I've posted a resolution/contrast infinity test showing how the VM version performs on both the Leica and 'Sony + adapter.' This could be useful for those who want to use the VM version with both systems: https://www.fredmiranda.com/forum/topic/1647789/4#15256999
imoretti wrote:
I don’t see any coding on my 21/1.4. Here’s a site, though, that will show you how to code it. Apparently, the newer VM glass has a groove on the mount which can be used for this purpose. (I see that on my 75/1.5 also.)
If you decide to make your own 6-bit code template from the JPEG file, just print it to Paper Size (US Letter) and use Scale (53%). Link to Original Article
I played with it today and was able to print a template that fits my M-lenses perfectly. I don't think you can use a sharpie anymore because Leica tweaked the 6-bit sensor with recent bodies. You may need to use a 'flat' black and white paint when using the template.
Nice! Since I use my VM glass on a Z7, I’m hoping that this will be useful to those with a Leica body (or adapter) that could actually read the codes. But, who knows, I may yet join those ranks!
Thanks, too, for your other reminder that the profiles would only be approximations as the codes were created for Leica lenses. I also realized after I posted the links that they were from about 10 years ago, so probably missing newer lenses.
imoretti wrote:
Nice! Since I use my VM glass on a Z7, I’m hoping that this will be useful to those with a Leica body (or adapter) that could actually read the codes. But, who knows, I may yet join those ranks!
Thanks, too, for your other reminder that the profiles would only be approximations as the codes were created for Leica lenses. I also realized after I posted the links that they were from about 10 years ago, so probably missing newer lenses.
The Z7 uses a Sony sensor but many say the stack is slightly thinner. Still, it's likely the 21/1.4 E-mount + adapter would perform better on the Nikon body compared to the VM version.
Unless you intend to use the VM version on a Leica body as well, I would go with the E-mount version for both Sony and Nikon cameras.
I didn’t realize this... just did the proverbial “assumed”...
I already have the 21/1.4, 28/2, 40/1.2 and 75/1.5 all in M so a little difficult to turn back now. makes a good excuse down the road (a few years?) to pick up a Leica.
Fred Miranda wrote:
The Z7 uses a Sony sensor but many say the stack is slightly thinner. Still, it's likely the 21/1.4 E-mount + adapter would perform better on the Nikon body compared to the VM version.
Unless you intend to use the VM version on a Leica body as well, I would go with the E-mount version for both Sony and Nikon cameras.
imoretti wrote:
I didn’t realize this... just did the proverbial “assumed”...
I already have the 21/1.4, 28/2, 40/1.2 and 75/1.5 all in M so a little difficult to turn back now. makes a good excuse down the road (a few years?) to pick up a Leica.
I'm using the 21/1.4 on the Nikon Z7. It needs some color cast correction (use Capture One, for example), but probably less than other lenses with shorter exit pupil to sensor distances. The lens has some coma wide open, and any aberrations induced by the sensor stack at oblique ray angles is probably much less than the coma. My initial assumption would be that both version (VM and E) are essentially the same on the Nikon Z7. I bought the VM version because I have a couple of Leica M lenses that I'm also using on the Z7.
Jun 26, 2020 at 02:42 AM
Steve Spencer Offline • • • • • • • Upload & Sell: On
AcuteShadows wrote:
I'm using the 21/1.4 on the Nikon Z7. It needs some color cast correction (use Capture One, for example), but probably less than other lenses with shorter exit pupil to sensor distances. The lens has some coma wide open, and any aberrations induced by the sensor stack at oblique ray angles is probably much less than the coma. My initial assumption would be that both version (VM and E) are essentially the same on the Nikon Z7. I bought the VM version because I have a couple of Leica M lenses that I'm also using on the Z7.
I have the E version that I use on the Z7. I don't see any color cast that needs to be corrected. The lens does have quite a bit of coma wide open, but it clears up mostly by f/2.8 and pretty much totally be f/4. It is sharp across the frame from f/2.8 as well. The Z cover glass is a lot more like Sony cover glass than Leica cover glass, so I suspect the E version is a little better on the Z camera, but I have not seen direct comparisons.
Steve Spencer wrote:
I have the E version that I use on the Z7. I don't see any color cast that needs to be corrected. The lens does have quite a bit of coma wide open, but it clears up mostly by f/2.8 and pretty much totally be f/4. It is sharp across the frame from f/2.8 as well. The Z cover glass is a lot more like Sony cover glass than Leica cover glass, so I suspect the E version is a little better on the Z camera, but I have not seen direct comparisons.
