As an aside -- this is the largest test I've seen of various Leica lenses on a somewhat homogenous, non-studio scene. I hope that full versions of these images might sometime be available. (If it would be a matter of buying a used SD card that happened to have some images on it, let me know. )
Looking through the images again, I see ... and I might have simply fallen off my rocker from drinking too much koolaid ... but I see higher contrast with the Kolari mod, even when no higher resolution is apparent. I think that what I'm seeing is due to the (obvious) reduction in blue filtering resulting in (a subtle?) reduction of atmospheric haze. Before I think too much about this, I'd be interested to learn if anyone else sees the same thing.
This makes me wonder if there isn't any UV contamination here. Does anyone in the southern hemisphere have a Kolari modded camera who would be willing to take a test portrait of someone with half a face's worth of sunscreen? Personally, this would make me jump at the mod -- I like the way UV looks in images -- but it would be worth knowing about if something beyond visible is making an appearance here. Unlikely -- I trust Kolari to know their stuff -- but perhaps worth a look.
Jon, thank you and you're welcome. We are planning to do a second series with just the problematic M lenses between the A7RII.Mod and the M9, possibly with the A7RII throwing in as a baseline. This would allow us to be more methodical.
So to the second point, I would not draw anything regarding the contrast just yet because the lighting changed visibly between the stock and mod tests. To keep thing more manageable, we shot with a bunch of lenses on the stock and then we switched to the mod (or vice versa). It would not affect the outcomes hugely since the reduction in smearing is still the most prominent one. Anything more subtle than that has to be evaluated with a more careful test.
Lastly, I'm not sure I understand the UV looks thing. But below is a portrait taken with the A7RII.Mod + Contax N 85. I only jacked up the Exposure, pushed down highlights and used strong contrast curve. No additional PP has been done. This is on the custom DNG profile I made.
I wouldn't expect this to be nearly as strong, but I'd be curious to see if, perhaps, the Kolari mod's color balance is caused more by transmitting more blue toward UV rather than by blocking excess red to stay safe from IR. Either way, I like the result, just wondering if there's more to it than shows up at first glance.
p.6 #4 · Kolari Mod's Improvements: A Summary of Tests
Full size test samples for G28, ZM25, and CV 15/4.5 vII on modded A7rII. Forgot to turn off auto ISO so not ideal testing conditions but you can get a good idea of performance. Focus point in the center for all shots. G28 and ZM25 both still have some field curvature and can be improved a bit by backing off focus.
p.6 #5 · Kolari Mod's Improvements: A Summary of Tests
Sam_W wrote:
I'm really, really curious how the ZM D15 performs with 10mm/16mm/26mm of extension tubes.
I imagine it would be similar to the Venus 15mm macro, except better overall performance...
I'll try 5mm extension first. I think maximum you can go is around 16mm since the working distance is the limiting factor. At around 1:2.5 mag, lighting is already a challenge. Will post some flower shots tomorrow.
---------------------------------------------
I wouldn't expect this to be nearly as strong, but I'd be curious to see if, perhaps, the Kolari mod's color balance is caused more by transmitting more blue toward UV rather than by blocking excess red to stay safe from IR. Either way, I like the result, just wondering if there's more to it than shows up at first glance.
Cheers,
Jon
Interesting. Can't promise, but I'll see what I can come up with.
---------------------------------------------
brendans wrote:
Full size test samples for G28, ZM25, and CV 15/4.5 vII on modded A7rII. Forgot to turn off auto ISO so not ideal testing conditions but you can get a good idea of performance. Focus point in the center for all shots. G28 and ZM25 both still have some field curvature and can be improved a bit by backing off focus.
p.6 #6 · Kolari Mod's Improvements: A Summary of Tests
JaKo wrote:
...BTW, have you noticed any differences in colour shift caused by Kolari mod that varied between mod versions and cameras, especially A7R2M sensor vs. 'older' Sony sensors?
These were taken around noon today. WB is set on Daylight for both the A7RII.Mod and the A7S.Mod. There was a slight exposure difference, so I matched them up in LR (-0.25 EV for A7RII). I used the Adobe Standard Profile since I'm not sure how else to approach this color thing. I don't have Raw Digger to see.
p.6 #8 · Kolari Mod's Improvements: A Summary of Tests
JaKo wrote:
Thanks Hiep!
Both mods/cameras' outputs look similar. Hopefully the original and 2nd Kolari mod versions won't require separate colour tweaking.
Jack,
That is one of the advantages of being late to the party since my modded A7r was delivered yesterday.
p.6 #9 · Kolari Mod's Improvements: A Summary of Tests
Maybe a short summary of our (Hiep and yours truly) tests are in order.
M lenses
ZEISS Distagon 2,8/15 ZM Lens seems better in the corner on the A7r2.Kolari.mod than on the M9 - but need to verify.
