Hello! Im eagerly awaiting my ordered Aivascope, and in the meantime wondering about the character of the lens. Most tests ive seen is with very imperfect lenses, which makes it hard to see what character is from the taking lens, and whats from the scope. Have anyone tested the Aivascope with a very neutral, perfect and accurate lens to try to identify the character of the Aivascope? Does it mute certain colors or shift cool/warm etc? How is the CA, and which colors of abberations does it have? I for instance dont like purple CA, but dont mind yellow/blue/green. Does CA from taking lens and scope mix? Does a lens with very smooth bokeh yeild smoother bokeh with the scope than if youre using a taking lens with busy bokeh? or is the Aivascope glass itself mostly accountable for the look? In other words, to what regard does bokeh of the taking lens take effect with the Aivascope? Since focusing is done with the scope, does it kind of override a lot of the focusing characteristics of the taking lens? Does for example the cats eye shape or perfect Circle shape of the bokeh of the taking lens matter to the Aivascope, or is the Aivascope’s bokeh doing its own thing regardless?

tobiaseek Hey Tobias! All very legit questions which we will try to answer in depth, and we hope down the road others will chime in to add things to this interesting topic. Generally whether you shoot in a all in one anamorphic lens or a anamorphic adapter, there’s always a spherical lens design and cylindrical part. In most Anamorphic lenses which yield the most distinguished anamorphic look the anamorphic element is placed in front, and thus is the first pass when the light comes in, the only thing that sits in front of it (some designs work differently) is the variable diopter. In the case of the Aivascope 1.5X anamorphic adapter focus is done with a variable diopter. The scope is tuned for infinity focus and if you rig it in front of the taking lens you need to make sure the taking lens also is adjusted for infinity focus, which is most of the case not where the infinity marker hits its mark (most taking photography lenses are badly bad focussed), so take a very distant object (1km away) and open up the aperture, use a digital magnify on your monitor and make sure infinity focus is nailed, preferably put some electrical tape around the taking lens, so it stays where its at. (not too much tape because you might want to redo this after a while because things can go off calibration), now its the variable diopter which does all the focussing work. we call this a ‘ single focus solution’ . It sounds like we are drifting away from answering your questions, but its important to have a good understanding of how light enters your optical formula. Because the anamorphic element squeeze’s the image by 2X cilindrical in one direction, the aperture will be oval as well. NOw let’s get this quickly overboard and if you need the exact scientific reason for it make a other topic… 🙂 bokeh works like a squared formula, so even though you desqueeze the 2X image by a number of two, the bokeh is still oval because its affected by a way bigger number (squared) than the 2X forumula. if you would put the anamorphic element behind the conversion point (aperture) it would just be squeezed by 2X as a whole and after desqueezing the aperture will return to a normal round circle. SO anamorphic boke is not 2X but instead squared, and by desquezing it 2X you still have a somewhat oval, depending on the dof the bokeh can look more squared or almost round, this is effected by both the distance to foreground, the focus point, how far away the distant object is, the squeeze ratio of the anamorphic lens, the f stop and of course the focal length of the taking lens. PLease note that the anamorphic lens only produces the squeezed ratio, spherical bokeh is squared in its behaviour, thus a bokeh bubble of a 80mm lens with the same f stop as a 40mm is squared in terms of size, but since the anamorphic lens comes in play its squared with a offset in regard to the squeeze ratio of the lens. Hope this is somewhat understandable. However if you want to render perfect circular (oval) bokeh you need to make sure that your taking lens has a circular aperture, most lenses wide open do have a circular bokeh but once stopping down, the amount of aperture blades and their system (you have different aperture systems, like rounded blades, flower shapes triangles etc.), will determine how circular thus how perfect your ovals will look like. Generally in cinema lens design (look at Cooke) a high aperture count is not really in its tradition and it can effect mtf (sharpness), due to airy rings, refraction , hence most cinema lenses (except lomo standard speeds) will have blocky bokeh once stopped down, but lets leave these discussion in other topics. Generally a lot of photo lenses do have a high blade count including a rounded blade system. The famous 17 blade Helios (originated from biotar 58mm) , are very popular among anamorphic shooters. Generally stopping down on a taking lens will yield the best results when shooting anamorphic, thats because both the diopter (which is designed to work the best for one focal length in the case of the aiva a 50mm) and the anamorphic lens will cut off some high frequencies, by stopping down, you diminish these effects. But once stopping down a lot of lenses are not circular in terms of bokeh anymore. Thats why there are some lenses better for anamorphic shooting than others.. I myself am much more obessed with 3d depth distortion of anamorphic lenses and waterfall bokeh, than perfect circular ovals.. On top of that most anamorhhic lenses have cat eye bokeh (cut of bokeh by the housing ) on the edges, and slanted bokeh (only the ovals are straight in the middle but rotated on the sides) , this is a combined effect of both the anamorphic element and the spherical lens. Due to the amber coating on the Aivascope (both front and back element), the scope will render somewhat warm, especially when a light source hits directly the glass, but again this is a combined feature of the coating of the taking lens and the Aivascope. If a taking lens has green coating, and the aivascope amber, than there will be secundary flares in green. I prefer to use lenses that are either single coated or multicoated with a balanced coat that has magenta green and yellow, and complementaries, that works super sexy together with the aiva. The aivascope doesnt add much chromatic abberations, its a nice design, but generally if your taking lens has a lot of ca, the aivascope might articulate this more. BUt Ca with tha aiva is mostly coming from the taking lens, if you take a taking lens with lack of ca you probably find not much of ca in your footage. \ SO for bokeh, the scope mostly adds a squeeze ratio to your bokeh with a squared law, the same as bokeh works with a spherical lens but with the added squeze ratio of a anamorphic lens in that direction. But due to distortions (barrel mostly) the ovals are kind of messy on most scopes, and the wider you go or the further you put the taking lens away from the scope the more this is pronounced. About the cat eye bokeh which is mostly produced inside the taking lens due to lack of coverage, the fun part here is that when you infinity focus on a lens, the element is as close to the sensor as it can be so mechanicala and optical vignetting are at its worst, if you focus closer with the taking lens (without the scope) the front element is further away and mechanical vignetting is lower, but again, just look for a nice taking lens that has characterstics that you would like to see to leak through in your anamorphic formula, a perfect lens doesnt exist in that regard, and that makes most Anamorphic lens collectors and shooters so obsessed. always on the haunt for that perfect lens! If you happen to have more questions . please shoot! https://gafpagear.com/shop/aivascope-1-5x-anamorphic-adapter/