The color cast is less than on the current Summicron 28mm, but it's there. Maybe you can see it if you take a picture of a gray sky through a translucent glass or plastic, and check the color values. Otherwise, the effect can be subtle, like if you would light the centre with a small lightbulb. The main downside is that the surrounding light (towards the edges) appears to become increasingly cold, and that may introduce some claustrophic atmosphere to an image. I have no way to determine the actual thickness of the sensor stacks, and the optical properties of the layers. But the results of using the same lenses on Nikon Z and on Sony cameras are different, and - with regard to aberrations - the only plausible explanations i can give is differences in the sensor stack, including possibly the microlens array design.
Jun 26, 2020 at 12:28 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
AcuteShadows wrote:
The color cast is less than on the current Summicron 28mm, but it's there. Maybe you can see it if you take a picture of a gray sky through a translucent glass or plastic, and check the color values. Otherwise, the effect can be subtle, like if you would light the centre with a small lightbulb. The main downside is that the surrounding light (towards the edges) appears to become increasingly cold, and that may introduce some claustrophic atmosphere to an image. I have no way to determine the actual thickness of the sensor stacks, and the optical properties of the layers. But the results of using the same lenses on Nikon Z and on Sony cameras are different, and - with regard to aberrations - the only plausible explanations i can give is differences in the sensor stack, including possibly the microlens array design.
My experience comparing the Z7 and A7r II is quite different. The E version of the Voigtlander 21 f/1.4 performs almost identically on the two cameras. Any differences are very subtle and I don't see a color cast with either camera. Below is an example of a sunset shot with the E version with no correction for color cast. I just don't see it in this image or in any image I have shot with the lens on the Z7. The Sony and Nikon cameras have a similar set up for their cover glass with a pice of glass on the sensor itself and a piece of glass mounted with the dust shaker a bit off the sensor. The Leica has no dust shaker and there is just a thin piece of glass on the sensor itself. The Leica cover glass is like substantially thinner than either the Sony or the Nikon mirrorless camera.
Steve,
I do think the Leica M sensor stack is the thinnest we have and in general M lenses perform their best on M camera bodies (There are exceptions though). Even the SL/SL2 sensor stack is thicker than the Leica M.
The stack thickness from both Sony and Nikon Z sensors seem to be similar as you reported. (Sony manufactures sensors for Nikon)
I've seen some reports showing Loxia lenses performing a tad better on the Nikon body but I have never tested this myself. Perhaps any difference could also be attributed to mount to sensor and/or adapter tolerance.
Jun 26, 2020 at 01:07 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
Fred Miranda wrote:
Steve,
I do think the Leica M sensor stack is the thinnest we have and in general M lenses perform their best on M camera bodies (There are exceptions though). Even the SL/SL2 sensor stack is thicker than the Leica M.
The stack thickness from both Sony and Nikon Z sensors seem to be similar as you reported. (Sony manufactures sensors for Nikon)
I've seen some reports showing Loxia lenses performing a tad better on the Nikon body but I have never tested this myself. Perhaps any difference could also be attributed to mount to sensor and/or adapter tolerance.
Yes, Fred, I totally agree with what you said here. Your thread on Leica M lenses that also perform well on Sony cameras has been very useful, and I think there are a number of lenses that do as well on Sony cameras as on the Leica M as you have shown (and I reported on a number of lenses as well). That said the corners of the M version of the CV 21 f/1.4 seem to be noticeably worse on a Sony than on a Leica, whereas the E mount version of the CV 21 f/1.4 has wonderful corners from pretty much wide open. So, IMO, it is a lens that benefits from getting the Sony E mount version rather than just using the M mount version on Sony. Of course, other may care less about the corners and are fine with the M version on Sony.
I also have now shot a bunch of E mount lenses (Loxia 21, Loxia, 25, Loxia 50, Loxia 85, Voigtlander 21 f/1.4, Voigtlander 40 f/1.2, Voigtlander 65 f/2 APO, Voigtlander 110 f/2.5 APO) on both the Sony A7r II and the Nikon Z7 and I have yet to see any substantial difference in how the lenses perform on the two cameras. The results have always been very very similar. This suggests to me that the Leica M version of the 21 f/1.4 would likely perform like it does on Sony sensors, (excellent in the center and even edges, but a noticeably drop in performance in the corners) on the Z7 as well. I haven't tested that so I may be wrong, but I suspect the Sony E mount version is likely noticeably better on the Z7 (I know my Sony E mount is excellent all the way to the corners on my Z7) in the corners at wider apertures (but probably only there), than the Leica M version is on the Z7.