LEICA TRI-ELMAR-M 1:4/16-18-21 APH. A7r2.Kolari.mod slightly improved over Stock A7r2 for all focal lengths.
LEICA ELMARIT-M 1'2.8/21 ASPH. E 55 Surprise. Lens seems fully usable WO on A7r2.Kolari.mod, slight residual field curvature remains though.
LEICA SUMMICRON-M 1:2/28 ASPH. E 46 Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA ELMARIT-M 1:2.8 ASPH. E 39 Sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
ZEISS Distagon 1,4/35 ZM Extremely sharp at f/5.6 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA SUMMILUX-M 1:1.4/35 ASPH. E 46 (non-FLE) Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA SUMMICRON-M 1:2/35 ASPH. E 39 Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA SUMMILUX-M 1:1.4/50 ASPH. E 46 Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA APO-SUMMICRON-M 1:2/50 ASPH. Best at f/5.6 to f/11 on A7r2.Kolari.mod, minimall (insignificant?) improvement over Stock A7r2.
LEICA SUMMICRON-M 1:2/50 E 39 Very sharp at f/5.6 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
Ernst Leitz GmbH Wetzlar Summicron f = 5 cm 1:2 (Rigid Summicron) Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEICA ELMAR-M 1:2.8/50 E 39 (Collapsible) Very sharp at f/8 on A7r2.Kolari.mod. Stock A7r2 can't catch up.
LEITZ CANADA ELMAR 1:3.5/65 (Visoflex) Very soft - had CLA done, but wonder whether the lens was assembled correctly? I doubt it!
LEICA APO-SUMMICRON-M 1:2/75 ASPH. E 49 Best at f/5.6 to f/11 on A7r2.Kolari.mod, minimal (insignificant?) improvement over Stock A7r2.
LEITZ CANADA SUMMICRON 1:2/90 No discernable degradation of Stock A7r2 relative to Kolari modified A7r2.
LEICA ELMARIT-M 1'2.8/90 E 46 No discernable degradation of Stock A7r2 relative to Kolari modified A7r2.
LEITZ WETZLAR ELMARIT 1:2.8/90 No discernable degradation of Stock A7r2 relative to Kolari modified A7r2.
LEITZ LENS MADE IN CANADA TELE-ELMARIT-M 1:2.8/90 No discernable degradation of Stock A7r2 relative to Kolari modified A7r2.
LEITZ WETZLAR TELE-ELMAR 1:4/135 No discernable degradation of Stock A7r2 relative to Kolari modified A7r2.
FE lenses
ZEISS SONY Carl Zeiss Sonnar FE 2,8/35 ZA 49/0,35m/1,15ft (Zony) Stock A7r2 extremely sharp at f/4, minimal (insignificant?) degradation for A7r2.Kolari.mod in far corner.
MITAKON ZHONGYI SPEEDMASTER 0.95/50MM M67 Version 2 Stock A7r2 sharp at f/8. A7r2.Kolari.mod slightly better?
ZEISS SONY Carl Zeiss Sonnar FE 1,8/55 ZA 49/0,50m/1,64ft (Zony) Stock A7r2 sharp at f/5.6. Slight degradation for A7r2.Kolari.mod in far corner.
SONY FE 2.8/90 MACRO G OSS 62/0.28m/0.92ft Stock A7r2 sharp at f/5.6. Minimal (insignificant?) degradation for A7r2.Kolari.mod in far corner.
p.6 #10 · Kolari Mod's Improvements: A Summary of Tests
For those looking for lenses KEH currently has a killer sale going on: 20% $250, 25% off $500, 30% off $1000+. Just picked up a 24/3.8 elmar and ZM18 for great prices.
p.6 #11 · Kolari Mod's Improvements: A Summary of Tests
brendans wrote:
For those looking for lenses KEH currently has a killer sale going on: 20% $250, 25% off $500, 30% off $1000+. Just picked up a 24/3.8 elmar and ZM18 for great prices.
Thank you very much for the heads up. I just placed an order myself for a lens . This made it a no brainer. Would you mind posting some aperture test shot later? I'm interested in seeing how that ZM 18 would perform on a modded cam. I know the Elmar is fantastic at WO.
Edit: , they removed all the high priced items: Otii, Leica UWA Lux, etc.
p.6 #12 · Kolari Mod's Improvements: A Summary of Tests
hiepphotog wrote:
Thank you very much for the heads up. I just placed an order myself for a lens . This made it a no brainer. Would you mind posting some aperture test shot later? I'm interested in seeing how that ZM 18 would perform on a modded cam. I know the Elmar is fantastic at WO.
Edit: , they removed all the high priced items: Otii, Leica UWA Lux, etc.
