Forum Replies Created
-
Posts
-
Hi peace 7,
sorry to reply this late, but I’ve been out for a while…Thing is that most people (found on web discussions) are taking some virtual visual perfection as a merit of aesthetical value. What’s so more, they would consider anything subpar in comparison to highest MTF as a pure trash and useless tool.
In my hummble opinion, it is a stupid approach at best.
It is like saying that only cotton canvas, camel hair brush and oil colors are acceptable for any painting artistic attempt. Things less perfect such as charcoal or even soil on a finger can’t be used for any serious art…
Regarding lens experiment, I am struggling with mechanical and optical workshop. I was able to calculate few (IMHO) interesting lenses, similar to those in this thread, but my prototype production capabilities went to zero, since my friend – optician recently get retired.
I am still playing with some legacy optics that I have at home, using super cheap projection lenses as a base, but those experiments are pure fun and trial/error method. If I make something interesting again, I will post it here, and I encourage all others to do the same.
Let’s celebrate imperfection!!!
Hi,
I tried both Canon FD 50/1.2 L and Mitakon. Mitakon is special lens with very specific rendering, while Canon L is great lens with very modern rendering. Both are great but Canon is more versatile IMHO.
Cheers,
ViktorHi Arthur,
I learned my lesson with field curvature when I shot group of managers with Canon EF 24-70 at the wide end, focusing on the center. At f/16, I thought that I am safe from FC, but few pals who were forced to be on the frame edges, weren’t happy with their blurred and distorted versions… and thus I lost great client. (It was beer brewery, whic translates in many events with free beers, nice chicks, etc :-()
In some critical situations (it usually works for landscape), you can try focus stack (5 shots should be enough) for outstanding sharpness across the frame. In other situations, you might find focusing on the edges at f/8 and smaller sufficient for good results.
I don’t have Loxia any more, but I quite liked it when I has.
Cheers,
ViktorHi Arthur,
thank you for your comments and observation. When we speak about field curvature and lens design in general, we should consider few things in addition…Most important is focal length. While 35mm is sometimes considered as a standard focal length, in reality it is quite wide and wide lenses have to deal with many aberrations. Like with most other optical designs, it is always question of priorities and compromises.
Biogon is indeed known for “rather” flat field curvature in comparison to Distagon e.g.,but it is even better known for low distortion. (Biogon was originally based on Biotar, which was another way to call Planar and Planar name was based on planography – prints from flat surface)
Distortion is usually one of the most problematic aberrations with wide angle lenses and Loxia 35/2 shows extremely good control over it. However, good control of the distortion needs some compromises and some of them are partially paid off in field curvature while other partially reflects in increased sagittal oblique spherical aberration (butterfly effect).
Thing is, that wide angle lenses which seems to perform better in the extreme corners then Biogon/Loxia, usually have so called mustache (irregular, complicated) distortion. This type of correction is today very popular (thanks to much more precise and cheaper production of aspherical elements), especially when combined with automatic software correction. In most situations, image looks sharp from edge to edge with such a lens. But looking more carefully for the critical distortion control (when shooting architecture e.g.), one might be rather disappointed.
Loxia/Biogon would perfectly sharpen the extreme corners from f/11 onward, but we are facing here problems with diffraction on modern sensors.
While I understand your concern, it won’t be easy to find perfectly “flat” 35mm lens yet. As you can see from my tests, only Sigma 35/1.4 Art will beat Loxia in that regard (showing indications of mustache distortion though) and I was quite impressed with new Zeiss 35/1.4 Distagon ZM on Leica M body, even if Distagon usually means quite heavy field curvature.
In my review Loxia shown better corner performance (by very slight margin) then Sony FE 35/2.8 from f/4 onward, and except Sigma and new Distagon ZM, neither of tested lenses did come even close.