Steve Spencer wrote:
My experience comparing the Z7 and A7r II is quite different. The E version of the Voigtlander 21 f/1.4 performs almost identically on the two cameras. Any differences are very subtle and I don't see a color cast with either camera. Below is an example of a sunset shot with the E version with no correction for color cast. I just don't see it in this image or in any image I have shot with the lens on the Z7. The Sony and Nikon cameras have a similar set up for their cover glass with a pice of glass on the sensor itself and a piece of glass mounted with the dust shaker a bit off the sensor. The Leica has no dust shaker and there is just a thin piece of glass on the sensor itself. The Leica cover glass is like substantially thinner than either the Sony or the Nikon mirrorless camera....Show more →
I'm not saying you necessarily see a color cast in a different picture, but you need a picture without color gradients in order to really be able to ascertain any potential color cast. Most unicolor walls, for example, are not suitable due to gradients in the ambient light. Once you determine the existence and character of the color cast, you may see it a lot of other pictures, or you just apply the color cast correction file in post, resulting usually in a more lively and evenly "lit" image.
Steve Spencer wrote:
Yes, Fred, I totally agree with what you said here. Your thread on Leica M lenses that also perform well on Sony cameras has been very useful, and I think there are a number of lenses that do as well on Sony cameras as on the Leica M as you have shown (and I reported on a number of lenses as well). That said the corners of the M version of the CV 21 f/1.4 seem to be noticeably worse on a Sony than on a Leica, whereas the E mount version of the CV 21 f/1.4 has wonderful corners from pretty much wide open. So, IMO, it is a lens that benefits from getting the Sony E mount version rather than just using the M mount version on Sony. Of course, other may care less about the corners and are fine with the M version on Sony.
I also have now shot a bunch of E mount lenses (Loxia 21, Loxia, 25, Loxia 50, Loxia 85, Voigtlander 21 f/1.4, Voigtlander 40 f/1.2, Voigtlander 65 f/2 APO, Voigtlander 110 f/2.5 APO) on both the Sony A7r II and the Nikon Z7 and I have yet to see any substantial difference in how the lenses perform on the two cameras. The results have always been very very similar. This suggests to me that the Leica M version of the 21 f/1.4 would likely perform like it does on Sony sensors, (excellent in the center and even edges, but a noticeably drop in performance in the corners) on the Z7 as well. I haven't tested that so I may be wrong, but I suspect the Sony E mount version is likely noticeably better on the Z7 (I know my Sony E mount is excellent all the way to the corners on my Z7) in the corners at wider apertures (but probably only there), than the Leica M version is on the Z7....Show more →
It's perfectly possible that the Sony sensor is closer to the Nikon Z sensor, and that E lenses may be the better choice on the Nikon Z. Yet my experience with M lenses differs from what i hear and see about the performance of M lenses on Sony bodies. Maybe it's not sensor thickness, but the microlens array. I'm not willing to do a cut-out of my Z7, but would be very interested in what actually causes the different results.
AcuteShadows wrote:
It's perfectly possible that the Sony sensor is closer to the Nikon Z sensor, and that E lenses may be the better choice on the Nikon Z. Yet my experience with M lenses differs from what i hear and see about the performance of M lenses on Sony bodies. Maybe it's not sensor thickness, but the microlens array. I'm not willing to do a cut-out of my Z7, but would be very interested in what actually causes the different results.
IMO the stack thickness is the major culprit. Microlenses design play a part too.
If a lens is optimized for a thin sensor and it's used on a thicker one, there will be induced field curvature. I believe the second best camera to use with M lenses is the Leica SL series but it's still not perfect. Sony and Fuji are two of the worst.
In the case of E-mount lenses, they are optimized for Sony sensors and Nikon uses Sony, so they are pretty much compatible. Perhaps not perfect with all lenses though.
Fred Miranda wrote:
IMO the stack thickness is the major culprit. Microlenses design play a part too.
If a lens is optimized for a thin sensor and it's used on a thicker one, there will be induced field curvature. I believe the second best camera to use with M lenses is the Leica SL series but it's still not perfect. Sony and Fuji are two of the worst.
In the case of E-mount lenses, they are optimized for Sony sensors and Nikon uses Sony, so they are pretty much compatible. Perhaps not perfect with all lenses though.
I think I ultimately need to buy a Sony camera just to find out what the actual differences between both sensors are with regard to M lenses.
AcuteShadows wrote:
I think I ultimately need to buy a Sony camera just to find out what the actual differences between both sensors are with regard to M lenses.
Voigtländer 21mm f/1.4 on the Nikon Z7 at f/1.4. 100% pixel resolution, distance is 50 to 100 meters, rather than infinity. First image is from the extreme corner, second is from about 4mm up on the edge. The Leica sensor is probably still better. However, I didn't by the Voigtländer in order to shoot f/1.4 at infinity, but because it probably has one of the more distant exit pupils (i.e. virtual origin of exiting light rays) of the smaller wide-angle lenses.
makes a good excuse down the road (a few years?) to pick up a Leica.
