Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Sony Forum | Join Upload & Sell

1       2       3              14      
15
       end
  

Archive 2015 · Kolari Mod's Improvements: A Summary of Tests

  
 
bjornthun
Offline
• • • •
Upload & Sell: Off
p.15 #1 · p.15 #1 · Kolari Mod's Improvements: A Summary of Tests


Corrective lens elements to work with the diminutive Voigtländers 21/4, 25/4, 28/3;5 and 35/2.5 would be very nice. Same for the ZM Biogons. Wonder if Cosina could work out something themselves?


Oct 03, 2016 at 08:14 PM
hiepphotog
Offline
• • • •
Upload & Sell: On
p.15 #2 · p.15 #2 · Kolari Mod's Improvements: A Summary of Tests


Fred Miranda wrote:
Yes, perhaps we can get a one solution that's good enough for most lenses. However, it only seems to be a solution for astigmatism.
How about induced field curvature from the Sony sensor stack?


Fred, from my experience with the G21 before and after the mod, I would say this one actually fixes both of the problem. Induced field curvature will throw the corners out of focus, so you would get the regular de-focus blur. Astigmatism always appears as double image since tangential part will be more affected.

So from the picture above, the corners are not only in focus (i.e. on the similar plain with the center) but also I don't see any heavy double image (minor astigmatism); the rest is just mainly the inherent softness. So with that, we can just buy a plano-convex from Edmund Optics here in the States and experience away. I have been looking at their inventory quite extensively lately for a non-photography application.



Oct 03, 2016 at 10:35 PM
artur5
Offline
• • •
Upload & Sell: Off
p.15 #3 · p.15 #3 · Kolari Mod's Improvements: A Summary of Tests


Kudos to HaruhikoT for his contribution. This opens a a new world for people reluctanct to mod their sensors. It’s not only that the warranty is void if you do the mod, it's also the cost. Specially for people outside the USA is quite expensive, due to the added shipping charges and huge Customs fees ( at least for EU countries).
An external “filter”, so to speak, is definitevely more affordable, even using a different plano-convex unit for each lens. Reminds me of the M8 UV/IR filters of some years ago.
Another thing. Given that this thread is dedicated to the sensor stack modification and the external solution, if proved effective, might somehow hurt Kolarivision sales, perhaps it would be fair to create a new thread for the plano-convex external lenses and leave this one only for the original sensor stack mod ?



Oct 04, 2016 at 03:01 AM
HaruhikoT
Offline
• •
Upload & Sell: Off
p.15 #4 · p.15 #4 · Kolari Mod's Improvements: A Summary of Tests


Thank you having many comments and questions about my posts. I agree with artur5, I will create new thread, just copy my last posts and reply questions hours later.
Haruhiko



Oct 04, 2016 at 04:49 AM
HaruhikoT
Offline
• •
Upload & Sell: Off
p.15 #5 · p.15 #5 · Kolari Mod's Improvements: A Summary of Tests


Just created new thread to talk about external filter solution.
https://www.fredmiranda.com/forum/topic/1453834
I also posted replies for questions from you to the new thread.

Haruhiko



Oct 04, 2016 at 08:14 AM
Brandon Dube
Offline
• •
Upload & Sell: Off
p.15 #6 · p.15 #6 · Kolari Mod's Improvements: A Summary of Tests


HaruhikoT wrote:
Hi folks,

While Kolari mod is great solution for wide-angle rangefinder lens users,
personally I still hesitate to modify my A7 because such unofficial mod voids warranty.
Today I would like to introduce another solution to fix corner smearing issue.

I believe Astigmatism is the principal cause of corner smearing.
I tried some optical simulations and found that a plano-convex lens placed in front of the WA RF lens can control astigmatism.
I use OSLO free edition:
http://www.lambdares.com/oslo-university-program

For simplicity, simulation use perfect lens instead of complicated actual lens design.
I think symmetrical lens design such as Biogon matches this simplified simulation, but retrofocus design does not.
Here are the
...Show more

Hi,

Some may recognize my name from my work with Olaf Optical Testing / LensRentals. My primary engagement in optics is as an aberration theory researcher / student. My perspective may be of some interest here.

The coverglass generates aberrations because not all rays in a given cone, fan, bundle, or "pencil" of light that forms a point in the image hit it at the same angle if the beam is not from infinity. Snell's law says that a ray that is incident at an angle different to another ray is going to go somewhere different to the other ray. Thus, you get aberrations.