Cheers,
ViktorHi Juan,
thank you for moving here, I should be able now to respond faster 🙂I should probably make a post about my suggestion how to test legacy lenses, but so far, I can only give you basic ideas…
First and most important is to think about what are you willing to test. It is not good/possible to test all aspects of the IQ at once.
1. Absolute sharpness (sharpness in the mid zone)
This test might tell you lot about the condition of your lens. Most older lenses that had been opened and cleaned from fungus or some sort of haze, would have some level of “blooming” or “glow” wide open. If you find that the amount of this “glow” is high, your lens is most probably subpar (especially both Planars) and you should consider return.
To test the lens, find high contrast area at a distance of about 5m, that will cover mid frame (around 20% of the frame in the center). Use tripod, focus magnification, remote control or self timer (2s should be enough), use electronic first curtain set to on, If you are using lens or body with image stabilization, switch it off. Shot in a good light (sun from the side, ideally overcast). Shot in A mode, set ISO to 100, WB to preset value (daylight, cloudy etc.) open your lens to its max aperture and repeat focusing and shooting at least 10 times. Shot in RAW (for better evaluation of the detail) and check your files in your RAW converter at 100%. Check only center of the frame. If you find at least one shot (out of 10) which is sharp enough and doesn’t show much glow there, chances of having good lens are very good.
2. Sharpness across the frame and de-centering
First, you need to understand “field curvature” and your lens native performance in that regard. Planar design is supposed to have low field curvature, or if you like – even sharpness across the frame.
To test your lens if its rendering suffer beyond native field curvature, you will need flat target at a distance of approx. 3-5m (for 50mm lens). This is tricky part, because 100% precise alignment (parallel to the target) is crucial for good results. People usually use brick walls, murals and graffiti walls. Most important is to try to perfectly align camera with the subject as mentioned above. You can use all sort of guides and tools to accomplish the mission, but take your time.
You are now looking for the even sharpness across the frame, so the aperture that you will select should be similar to the one that you will use in a real life situation. Wide open aperture will show rather field curvature issues than de-centering, to narrow aperture can hide any softness, thus I would suggest to go with f/2.8 or f/4.Use tripod etc. (all settings as in the previous test) and take few shots. You don’t need to take 10 shots like in the previous test, because you are not looking for absolute sharpness now, but rather for flatness of the focus field. Therefore 3 shots should suffice more than enough.
Trying not to move anything much, take SD card from the camera and upload files in the computer. Look how sharp is center vs edges and extreme corners. If you find that one side (left i.e.) is softer than other side, it could be that either your lens is de-centered or that you didn’t align your camera perfectly against the target. First try to eliminate second option by making few more shots with a camera panned slightly to the opposite direction (right in our case) and in the same direction (left). Rotate the camera by approx. 1°per shot. Check your files again. If in all cases the same side is still softer, de-centering is most probably the problem. If in some shots the sharpness improves on one side (left in our example and decrease on other side, you should try to align camera with target again. PITA? You bet. But if you really want to know what is the problem with your lens, you’ll need to go through the torture. Forget about DOF theory, it wasn’t made for pixel peeping at the recent resolutions…
If you suspect your lens being de-centered, to be sure do the following. Go back to your set-up and turn the camera in the portrait orientation by rotating it (on the tripod head) at 90° CW. Take few shots at the same settings as before.
Check the images in the computer. If you see that the softer part moved in the same direction as you rotated the camera (in our example, the top part of the image will be now softer than bottom), you have de-centering problem, but…it could be the lens as well as the adapter. If you have other lenses in the same mount, check if they shows similar pattern as the tested lens. If yes, you might need better adapter. If not… it is most probably the lens, but the problem might occur in the lens to adapter connection. Slightly deformed mount or similar problem e.g. To be 100% sure if the problem is in the lens optics, you will need to test it on the optical bench, which for most mortals is impossible.
