How Do Beautiful Lens Flares Like This Form?

[Peter Ellner]

I love the beautiful lens flares in the form of radiating lines of light extending around the sun in some shots from the trailer for The Tree of Life: http://trailers.apple.com/trailers/fox_searchlight/thetreeoflife/

 

But I am very curious as to how that effect is produced by the camera, what causes it, etc. Most importantly, how can that beautiful look be achieved, not just any camera lens pointed towards the sun will produce that effect.

 

Here's another example, but in photo form:

 

Thank you!

[Phil Rhodes]

I've never been sure about that myself. Diffraction around the edges of the iris blades, maybe?

[linus rosenqvist]

I think that the flare in The Tree of Life is digital. Optical flare from video co-pilot is one way to go. I do however think that it would be intresting to know what kind of lens/lensobject that would render this kind of flare

Edited May 27, 2011 by linus rosenqvist

[John Holland]

I havent seen the film but i am sure the flare isnt a digital one ! Just a well shot film from a Dp who knows what he is doing .

[Ben Syverson]

It's called a "sunstar," and it's caused by the shape and number of the iris blades when stopped down. Typically you'd see sunstars starting at about ƒ/8, becoming progressively more prominent as you stop down. But it varies by lens. If you stop down a lens and look at the aperture, the more "spiky" or angular it looks, the more apparent the sunstar will be.

[David Mullen ASC]

I'm not sure why someone would say that this was a digital effect when it's a very common photographic effect when the lens is stopped down and you are pointing into the sun.

 

I did a whole series of flare tests for that S.A.L.T. test on RedUser by waving a clear light bulb around the frame against a black background and almost every lens ever made, spherical or zoom, gives you some kind of starry flare from a point source.

[Guest Ollie Walton]

I'm not sure why someone would say that this was a digital effect when it's a very common photographic effect when the lens is stopped down and you are pointing into the sun.

 

I did a whole series of flare tests for that S.A.L.T. test on RedUser by waving a clear light bulb around the frame against a black background and almost every lens ever made, spherical or zoom, gives you some kind of starry flare from a point source.

 

 

agree with david, can use anamorphic lenses or.. stop down , and get more dynamic range using knee settings or other. can do it in post, takes away the professionalism of it.

[linus rosenqvist]

The reason i said it might be digital was that on that trialer it actually looks like its moving a little bit strange (but looking at it again, its pretty clear that its the trees that effects it, making it look a little of) I know that this can be achived in camera as well but its also very easy to add in post. I think that today its pretty easy to get fooled by digital flares however your knowledge is alot greater than mine and if you are sure that it isnt digital I take your word for it.

[Karl Eklund]

Slow shutter speed will also create the same effect, but then we are talking shutters like 1-30 seconds... Obviously not so great for motion, but doing timelapses might work well, if the lens doesn't create the effect.

[Stuart Brereton]

agree with david, can use anamorphic lenses or.. stop down , and get more dynamic range using knee settings or other. can do it in post, takes away the professionalism of it.

 

The flares in question are not anamorphic flares, but spherical.

 

Slow shutter speed will also create the same effect, but then we are talking shutters like 1-30 seconds... Obviously not so great for motion, but doing timelapses might work well, if the lens doesn't create the effect.

 

Flares are an optical phenomenon. Shutter speed has no effect on whether a lens flares or not.

[Chris Keth]

Slow shutter speed will also create the same effect, but then we are talking shutters like 1-30 seconds... Obviously not so great for motion, but doing timelapses might work well, if the lens doesn't create the effect.

 

 

 

Perhaps you're thinking this because you've seen it most often in long-exposure photography? It happens when the iris is closed down pretty far which is done pretty often for that type of photography, even at night.

 

 

[Chris Keth]

I've never been sure about that myself. Diffraction around the edges of the iris blades, maybe?