There are two types of aberrations some lens designers like to talk about. There are intrinsic aberrations, the ones that things like Seidel Sums and wavefront expansions calculate. There are also induced aberrations, or aberrations generated by aberration somewhere else in the system.

A perfect thin lens model captures the intrinsic aberrations of the coverglass but misses the induced aberrations due to the aberrations from the master or taking lens. For a well corrected master lens, you can assume those aberrations are pretty minimal.

Regarding symmetric vs retrofocus, there is nothing so different between the two as far as aberrations go. The big difference is that for a symmetric lens, the chief ray angle is similar to the object angle. The chief ray angle is what the exit pupil distance information on this subject you can find on the LensRentals blog, as done by Brian Caldwell, really is a function of. For a retrofocus lens, the chief ray angle is less than the object angle.

You can toggle your simulation between retrofocus and symmetric, with a perfect master lens, by separating the aperture stop from the thin lens and putting it a bit in front of it. If it is at the front focal point of the thin lens, the system is telecentric and you have the limiting case of extremity for a retrofocus lens, and something similar to the behavior of most fisheyes. In the stop at the lens case, you are close to the symmetric behavior though it is not exactly so.

The aberration I would worry about in these simulations is field curvature, or astigmatism. Unfortunately you can't generate pure astigmatism in the number of surfaces afforded to you by the free version of OSLO. Field curvature you can do by placing a meniscus lens with equal radii right on top of the cover glass. It will contribute nothing to the focal length and image location, but adjust the curvature of the field. There may be some small astigmatism generated as wel. This should give you 5 of the 6 surfaces (or maybe all 6 if the thin lens is 2 surfaces) you are allowed for free.

If you would like, I can perform some simulation in my license of Code V, though the work must remain very strictly noncommercial. I also have an archive of every design in Modern Lens Design, as I have served as a sub-editor of its 3rd edition. There are maybe 15 biogons in there and 15 retrofocus lenses.

Regardless, I am happy to provide guidance.

All the best,
Brandon



Oct 04, 2016 at 08:44 PM
Brandon Dube
Offline
• •
Upload & Sell: Off
p.15 #7 · p.15 #7 · Kolari Mod's Improvements: A Summary of Tests


HaruhikoT wrote:
To prove the theory, I purchased a f=1.5m lens "SLB-50-1500PM" and a f=2m lens "SLB-50-2000PM" from Japanese lens supplier SIGMA-KOKI.
Both are 3mm thick, 50mm diameter and AR-coated. $85-$90 each.
You can order here:
http://www.global-optosigma.com/en_jp/Catalogs/pno/?from=page&pnoname=SLB-P%28%CF%8630%EF%BD%9E%CF%8650%29&ccode=W3043&dcode=W3043-2


Quick comment; Opto-Sigma is a fine vendor of optical mounts and hardware. I have used dozens of their mounts and they are high quality. I don't knowingly have any of their lenses, but they are probably good as well. Other vendors include Thorlabs, Newport, and Edmund Optics, among others.



Oct 04, 2016 at 08:48 PM
hiepphotog
Offline
• • • •
Upload & Sell: On
p.15 #8 · p.15 #8 · Kolari Mod's Improvements: A Summary of Tests


Brandon Dube wrote:
Hi,

Some may recognize my name from my work with Olaf Optical Testing / LensRentals. My primary engagement in optics is as an aberration theory researcher / student. My perspective may be of some interest here.

The coverglass generates aberrations because not all rays in a given cone, fan, bundle, or "pencil" of light that forms a point in the image hit it at the same angle if the beam is not from infinity. Snell's law says that a ray that is incident at an angle different to another ray is going to go somewhere different to the other ray. Thus,
...Show more

Brandon, any guidance in choosing the right front filter would be much appreciated by me, and no doubt by a lot of people in here. A more complex simulations would help to choose the right filter instead of having to experiment with the right focal length.



Oct 04, 2016 at 09:25 PM
Brandon Dube
Offline
• •
Upload & Sell: Off
p.15 #9 · p.15 #9 · Kolari Mod's Improvements: A Summary of Tests


hiepphotog wrote:
Brandon, any guidance in choosing the right front filter would be much appreciated by me, and no doubt by a lot of people in here. A more complex simulations would help to choose the right filter instead of having to experiment with the right focal length.