No problem. Yeah, it's probably good they didn't have several other lenses I've been wanting to maybe give a try
p.6 #13 · Kolari Mod's Improvements: A Summary of Tests
Hmm. I wonder how the Otii will perform with the thinner sensor glass. LensRentals showed better MTF performance when they stuck in a 2mm piece of optical glass. Which makes me wonder whether it's designed to tolerate the ~1.9mm from a Sony body, or the ~1.2mm from Canon body. (I can't seem to find the Nikon glass thickness at the moment.)
p.6 #14 · Kolari Mod's Improvements: A Summary of Tests
Sam_W wrote:
Hmm. I wonder how the Otii will perform with the thinner sensor glass. LensRentals showed better MTF performance when they stuck in a 2mm piece of optical glass. Which makes me wonder whether it's designed to tolerate the ~1.9mm from a Sony body, or the ~1.2mm from Canon body. (I can't seem to find the Nikon glass thickness at the moment.)
Most would think they don't have much effect. JIm Kasson didn't test his Otii on the modded cam, probably because of this reasoning.
p.6 #15 · Kolari Mod's Improvements: A Summary of Tests
Sam_W wrote:
Hmm. I wonder how the Otii will perform with the thinner sensor glass. LensRentals showed better MTF performance when they stuck in a 2mm piece of optical glass. Which makes me wonder whether it's designed to tolerate the ~1.9mm from a Sony body, or the ~1.2mm from Canon body. (I can't seem to find the Nikon glass thickness at the moment.)
The 55 Otus is slightly worse without the sensor stack glass:
p.6 #16 · Kolari Mod's Improvements: A Summary of Tests
Let me raise a question to the experts here, concerning those Otus MTFs. Theoritecally, in the exact centre of image, glass stack thicknes shouldn't matter at all because the light rays are perpendicular. Therefore, both MTF graphs should start at the same point, with disregard to glass stack. On the curves above though, the Otus with 2mm. glass starts higher than the Otus without glass stack, This is not right. Or I'm missing something ?
p.6 #17 · Kolari Mod's Improvements: A Summary of Tests
artur5 wrote:
Let me raise a question to the experts here, concerning those Otus MTFs. Theoritecally, in the exact centre of image, glass stack thicknes shouldn't matter at all because the light rays are perpendicular. Therefore, both MTF graphs should start at the same point, with disregard to glass stack. On the curves above though, the Otus with 2mm. glass starts higher than the Otus without glass stack, This is not right. Or I'm missing something ?
I believe this would only be true with a theoretically perfect pinhole lens. Otherwise, with an aperture of any diameter, rays pass through the different parts of the aperture and are focused on the exact center of the image. Of course there are rays that are perfectly perpendicular to the sensor, but the large aperture guarantees that there will be angled rays as well. The larger the diameter (and closer the exit pupil), the larger the average angle, even for points in the exact center.
p.6 #18 · Kolari Mod's Improvements: A Summary of Tests
artur5 wrote:
Let me raise a question to the experts here, concerning those Otus MTFs. Theoritecally, in the exact centre of image, glass stack thicknes shouldn't matter at all because the light rays are perpendicular. Therefore, both MTF graphs should start at the same point, with disregard to glass stack. On the curves above though, the Otus with 2mm. glass starts higher than the Otus without glass stack, This is not right. Or I'm missing something ?
For fast glass, Brian Caldwell said that thicker glass somewhat reduces the on-axis spherical aberration, so you get a slightly higher on-axis reading:
"Brian tells me that the amount of glass in the path creates on-axis spherical aberration that could affect center MTF on wide-aperture lenses."
Here is an interpretation from Lloyd:
"On axis the added plate changes the correction towards spherically overcorrected when the lens speed is fast."
FWIW, this effect is also seen on the Zeiss MTF measurement with 2.5mm thick optical glass for the ZM 35/1.4.
p.6 #19 · Kolari Mod's Improvements: A Summary of Tests
Optics is always a series of tradeoffs. A push here, can mean a pull there. I think folks tend to get enamored by the gains they are chasing, and often sometimes tend to gloss over where the trade was to get it ... then discounting it as insignificant. Of course, one person's significance can be another persons insignificance.
The thing that is more notable to me in the comp graph isn't the starting point @ higher (which it is), but how tight the sag/tan lines are vs. how soon the start coming apart and how far apart they are in the end.
p.6 #20 · Kolari Mod's Improvements: A Summary of Tests
RustyBug wrote:
Optics is always a series of tradeoffs. A push here, can mean a pull there. I think folks tend to get enamored by the gains they are chasing, and often sometimes tend to gloss over where the trade was to get it ... then discounting it as insignificant. Of course, one person's significance can be another persons insignificance.
The thing that is more notable to me in the comp graph isn't the starting point @ higher (which it is), but how tight the sag/tan lines are vs. how soon the start coming apart and how far apart they are in the end....Show more →
Yes, I don't know of any technique (shooting or post process) to reduce the astigmatism of a lens. This aberration is probably the third most hated aberrations on my list, after field curvature and excessive purple fringing. The reason it's not at the top is because it's not always apparent until you spend quite some times with a lens.
)
, they removed all the high priced items: Otii, Leica UWA Lux, etc.