 

 

I'm certain diffraction is the culprit, or at least part of it. I think what makes it show more at closed-down stops is that there is less light allowed through the iris to drown out the diffraction star. It's a physical phenomenon that always happens, but I think at larger apertures we just don't see it because of a greater quantity of light allowed through. There is probably also a measure of pinhole effect where the diffraction stars are concentrated in a smaller area than when the aperture is large.

[Dom Jaeger]

It's a physical phenomenon that always happens, but I think at larger apertures we just don't see it because of a greater quantity of light allowed through.

 

I've noticed that the iris apertures of some older lenses form a star shape as they open, perhaps to facilitate this effect at larger apertures?

[Chris Keth]

I've noticed that the iris apertures of some older lenses form a star shape as they open, perhaps to facilitate this effect at larger apertures?

 

Possibly but I think that's more likely a practical matter. If you want an aperture to be round, you need many, many aperture blades. I have an old lens for 8x10 with 64 aperture blades and it's round as can be the whole way down. The problem with that, is one of delicacy and space as the lens gets smaller. The shutter on that 8x10 lens is 5" in diameter. There is plenty of space to work with. It's much more difficult and expensive to manufacture an iris mechanism of that quality when the iris is only an inch across, for instance. The other problem is that even if you can make it, it's so delicate the lens must be overly babied to avoid being in for service ever other week.

[Chris Millar]

Possibly but I think that's more likely a practical matter. If you want an aperture to be round, you need many, many aperture blades.

 

Or use waterhouse stops:

 

 

 

 

or for instance....

 

 

[Chris Keth]

Or use waterhouse stops:

 

 

 

True, if you can find lenses that take them. I've never seen a lens made for a motion picture camera that takes waterhouse stops. I have 3 or 4 other lenses that do, but they're all well over a century old now.

 

 

[Chris Millar]

True, if you can find lenses that take them. I've never seen a lens made for a motion picture camera that takes waterhouse stops. I have 3 or 4 other lenses that do, but they're all well over a century old now.

 

 

 

If I recall correctly rather than using an iris the kinoptic wide angles use slots for ND, not sure where they are in terms of where the iris usually sits, but maybe a waterhouse stop could be chucked in there ?

 

With the right tools and brains/balls ratio you could open up any lens and put something in there yourself too

[Dom Jaeger]

Possibly but I think that's more likely a practical matter. If you want an aperture to be round, you need many, many aperture blades.

 

Perfectly round, yes, though I've seen them come pretty close with as few as 16. Most iris designs just form polygons or rounded polygons. In contrast, I'm talking about a very specific star shape that forms as the iris approaches wide open (and sometimes using many blades). Quite a few older lenses have it and I'm sure it's a deliberate design. The only modern cine lenses with this kind of iris are Cooke S4s (as far as I know). I asked Cooke about it but never got a specific answer.

 

Or use waterhouse stops:

 

Fine if you don't need an aperture that's between stops. Or you don't want to change aperture during the shot. Interesting shapes though!

[Chris Millar]

Fine if you don't need an aperture that's between stops. Or you don't want to change aperture during the shot. Interesting shapes though!

 

I cut them out of cardboard for my 18" Voigtlander - for stills yeh sure, but same difference ;)

[Karl Eklund]

Flares are an optical phenomenon. Shutter speed has no effect on whether a lens flares or not.

 

Flares are, but not CCD blooms, and blooms I guess can be affected by the iris blades as well.

 

Perhaps you're thinking this because you've seen it most often in long-exposure photography? It happens when the iris is closed down pretty far which is done pretty often for that type of photography, even at night.

Yeah, I have only seen it debated about long exposure photography and not motion photography.

 

Here they talk about blooming.

http://www.digital-photo-secrets.com/tip/1175/how-to-avoid-blooming/

 

 

Anyways, I need to get a hold of a still camera, and test to see if shutter changes anything in bloom/lens flares, and obviously keep the f-stop the same. Because I realize that usually it is the f-stop that is closed down and the shutter is lowered to compensate, and people might then give the "shutter credit for the bloom/flares", when it is the lens that should get credit. But, when I can grab a hold of one, i will try to test and see.