Ideally, the lens is afocal. If it must have a focal length, then its "bending" is a question of optimization and depends on the particular design you are trying to fix. If I use a biogon in Code V, depending what I feed the optimizer it wants one of two scenarios, either a meniscus lens as far away from the front element as possible with roughly equal and negative radii (optimize at large aperture targeting the edge of the FoV), or essentially a window/regular filter (optimizing at f/11).

The bending is the other way (think fisheye front element) for a retrofocus lens. The retrofocus optimization is very difficult because the lens designs are unstable and prone to ray trace failure.

For an actual fisheye, the optimizer wants to stack another meniscus-like element on the front.

Most of the time the optimizer seems to end up increasing the amount of astigmatism. I don't know if this is to do with the lenses I am choosing being particularly well corrected (they are, but... shouldn't matter) and it not really having a choice in the matter or what. Code V can't (or I don't know how to) do an actual perfect thin lens, so I can only do real lenses.

If you have any particular lens in mind, which has its optical prescription available, I would be glad to take a look at it.



Oct 04, 2016 at 11:11 PM
HaruhikoT
Offline
• •
Upload & Sell: Off
p.15 #10 · p.15 #10 · Kolari Mod's Improvements: A Summary of Tests


Brandon Dube wrote:
You can toggle your simulation between retrofocus and symmetric, with a perfect master lens, by separating the aperture stop from the thin lens and putting it a bit in front of it.

Field curvature you can do by placing a meniscus lens with equal radii right on top of the cover glass. It will contribute nothing to the focal length and image location, but adjust the curvature of the field.


Thank you so much Brandon,

With your advice, now I confirmed OSLO free version can also handle retrofocus type ideal lens by separating the aperture stop.
I also see the additional meniscus in front of the cover glass controls field curvature in the simulation.
However, even if I could get optimal meniscus, I'm not sure how to attach it between cover glass and the primary lens's last element.



Oct 05, 2016 at 06:13 AM
Brandon Dube
Offline
• •
Upload & Sell: Off
p.15 #11 · p.15 #11 · Kolari Mod's Improvements: A Summary of Tests


HaruhikoT wrote:
Thank you so much Brandon,

With your advice, now I confirmed OSLO free version can also handle retrofocus type ideal lens by separating the aperture stop.
I also see the additional meniscus in front of the cover glass controls field curvature in the simulation.
However, even if I could get optimal meniscus, I'm not sure how to attach it between cover glass and the primary lens's last element.


The meniscus isn't really to serve as a working corrector. Of course it would be lovely if we could put optics inside the mirrorbox for correction. If we could we would always do so.

The perfect thin lens model captures the added behavior of the coverglass, but not the behavior due to the aberrations of the master lens. By adding the meniscus on top of the coverglass, you can add field curvature to simulate that aberration in the master lens.



Oct 05, 2016 at 10:09 AM
jackcklam
Offline

Upload & Sell: Off
p.15 #12 · p.15 #12 · Kolari Mod's Improvements: A Summary of Tests


Here I have a few questions regarding the Kolari mod.
I have been reading loads about it on this forum, but I still haven't found conclusive answers to some of my questions.

1) People seem to have different opinions about the v2 vs v3 glass. What are the pros and cons of each version and why should I pick one over the other?

2) I read that the mod has minimal detrimental effects on Sony FE glass. The 55/1.8 is found to be only slightly degraded. How about the zooms? Zooms are much more complex by design, and I wonder if anyone has done a conclusive test with Sony FE zooms?
In particular, I'm most interested in the 24-240mm.
(I know this lens is probably not anyone's favourite child on this forum. But I need a superzoom for location scouting).

3) Is IR pollution something that I should worry about after the mod?

4) Is the color shift just a simple, linear shift of color balance that can be fixed with the warm/cold & magenta/green sliders at lightroom (which, in my understanding, behaves like an optical filter)? Or is it something more complex that makes color profiling a mandatory step in fixing it?

thanks a lot,

Jack







Oct 08, 2016 at 05:54 AM
mdemeyer
Offline
• •
Upload & Sell: Off
p.15 #13 · p.15 #13 · Kolari Mod's Improvements: A Summary of Tests


Jack,

To your questions, I can comment on 1, 3, and 4 but don't own any Sony lenses so no personal experience on 2.