Other tests:
You might want to shot against the sun or any other light source to check if the coatings are OK. (I had several lenses, on which dumb owners tried to clean some scratches or dirt and they almost completely removed coatings!) To test the lens, you don’t even need to make images, just pan and tilt around your light source and follow resulting loss of contrast. Both Zeiss Planar lenses should have very effective coatings, so you shouldn’t see large loss of contrast. (Lens flare are something else, they depend also on the sensor reflections, so don’t be too picky.Your lens might (and will) have many native aberrations, such as Longitudinal and Lateral CA, Spherical aberration etc. but if the above tests didn’t show some extreme anomalies, you can be sure that you have good sample. Best way to test lens for most aberrations (to see the character of the particular design) is to shot high contrast targets, such as ISO 12333 test chart e.g. but I don’t think that you need it though…
I wish you good luck with your Planars, and let me know which one you’ll decide to keep.
Cheers,
ViktorWell, thank you very much 🙂
My name is Viktor Pavlovic alias Verybiglobo. I am living in Prague, Czech Republic, where I have studied Film Academy many years ago.
I am working as advertising consultant, film director, photographer and photography teacher for the last 3 decades.
While I have shot around 200 TV commercials in my life, I found lately that photography is much closer to my hearth. That’s why I started to work on this site and I am really glad if you like it.
Thank you for your kind words,
ViktorHi JS,
Which 50mm lens for the A7 is one of those question where almost everyone will give you different answer. Bellow is the list of the lenses that I would recommend you to consider…
1. Zeiss Planar 50mm f/1.7 C/Y – Very good standard lens with nice coatings, good sharpness across the frame, nice colors. Bokeh can be smoother.
2. Canon FDn 50mm f/1.4 – One of the sharpest lenses in its category. Affordable price, clean sharp images in a good light, not too big or heavy. In the back light it can loose the contrast, so always use the hood.
3. Olympus OM 50mm f/2 Macro (not easy to find for 400 USD though), very sharp lens with unique rendering, one of my legacy favorites. It could be faster and cheaper.
4. Canon FDn 50mm f/1.2 – probably my strongest recommendation for you – except slightly lower contrast, this lens is very close to much more expensive L version.You’ll find many other preferring Minolta or Pentax, Takumars or Leica. Truth is that all of them are rather good, so it is more about individual preference of the subtle differences in the rendering (character) than about huge differences in the image quality.
You can also consider rather cheap Carl Zeiss Jena Pancolar 50mm/1.8 in M42 mount if you want to start with something good but don’t want to spend too much.
Cheers,
ViktorHi Marc,
To answer your question we should consider few things…1. Mechanical precision – if the adapter isn’t 100% accurate, user might experience several disorders – de-centering, focus shift, accidental light flare.
2. Material – Lens bodies are in general built from different materials with a mount being made from brass, steel, aluminum or even plastic. If the adapter is made from different material, in theory we can experience some deformations resulting from changes in temperature, or simply from mechanical intolerance. Cheap Chinese adapters, usually made from a thin metal, could be affected by large temperature oscillations.
3. Reflections – This is very often problem for any adapter, because in ideal case, inner surface of the tube part of the lens construction is part of the original lens design. It means that the level and character of the paint or special coating material inside the lens adapter, might not be best for all lenses that will be mounted on it. To reduce reflections to the minimum, one would need to know exact character of the light spread of the lens after the exit pupil (and distance to the film plane). In practice, cheap adapters have very basic matte paint inside and this can result in inner reflections, flaring and significant loss of contrast, especially toward the edges of the image.In general, every extra piece of equipment, means more potential problems, including production tolerances, QC and above mentioned factual challenges. Each extra joint in the light path can cause lot of troubles…
On the other hand, we are far from being defect safe with a native mount lenses. Automation of the lens production, which is extremely demanding process in terms of accuracy and precision, doesn’t come for free. We can experience lot of “duds” on the shelves, even with the most expensive brands such as Zeiss or Leica.