A lot of great info about anamorphic in general there! Perhaps i wasnt specific enough about bokeh, but one thing i feel is left unanswered is to what degree the scope will change to look of the bokeh (Other than the oval-ness ofc). Heres two examples of smooth vs busy bokeh rendering. The left is very clean, whereas the right has sharp, bright and colored edges. Would the final image with the scope retain this of the bokeh from the taking lens (clean or busy), or does the scope more or less override this with its own character? This next example is about shape image 1 is the aivascope in the left corner on full-frame 16:9 cropped to 2:39 Image 2 is smooth oval shaped bokeh (generic image) Image 3 is the corner of a spherical lens with nice circular bokeh-shape. Could you achieve the look of image 2 (more or less) across the frame with the aivascope paired with a taking lens like in image 3? Or will the scope ultimately sharpen up the top and bottom part of the bokeh balls/ovals like in image 1 regardless? The reason for asking really is control. To know if one could go for a clean or dirty look in terms of bokeh shape with one scope! If the answer is no, i guess a workaround would be sensor crop factors. Shooting fullframe will give you the most vintage look using the entire scope, where as cropping to S35, M43 etc. and compansating with a wider lens would crop out the the corners of the aivascope, leaving you with a cleaner bokeh rendering (and not to mention distortion), right?

tobiaseek Hey tobias in general a scope is altering the bokeh to a certain degree, and indeed not only affecting the shape of the bokeh but also for instance abberations, but I would say its 95% the taking lens and 5% the scope thats altering these characterstics, most of the character of your taking lens will translate through, but ofc it will be distorted both by barrel distortion and uneveness in the squeeze. Generally a wider lens will show more of the scope, and putting the scope further away from your taking lens will yield the same effect. cropping is not really how you should look at it, its in a way 100% true but I think you overcomplicate it. A lens has a certain AOV (angle of view), which is determined by 1focal length 2 sensor size , and of course some lenses don’t cover certain sensor sizes and will vignette. with going full frame (i guess you refer to photography full frame 36X24 which is referred to as lf in cinematography), you not necessarily get more vintage character, the amount of the coverage of the scope is determined by the AOV of the taking lens and the distance from taking lens to scope. In lf a 50mm lens will yield the same coverage as a 35mm in s35, only difference is that the 50mm will have thinner DOF at the same given f stop, but the AOV will be the same. The aivascope starts to bottle neck around 50mm on a LF sensor like 36X24 (or a 24X16 sensor with a 35mm lens), and it depends on the spherical lens design, how good of a coverage you get, some 50mm have a pancake design (look up the 50mm Nikon pancake its beautiful and cheap but low aperture blade count), and that will be ideal in order to get the taking lens a bit closer to the scope. But about every scope will have over compression on the edges and barrel distortion, this will effect in way more oval on the sides, and slanted oval. Even though you could correct side compression in post (you need really good plugins or make custom settings within fusion like I did), but it wont correct bokeh fully due to the previous explained square law of bokeh. In order to get bokeh like in your example 2 you need a slightly longer focal length in LF like a 75-85mm (or 50-65mm in s35). And you need a lens that produces a bokeh without the onion rings or onion skin and the Aivascope will leave it relatively untouched and adds the Waterfall bokeh magic to these bokeh bubbles. I might be able to invite my experienced anamorphic friend Tomas Overtoom, he has a lot more to add!

