Archive 2015 · Kolari Mod's Improvements: A Summary of Tests
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.13 #1 · Kolari Mod's Improvements: A Summary of Tests
Fred Miranda wrote:
Does the Kolari mod improves or degrades the corners for the CV 15/4.5 III (M-mount) ?
I have a copy working really well on the stock A7RII even wide-open and wondered if a Kolari modification would change this performance for better or worse since this lens seems to be optimized for the Sony sensor stack...
Fred, based on sensor stack thickness the mod should improve the CV 15 f/4.5 III in M-mount, but interfere with performance of the E-mount lens.
p.13 #2 · Kolari Mod's Improvements: A Summary of Tests
Steve Spencer wrote:
Fred, based on sensor stack thickness the mod should improve the CV 15 f/4.5 III in M-mount, but interfere with performance of the E-mount lens.
Steve,
That would make sense but the CV 15/4.5 III is a curious lens as the M-mount and E-mount versions seem to be identical regarding sensor stack optimization.
p.13 #3 · Kolari Mod's Improvements: A Summary of Tests
MAubrey wrote:
Thanks everyone!
I had the original A7. I've now switched to the A7rII, but I'm looking at a second body with the mod. I'd like the superior EVF of the A7r, but I know from experience that the color shift is worse on the A7r compared to the A7.
I was hoping that I had remembered the info about the mod wrong and that it would help the A7r's color shift problem a bit. Looks like the original A7, it is, then.
the a7 and a7r have the same evf. it's the a7ii and a7rii that have different evfs. the a7 is much better than the a7r with regard to color shift but still not near as good as the a7rii.
p.13 #4 · Kolari Mod's Improvements: A Summary of Tests
sebboh wrote:
the a7 and a7r have the same evf. it's the a7ii and a7rii that have different evfs. the a7 is much better than the a7r with regard to color shift but still not near as good as the a7rii.
Sorry, should have clarified that a bit. The original A7 & A7rII zoom in to 12x for focus magnification, but the original A7r could go to 14.4x, which is really useful for MF lenses.
Aug 24, 2016 at 01:23 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.13 #5 · Kolari Mod's Improvements: A Summary of Tests
Fred Miranda wrote:
Steve,
That would make sense but the CV 15/4.5 III is a curious lens as the M-mount and E-mount versions seem to be identical regarding sensor stack optimization.
Interesting. Perhaps they built it for a sensor stack in between the M240 and the A7rII on purpose. That way they could just swap out the mount and keep the optics the same. They may have optimized the M mount for the Sony sensor stack, but that would seem like an odd way to go as presumably at least a fair number of people would like to use the M mount lens on an M camera.
p.13 #6 · Kolari Mod's Improvements: A Summary of Tests
MAubrey wrote:
Sorry, should have clarified that a bit. The original A7 & A7rII zoom in to 12x for focus magnification, but the original A7r could go to 14.4x, which is really useful for MF lenses.
i could be wrong, but i believe those magnification factors just correspond to the 1:1 for their sensors (1 pixel on sensor = 1 pixel on evf at 12x for 24mp and 14.4x on 36mp).
i haven't really noticed any improvement in my manual focus accuracy for the 1:1 compared to the 2:1 magnification. everybody works differently though.
p.13 #7 · Kolari Mod's Improvements: A Summary of Tests
Steve Spencer wrote:
Interesting. Perhaps they built it for a sensor stack in between the M240 and the A7rII on purpose. That way they could just swap out the mount and keep the optics the same. They may have optimized the M mount for the Sony sensor stack, but that would seem like an odd way to go as presumably at least a fair number of people would like to use the M mount lens on an M camera.
This begs the question: Is it possible to have a M to E-mount adapter with corrective optics overcoming the issue with the different sensor glass thickness? It would make much more sense than altering the sensor stack and making the camera incompatible with native lenses.
Aug 24, 2016 at 03:04 PM
Steve Spencer Offline • • • • • • • Upload & Sell: On
p.13 #8 · Kolari Mod's Improvements: A Summary of Tests
Fred Miranda wrote:
This begs the question: Is it possible to have a M to E-mount adapter with corrective optics overcoming the issue with the different sensor glass thickness? It would make much more sense than altering the sensor stack and making the camera incompatible with native lenses.
