marcelography wrote:
shooting with rangefinders is a PITA sometimes, more modern cameras with AF that don't depend on calibrations and tight tolerances are way less troublesome to shoot with. I also shoot with Sony to balance that...
I actually love shooting the rangefinder, it's the struggle of getting lenses that are perfectly dialed-in across the board:
– Does the lens hit infinity at the hard stop and does the rangefinder agree?
– Is rangefinder focus sharp at close distances wide open?
– How does midframe look?
– How do the corners look?
– Is there focus shift?
– Is the lens decentered/tilt/swing/etc.?
– Is the focus action smooth without any rough spots?
– Is the aperture ring tightness/resistance ok?
I would definitely pay up to 50% more for CV lenses if they had much more rigorous quality control and thoroughly tested each lens coming off the line. Maybe it's naive of me to think that would only cost 50% more, though.
speedgraphic wrote:
Well, I came to a pixel peeping thread and complained about pixel peeping, what else should I expect?
I'm just glad I get to use a great lens to its fullest and not worry about detail on a 40" print when I shot wide open.
No, the problem you injected into the thread was you're a film shooter and don't care. That's great, but the rest of us are interested in midframe performance. The kind of midframe softness I have an aversion to impacts print sizes much smaller than 40". You putting the subject's head squarely in the center of the frame, your 35mm film grain and the inability to shoot wide open in bright light have you shielded from such impacts. Don't make this about pixel peeping when you said the solution was to center the subject.
It's a moot discussion anyway since IMO a good copy of this lens performs very well wide open in the midframe for portraits.
highdesertmesa wrote:
No, the problem you injected into the thread was you're a film shooter and don't care.
Ahem, Adox CMS20II Pro resolves around 800lp/mm. Much higher resolution than any digital sensor available. I scanned the frames with an Eversmart Supreme II for evaluation. I also have made optical prints on 16x20 paper. If there is any softness, it's not apparent. Maybe there is some sample variation and if it's bad enough, buyers should seek warranty replacement.
Jack Takahashi here, just thought I'd chime in with some thoughts given the controversy about this lens:
-Yes, my lens was not perfectly centered, but the midframe performance I saw is the same as Fred's tests (and I believe he tried 3 copies and used an optimal one for his tests). If you look at his rendering comparison vs the 35 FLE the difference in sharpness is obvious.
-While I was not happy with this lens, for many people the sharpness is good enough. Will the mid-frame softness show up on instagram? No. Will it show up on film? Not really. Will it show up on 11x14 prints? Probably not. Sharpness needs are entirely subjective.
fwdesign wrote:
Jack Takahashi here, just thought I'd chime in with some thoughts given the controversy about this lens:
-Yes, my lens was not perfectly centered, but the midframe performance I saw is the same as Fred's tests (and I believe he tried 3 copies and used an optimal one for his tests). If you look at his rendering comparison vs the 35 FLE the difference in sharpness is obvious.
-While I was not happy with this lens, for many people the sharpness is good enough. Will the mid-frame softness show up on instagram? No. Will it show up on film? Not really. Will it show up on 11x14 prints? Probably not. Sharpness needs are entirely subjective....Show more →
I think some of us are wondering if the early copies, including yours and all of Fred’s may have been sub-optimal, at least at infinity. Did you see my test shots with the M11?
highdesertmesa wrote:
Also not everyone puts a portrait subject’s head in the middle of the frame, that’s a ridiculous assumption. Midframe weakness at portrait distance wide open is important to many people even if you personally don’t need it.
bjhurley wrote:
The midzone softness wide open can be seen very clearly near the end of Jack Takahashi's review (which has been linked to several times already in this thread): https://jacktaka.com/voigtlander-35mm-f15
Even for someone as undemanding as me (I've never checked any of my lenses for centering and am happy with most aberrations except corner smearing), this would be a dealbreaker.
highdesertmesa wrote:
Midframe looks great on my copy at typical portrait distances. Past 3.5m there is some off center weakness wide open, but that’s it.
I wish that once someone recognizes they tested and reviewed a bad copy they would remove the review.
phinix wrote:
That's exactly my thoughts. There are not many reviews of this lens and when I was looking for one, I hit that review straight away and couldn't believe that this brand new design lens would be such a disappointment, like reviewer suggested. So, now seeing guys here showing more samples and seeing that its not that bad is a huge up lift, if you know what I mean.
That review should be either taken off, or re-done with different copy.
That's a challenge with single copy reviews. Most are never revisited. To be fair, many who are interested in this lens probably bought it to use rather than to test. If their copy didn't meet expectations, they'll review it as such and probably move on.
At least here there is a longer timeline and input by multiple knowledgeable members. We can compare results over time and raise questions if there are differences.
Maybe it's copy variation? Maybe early production wasn't as 'tight' as it should have been? Or maybe some other aspect of performance was prioritized at the expense of the mid-frame? It kind of reminds me of my experience with the 28 Cron v2. I bought it ~6 months after announcement after seeing a number of positive reviews. It appeared to have better across-frame performance at wider apertures. The first copy I got was badly rangefinder miscalibrated (like off by 3+ meters) but it also had strong field curvature. The dealer exchanged it and the second copy was good with the RF but also had strong field curvature. I noted both had production dates (based on the stickers on their boxes) that suggested they were made prior to the official announcement. I posted my performance results over at LUF and got mixed responses. Some were adamant their copies performed as claimed by Leica. But I also received full-res samples from others in which I could see the same degree of field curvature as my copies. So that turned me off the lens (and I moved to the Lux instead and have generally been happy with its tradeoffs). Fast forward 2-3 years and it seemed like those buying the lens were not seeing field curvature problems. Was my experience a fluke attributable to a couple bad copies very close in production? Probably not because I saw it in photos provided by others. That had me thinking it was possibly something amiss early in production that was likely quietly addressed.
open to suggestions and if other aperture values would be interesting to see, let me know and I can post them here as well.
https://i.imgur.com/0OYliNi.png