rscheffler
Offline
• • • • • •
Upload & Sell: Off
|
| p.3 #1 · p.3 #1 · Voigtlander Nokton 50mm f/1.0 announced for Canon RF mount | |
rscheffler wrote:
As for the magenta fringing problem with this lens and some R bodies - that's a different problem from tweaking for the sensor stack and is dependent on the sensor's design. The stated cameras that will exhibit this fringing are all FSI sensor tech
exdeejjjaaaa wrote:
R5, R6II are FSI too :-) ... so keep inventing
PS: R3 is the only BSI sensor that Canon managed to make for RF cameras...
What am I inventing?? Yes, R6II is FSI and I own it. Some of my rangefinder lenses adapted to it result in magenta color shift towards the edges/periphery of the image. I will have to revisit my tests of my rangefinder lenses on a loaner R5, but don't recall it having as noticeable a problem.
In any case, color shifts are typically related to a mismatch of optical design (short exit pupil distance) and sensor design (pixel well size, depth, microlenses), though with steep ray angles, differential IR blocking/absorption by the sensor stack can also result in color shift with BSI sensors, but IME this is blue/cyan rather than magenta. For Cosina to avoid the magenta color shift problem with some Canon FSI sensors, they would have had to more significantly redesign the lens optics so that peripheral ray angles would travel more perpendicular to the sensor plane. When the previous post stated Cosina optimized the lens for the RF mount, which Cosina specifically stated on their webpage for this lens (both RF and Z mount versions), and which you disputed, the implied meaning of such 'optimization' would be minor spacing adjustments to compensate for sensor stack thickness differences between Z and RF mount. Not wholesale optical redesign to avoid color shifts more inherent with FSI sensors than BSI.
Given that you mostly post on the Canon board, I don't know how much you follow Cosina's Voigtlander lenses for Leica M mount, which is the mount for which the 50/1 was first released. Or if you follow adapting rangefinder lenses to other mirrorless systems and how those compare in performance to the same lens models optimized for mirrorless. On Cosina's website, the block diagrams for lens models available as Leica M optimized and mirrorless optimized are exactly the same, including this 50/1, but also numerous other models. Yet numerous reviews indicate, including those by Fred Miranda himself, that Voigtlander's Leica M mount version lenses perform best on Leica sensors and the Sony FE mount versions perform best on Sony sensors. Adapting M mount versions to Sony often results in poorer image quality (increased field curvature, astigmatism) due to incompatibility with Sony's sensor stack thickness. This is solid evidence that Cosina indeed does optimize optics for whichever mount/sensor system the lens is marketed. The optical block diagrams published in their marketing materials won't show the subtle spacing differences between elements.
|