It might be easier for someone like Cosina to go the other way. You could make a lens adjusted for the Sony E mount and then just add about a mm of glass in the path behind the last element for an M mount lens.
p.13 #9 · Kolari Mod's Improvements: A Summary of Tests
Steve Spencer wrote:
It might be easier for someone like Cosina to go the other way. You could make a lens adjusted for the Sony E mount and then just add about a mm of glass in the path behind the last element for an M mount lens.
I meant a third party vendor like Fotodiox or Metabones for example. They could develop an adapter with corrective optics to take in consideration the thicker stack projecting the rays into the correct path. It could be a cheaper alternative to modifying the sensor and would not impair native lenses without the adapter.
p.13 #10 · Kolari Mod's Improvements: A Summary of Tests
Lenses seem to degrade differently, so I wonder if such a corrective optic would be equally effective with all lenses.
Fred Miranda wrote:
Does the Kolari mod improves or degrades the corners for the CV 15/4.5 III (M-mount) ?
I have a copy working really well on the stock A7RII even wide-open and wondered if a Kolari modification would change this performance for better or worse since this lens seems to be optimized for the Sony sensor stack...
Steve Spencer wrote:
Fred, based on sensor stack thickness the mod should improve the CV 15 f/4.5 III in M-mount, but interfere with performance of the E-mount lens.
Fred Miranda wrote:
Steve,
That would make sense but the CV 15/4.5 III is a curious lens as the M-mount and E-mount versions seem to be identical regarding sensor stack optimization.
Steve Spencer wrote:
Interesting. Perhaps they built it for a sensor stack in between the M240 and the A7rII on purpose. That way they could just swap out the mount and keep the optics the same. They may have optimized the M mount for the Sony sensor stack, but that would seem like an odd way to go as presumably at least a fair number of people would like to use the M mount lens on an M camera.
I saw some example images early in the 15/4.5III M-mount release period shot by one guy on both systems (IIRC) and the unmodded Sony images looked slightly sharper. My guess is Cosina probably optimized the lens to be somewhere between the two sensor stack thicknesses. I would guess similar is true with the new 10mm and 12mm.
p.13 #11 · Kolari Mod's Improvements: A Summary of Tests
rscheffler wrote:
No all lenses seem to degrade similarly so I wonder if such a corrective optic would be equally effective with all lenses.
I saw some example images early in the 15/4.5III M-mount release period shot by one guy on both systems (IIRC) and the unmodded Sony images looked slightly sharper. My guess is Cosina probably optimized the lens to be somewhere between the two sensor stack thicknesses. I would guess similar is true with the new 10mm and 12mm.
It could be but this lens has a lot of variation towards the edges and it's not easy to get a good sample. I was able to test two great copies (M and E mount) and they look virtually identical towards the edges at all apertures. That actually seem to reinforce what you wrote about the lens being optimized for both stacks. (averaging the thickness)
p.13 #13 · Kolari Mod's Improvements: A Summary of Tests
Steve Spencer wrote:
It might be easier for someone like Cosina to go the other way. You could make a lens adjusted for the Sony E mount and then just add about a mm of glass in the path behind the last element for an M mount lens.
Has anyone actually tried the Loxia 35 on a Kolari modded A7R II or A7R? I wonder if there are any added benefit with the thinner filter stack combined with the corrective optics inherent on the Loxia.
p.13 #14 · Kolari Mod's Improvements: A Summary of Tests
I suspect that what Cosina did was to redesign completely the optics of the new CV 15/4.5 making it a clearly retrofocal design, which greatly minimizes the importance of the sensor stack thickness. Thus, it won't make much difference if the lens is optimized for 1mm or 2mm. covers. Possibly, the infamous copy variation will be a more important factor, concerning corner sharpness, than the thickness of the sensor cover.
p.13 #15 · Kolari Mod's Improvements: A Summary of Tests
artur5 wrote:
I suspect that what Cosina did was to redesign completely the optics of the new CV 15/4.5 making it a clearly retrofocal design, which greatly minimizes the importance of the sensor stack thickness. Thus, it won't make much difference if the lens is optimized for 1mm or 2mm. covers. Possibly, the infamous copy variation will be a more important factor, concerning corner sharpness, than the thickness of the sensor cover.
Maybe, but if so, I'm not really seeing it in the design.
p.13 #16 · Kolari Mod's Improvements: A Summary of Tests
tzhang4284 wrote:
Has anyone actually tried the Loxia 35 on a Kolari modded A7R II or A7R? I wonder if there are any added benefit with the thinner filter stack combined with the corrective optics inherent on the Loxia.