1. The only difference between v2 and v3 is the addition of an anti-reflection coating to the filter. This coating also improves the UV rejection somewhat. From my extensive experience with both, the addition of the A/R coating is significant to remove the greenish reflections that can occur in the uncoated filter when scenes have very bright point sources of light in otherwise dark scenes. I shoot a lot of architectural interiors, so that matters to me. Your mileage may vary. Regarding the improved UV filtering, I keep UV filters on my lenses so have never seen any difference. If you don't use UV filters, I assume you could notice this difference.

3. No problems with IR that I have seen or seen documented.

4. The change to the white balance can be corrected with the controls in Lightroom, but I (personally) usually prefer the custom profiles, which are easy to create. Even though I usually use the custom profiles - especially for outdoor landscape where I find it easier to get the sky blue better, I also often use Camera Neutral (I find most profiles to be over saturated) for indoor shots with people because I find it a bit more forgiving of weird mixed lighting situations.

Regarding the Sony lenses, there is much theoretical understanding that says to the extent they are optimized for the stock Sony filter thickness, they will be adversely impacted by a thinner stack. However, it is hard to generalize about the performance since it depends on a variety of factors, primarily exit pupil distance (shorter has more impact) and speed (shooting wide open has more impact). Beyond that you have to test with a particular lens.

Michael

jackcklam wrote:
Here I have a few questions regarding the Kolari mod.
I have been reading loads about it on this forum, but I still haven't found conclusive answers to some of my questions.

1) People seem to have different opinions about the v2 vs v3 glass. What are the pros and cons of each version and why should I pick one over the other?

2) I read that the mod has minimal detrimental effects on Sony FE glass. The 55/1.8 is found to be only slightly degraded. How about the zooms? Zooms are much more complex by design, and I wonder if anyone has
...Show more



Oct 08, 2016 at 11:32 AM
thompsonkirk
Offline
• •
Upload & Sell: Off
p.15 #14 · p.15 #14 · Kolari Mod's Improvements: A Summary of Tests


I searched the thread to see how the Kolari mod works with Noctilux (f0.95; or also f1.0). But I found nothing. Does anyone have info/experience?

Kirk



Oct 26, 2016 at 11:06 PM
beetlephoto
Offline
• •
Upload & Sell: Off
p.15 #15 · p.15 #15 · Kolari Mod's Improvements: A Summary of Tests


Late reply, but I use the Noctilux f1 (v3) with my A7rM. I found little differences vs the stock A7r. My findings are that the lens exhibits a little bit less field curvature with the thin stack, and that CA is less pronounced.

I actually initially came back to this thread to ask a question about sensor cleaning. I have the v2 glass, and was wondering what liquid I should use to clean the sensor, and whether the regular Eclipse liquid was OK? the blower is not enough to remove all dust spots unfortunately.

Thanks!



Nov 15, 2016 at 05:34 AM
beetlephoto
Offline
• •
Upload & Sell: Off
p.15 #16 · p.15 #16 · Kolari Mod's Improvements: A Summary of Tests


Self-answer: Ilija replied, Eclipse liquid is recommended.


Nov 16, 2016 at 11:22 AM
mdemeyer
Offline
• •
Upload & Sell: Off
p.15 #17 · p.15 #17 · Kolari Mod's Improvements: A Summary of Tests


Not sure if I can reply to an Archive post, so here goes...

I've been doing some tests with the UT-modified Kolari A7. On an international assignment so limited ability to do things with too much science involved, but I did try making a tweaked Camera Neutral profile for the camera as follows:

Process: 2012
Profile: Camera Neutral

Red Primary
Hue +35
Saturation -5

Green Primary
Hue 0
Saturation 0

Blue Primary
Hue +10
Saturation +25

This is an experiment, but a reversible one, so please give it a try and let me know what you think. With this profile on the UT mod camera a daylight balance shot seems close to 3200K and -25 color balance settings.

And greetings from Beijing (where I'm on a 1-year assignment). Any togs here in Beijing who speak English, PM me!

Michael

PS. I did land an M10 before leaving the country, so will try to post some comparisons once the air in Beijing clears enough for some architectural shots.






Kolari A7 UT Profile




Aug 14, 2017 at 09:52 AM
1       2       3              14      
15
       end




FM Forums | Sony Forum | Join Upload & Sell

1       2       3              14      
15
       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.