My advise is thus – test your lens camera combination prior to the purchase whenever you can, don’t think too much about IQ perfection, because you’ll never reach 100% anyway and finally, don’t take all those lens tests, especially so called scientific (based on test charts and MTF calculation) too seriously. Roger Cicala from Lens Rentals is the guy who has lot of experience and I find his articles to be among best on the Internet (if not the best). You should however always read his article to the end. Here is what he has to say in a conclusion of his article about lens adapters intolerances and problems (tested mainly on optical bench, highly precise optical instrument with its own caveats):
“What Does It Mean in the Real World?
Like a lot of laboratory testing, probably not a lot. Adapters couldn’t all stink or people wouldn’t use them. Like a lot of tests, you can detect a very real difference in the lab that doesn’t make much difference at all in the real world.”
I have also many Leica R lenses and I am happily using them on my Sony E mount cameras.
Cheers,
ViktorHi all,
Here is a promised key to the sample comparison in the review.
Sony A6000 + SEL 10-18/4 OSS are behind following images:
A, D, E, G, J and KSony A7 + FE 16-35/4 ZA OSS were used for the images:
B, C, F, H, I and LIn addition, here are links for two original size files posted on Flickr, that we haven’t process (apart of standard Lightroom preset).
Cheers,
ViktorThanks for yours analyze. We will keep camera+lens ID hidden for a bit longer, but your reasoning is very good.
Cheers,
ViktorHi Andrew,
let us know please if you succeed with the mod. I’ll be happy to know if it works for you.Cheers,
ViktorHi Andrew,
I got few minutes to try conversion on the type 44-2 in order to see if I can replicate your problem…As I was in hurry, I couldn’t take pictures, but I might have found the cause of your problem…
I believe that you twisted 180° – metal collar ring between front element and secondary elements group.In order of disassembling, you have first to unscrew the little safety ring with letters “Hellios, 44-2” etc. After that you should carefully twist the lens to drop the front group in your palm. (If you twist and shake it too much, second group can fold out too.)
Bellow front group, (that you should have in the hand now) and above second group, (in other words – between two glass elements) there is a metal collar ring that is supposed to keep the proper distance between both optical groups. This collar has wider part that should face upward, and tighter part that touch the second lens group.If you turned this collar (accidentally) other way around – so that wider part is touching the second group and front element comes on the narrower part of the collar, you will most probably keep proper distance between elements, but you won’t be able to screw back safety ring (with signs).
If you keep the original position of this collar, both lens optical groups will hard touch each other, but you shouldn’t have problem to screw back the safety ring (with signs). (At least I haven’t had a problem, except that it takes a bit longer to properly position safety ring).In this case however, I would expect problems with focus (didn’t have time to test), because glasses will touch each other and distance between them will be shorter than intended. You might try to make few spacers (from the harder black paper, or plastic folder etc.) to lift the front element just enough to fix the focus issue, but still not too much in order to screw safety ring back. I guesstimate thickness of just bellow 1mm. (Lenses will probably still touch each other but less than without spacers.
I hope it make sense, and if you need any further support, let me know.
Best regards,
ViktorHi Andrew, As soon as I find some time, I will try to convert one of those and will see if I can replicate your problem. To post images here, you need to have them somewhere on the web, such as Flickr, Picassa etc. and to embed them to the post. Embeding link differs between services, but it shouldn’t be too complicated.
Cheers,
ViktorHi MTeee,
p
you ate correct. That’s the same type of Canon FD lens that I used for the comparison above.Cheers,
ViktorHi Andrew,
what exactly type of Helios did you use? (44M, 44-2 etc.) I have most of them and can try to replicate your problem. If I succeed, I can make few pictures to suggest the solution, if not, we’ll at least know which types are not suitable for the mod.I can make the mod for you, and it will cost 99 USD with the lens in fully working order (+ shipping) but I am sure you can pass much cheaper with DIY.
Cheers,
Viktor