No, but I have tried the FE 35/2.8 ZA and Batis 25/2.0 on a Kolari modded V2 a7. In both cases, at infinity, when focusing dead center wide open, the corners needed focusing toward infinity just a bit. This would imply field curvature as if you would hold an open umbrella out in front of yourself. Keep in mind that the thicker coverglass of a stock a7 type camera, pushes the infinity of corners out away. So, how flat is the field on a stock camera?
I am trying to compare these 2 lenses on a stock a7R II, but the atmosphere here is not that clear to be able to see a difference wide open in corners/center, but at the Batis 25 optimum of f/5.6 corner sharpness equals the center. Diffraction is evident at f/8.
Update: Due to the air clearing here, I can detect no focus difference between corner and center at wide open infinity on a stock a7RII. This was my hopeful outcome for 24-25 mm. The Batis is obviously designed for the thick coverglass of the stock a7 type camera. A Kolari modded camera with thinner coverglass pulls back the edge focus as described above.
p.13 #17 · Kolari Mod's Improvements: A Summary of Tests
JimBuchanan wrote:
No, but I have tried the FE 35/2.8 ZA and Batis 25/2.0 on a Kolari modded V2 a7. In both cases, at infinity, when focusing dead center wide open, the corners needed focusing toward infinity just a bit. This would imply field curvature as if you would hold an open umbrella out in front of yourself.
This seems to be correct Jim and if you do the opposite by adding a M-mount lens (ZM35 Distagon for example) on a stock A7RII body, the field curvature moves to the other direction because of the thicker sensor stack. Meaning, when focusing at center, the edges are not in focus and need focusing "away" from infinity. (I call this inwards field curvature).
That's exactly the opposite with the Loxia 21/2.8 and Loxia 35/2 which are probably corrected for the stack thickness. On these lenses, FC goes to the other direction (outwards)
Oh, and one comment on the ZM35 Distagon. I went back to look at the comparison shots with the FE 35/2.8 ZA and at corner infinity the ZM is sharper at f/2.8 and f/1.4!
The ZM 35/1.4 is an incredible lens even on the stock A7RII. If suffers from induced field curvature and astigmatism on a stock body but by f/5.6 and smaller, it's sharp as a tack all over the image field.
p.13 #18 · Kolari Mod's Improvements: A Summary of Tests
Fred Miranda wrote:
This seems to be correct Jim and if you do the opposite by adding a M-mount lens (ZM35 Distagon for example) on a stock A7RII body, the field curvature moves to the other direction because of the thicker sensor stack. Meaning, when focusing at center, the edges needs focusing "away" from infinity. (I call this inwards field curvature).
That's exactly the opposite with the Loxia 21/2.8 and Loxia 35/2 which are probably corrected for the stack thickness. On these lenses, FC goes to the other direction (outwards)
The ZM 35/1.4 is an incredible lens even on the stock A7RII. If suffers from induced field curvature and astigmatism on a stock body but by f/5.6 and smaller, it's sharp as a tack all over the image field....Show more →
Now it all makes sense! I've been referring to direction of curvature exactly the opposite. Was beginning to question my sanity. Maybe I should look into the Loxia 35/2, that's my preferred curvature direction
to be picked up by Camera Raw or LightRoom. You might have to unhide the folders if you don't see the AppData folder in your user folder.
In my earlier post (on the v1 profiles) I commented that I thought the colors were a bit saturated with these profiles. After more extensive use, I think this was just me using Camera Neutral for a long time, which is quite unsaturated. I still dial back the saturation a little with these, but I think it's more about personal taste than absolute correctness.
I just made these and have not tested them extensively. Owners with the humidity resistant glass filters, please give them a try and let me know what you find. People reported good results with the v1 profiles, especially improvements in blue skies.
Michael
Update Nov 9 - Did some shooting this weekend under difficult lighting with good results. Still needs a large Magenta shift (I need to understand why) in LR but good results overall....Show more →
Just got my camera back from Kolari and I'm trying to figure out where to put the .dcp file linked on the first page of the thread, so that I can apply the proper settings to my imported RAW files. I'm on a Mac. I currently have it in Library/Application Support/Adobe/CameraRaw/CameraProfiles, after selecting the hidden Library (holding down on the Alt/Option key) from the Go menu on the desktop. I also tried putting the file into the Lightroom Color Profiles folder to no avail. I'm not too savvy with this stuff, can anyone lend a hand?