Jerome Alright, that totally makes sense. Somehow i thought sensor cropping would work, but ofc the aov stays the same when compensating focal lengths! But cool. About pancake lenses. Does them being small matter, or are they recommended because they generally have the front element close to the tip of the lens? Or is how close you can get the front element of the lens to the back of the scope all that really matters?

tobiaseek If you have a 50mm lens the lens is probably longer than 5 cm yet the focal point (where lines are inverting) is 5 cm. A pancake is much closer in length in relation to the focal length . Look at the scope as a window frame. The closer you get to the window frame the less you see the frame , if you shift to a longer narrow lens this will also help. But still if you take a long focal length and you stand far enough from the window you can still see the frame. Hope this helps.

Alright alright! About aperture and sharpness. Could you get a pretty sharp image using a taking lens at a fast aperture like f1.4 with the aivascope, given that the taking lens is tack-sharp at that aperture? If no, could you work around that with sensor size? (shooting at a slower fstop, yet with equivalent dof with a bigger sensor)

tobiaseek So basically you ask if you can get a similar dof and angle of view by taking a longer focal length a bigger sensor and a deeper f stop? Yes the same math applies as with spherical lenses. Watch out for mechanical vignetting . When you take a taking lens with much bigger entrance glass than the Aivascope rear end (aiva rear is 38mm) you get mechanical vignetting. This is a effect which doesn’t result in vignetting (dark corners) but the aivascope will then prevent the taking lens to gather light from its outer edges. Basically giving you the effect when shooting at a deeper f stop. Can you get a decent image with a 50mm lens at f1.4 on the aiva? Most lenses at f 1.4 are not sharp yet they present themselves as sharp because the contrast between sharp and unsharp is very big so they look sharp by comparison with the bokeh. But if you could have sharp lens the scope and diopter of the scope will take quite a bit of a bite as well. That’s why you never see scopes that perform well below 2.8 . Atlas lenses only sharpen up around f4.0 /f5.6 . If you want I can go a bit more in depth how this works.. generally the aivascopes are considered to be very sharp, I have seen quite a good examples at high f stops which looked nice, field curvature of the taking lens is also a concern look for a tiny ultra sharp taking lens. Leica m lenses are great . The summilux ones. The appearance of sharpness can look different when knowing what your doing. Light can make objects appear completely different, work with high contrast ratios and avoid flares. Hope this helps.

Jerome I think you misunderstood what i meant, but with good reason. From my initial testing with the aiva, i was under the impression that it couldnt handle fast apertures. So i assumed - and asked for confirmation, (if that was the case) - would a larger sensor solve that, if you stop the lens down, get equivalent dof, and also a sharper image. This question arose from the thought that the aivascope just couldnt handle fast apertures, which ive found is wrong, so it can be dismissed! So from another test i just did, there was very little difference between sharpness with and without the scope, at max aperture. I think my previous tests has been somewhat '‘wrong’' due to wrong infinity focus on the taking lens. Taking lens focus seems to be a very delicate thing one has to get right. Heres a test i just did, left is spherical, right is with aiva. Very impressive indeed! Barely any sharpness difference, just a more pleasing image one could say. No need to blur in post 😛

tobiaseek One thing i also now noticed, how the CA on the metal with the aiva on this lens has a red/orange hue, reminiscent of halation. Thats cool!

tobiaseek hey Tobias, great tests your posting here! Thanx! Yes in general the Aivascope is among the sharpest scopes that exists. Like you said, infinity focus is very important and only the slightest offset can take the bite out of your image. Another potential deal breaker is the variable diopter which is made for only one focal length, namely around 50mm mtf. So if you go much longer (85mm and above) you might need to stop down a bit. The halation from your last test is indeed reminiscent of film stock halation (due to bounced back light on the film backing which bleeds into the red and sometimes blue layer). Looks super cool! Keep it going. Can you tell us which taking lens you used for this test?

AlexanderAlucard the sides have quite a bit if overcompression. Which can be an esthetic but I always take it out with fusion. If your in resolve I can help you with that..

AlexanderAlucard

Jerome I like the distortion, it feels a little psychedelic.
Hoping the 28mm will give me the same effect on Varicam sensor. The 35mm on Varicam doesn’t have anything interesting going on the sides. She’s pretty bland 🙁

55mm M. Zuiko on Super3

JamesC

I borrowed one of these a few days ago to test on the mavo edge.. I used Leica r 50 and 90..worked great but was getting some vig on the 50mm at 8k.

I was wondering what’s the best way to mount? I had a rail set up that just pushed shading the taking lens.. is this the best solution or should I be doing anything different ?

Many thanks

Jerome

JamesC hey with a 50mm leica the front element sits quite deep. Try a pancake lens!

Kinoliker

Hey, can i mount the lens support clamp upside down where its above the lens and would it still give support? like with this top handle?

Jerome

Kinoliker technically you can yes. However I don’t think you will find a top mounted lws rod support with the right standard (85mm center rod to center lens). But if you do so yes it will work great.

« Previous Page Next Page »