A question about breathing For Mr. Duclos

[Björn Benckert]

I agree to an extent. I think that breathing is an organic part of films.

I don't seek to actively use it in my film grammar ... but its often there so I must account for it. That usually means being creative and trying to motivate or otherwise use the effect.

My ideal however is to have the lens not breathe ... then if I need movement as part of my grammar I can add it by ::gasp:: moving the camera.

Well for post / compositing and CG integration having the LDS is big deal... having a digitally stable source with only focus moving is also a big deal.

 

[fALk Gärtner]

Great to see that somebody else is thinking of this. I got some thoughts on your comments...




I can tell you from my 15 year as a flame artist, trying to figuring out even the zoom factor from "focus pull breathing" using tracking dots of a moving camera shot might be doable, but it's freaking hard, involves the higher level of 3D tracking. Breathing is not only "zoom" it's a "fisheye" warp thing going on also so to create a perfect flat source with only focus moving involves warping swell, do not think you have it in resolve. But do not know for sure do not use the program.

Having a perfectly flat source however is improving 3D tracking and 3D integration by a lot so getting the numbers out of the canon lenses and have it feed into the meta data is huge for post...

yes yes yes. First have a stable clean as possible image - if you have done all your integration and the client/DP/Director really needs a more natural feel then you can ALWAYS add in artifacts afterwards - and yes I regard breathing an artifact and I personally do not like it - but I also do not like grain that much
Getting breathing out in post is difficult takes away valuable time that you could spend making good looking vfx

 

[Marc Wielage]

I believe there are sections on lens design and theory, plus the causes for "breathing" problems in focus changes in past editions of the American Cinematographer's Manual -- a must-have publication that every DP should own.

 

[Matthew Duclos]

"Breathing" is a term used to describe the apparent change in the field of view while racking focus. It's an optical-mechanical design flaw. It's much more expensive to design a lens that does not breathe. A "floating element" is a very common method for correcting breathing in a lens. Most still photo lenses will breath a since manufacturers such as Nikon and Canon don't really care about making the image as accurate as possible. Can you tell the difference between a 45° field of view and a 46° field of view looking through your view finder? Probably not and nobody cares about a 1° difference when you snap a still frame. But when the field of view changes while focusing during a continuous shot, it's devastating to the scene. Watch Super 8 and look for the long focus pulls from infinity to close focus. The breathing exhibited by the lenses in that movie is horrible, made even worse by the anamorphic lenses. Cinema lenses do their best to reduce breathing and in most modern lenses, accomplish this very well. Try to find some breathing in a set of Master Primes. ...Good luck.
For example, keep an eye on the right edge of the frame as i run through focus. Sorry about the shaky image, this was a quick and dirty example.

 

[Ketch Rossi]

Matthew you tucked my words out of my mouth and bit me to it...

Oh and if you would to see the Leica Summilux C. Primes, you would be very impressed as well, Absolutely NO breathing on those babies, but you do pay, as their cost is in line with the Master Primes...

You do pay for what you get when it comes to asking NO BREATHING in Cine Glass, for Still Glass, well deal with it, as it isn't going away,
and for this I say don't use still glass in 3D, it will make your life existence MISERABLE!!

 

[Antoine Fabi]

Matthew,

That leads me to ask why do you think that some major directors and DOPs prefer UPs over MPs.

Antoine

 

[Ryan Patrick O'Hara]

Matthew,

That leads me to ask why do you think that some major directors and DOPs prefer UPs over MPs.

Antoine

I'm not Matthew, but can you give an example of who prefers UP's over MP's? I suppose from a non-subjective standpoint, with optical differences aside... they are smaller (half the length), lighter (half the weight), have a wider range of focal lengths (8-180 vs 12-150) and are less expensive.

If it's not a money issue, I've never heard anyone turn MP's down for UP's unless it was a hand-held or steadi-cam shot/shoot. Who knows. I have plenty of UP experience, but don't get the MP's too often, so I'm not knowledgeable enough to really know the difference the two give in look to guess the aesthetic preference.

 

[Matthew Duclos]

That leads me to ask why do you think that some major directors and DOPs prefer UPs over MPs.

That's an excellent question. My best guess is that Master Primes have become the golden standard when it comes to cinema optics. They are the go-to for accurate, high-quality images. In a industry driven be creativity and individuality, I assume it's considered hip to choose the less-than-perfect set of Ultra Primes over the uncompromising Master Primes while still proclaiming "I've arrived". Or they simply can't afford Master Primes.

 

[Antoine Fabi]

Hi Matthew,

Yeah...i mean, i think i saw Peter Jackson chose UPs in Lord of the rings.

Maybe i'm wrong but i think i saw that on BTS pictures.

Thanks!

Antoine

 

[Ryan Patrick O'Hara]

Hi Matthew,

Yeah...i mean, i think i saw Peter Jackson chose UPs in Lord of the rings.

Maybe i'm wrong but i think i saw that on BTS pictures.

Thanks!

Antoine

The LotR movies were released in 2001 through 2003... Master Primes were unveiled in 2005.

That is likely why they didn't use them

 

[Terry VerHaar]

Another Question for Matthew...

Another Question for Matthew...

Matthew - If you are willing and able to offer an opinion on this: do you think the Zeiss CP.2s are as good as, or better than, the cine-modded ZE/ZFs? I am asking this primarily from the perspective of the image, not the mechanical advantages or disadvantages of one vs. the other. Thanks.

 

[Illya Friedman]

Different manufacturers of cine lenses combat breathing in different ways. On the inside Master Primes could technically be considered a "zoom" lens, just one without a externally adjustable stage. The MP zoom is tied directly to the focus stage, and the amount of zoom is so minimal, it exists only to counteract FOV change from the breathing of the lens. It's the reason you can't see any breathing in a MP. It's a remarkable achievement and an incredibly clever design- you do pay dearly for it, in both rentals or purchases. I've never seen less breathing in any lens, particularly the longer MP focal lengths.

I'm fairly cognizant of lens breathing. I confess a little bit of breathing is no deal-breaker for me. That being said, ENG zoom lenses and older 16mm/35mm zoom lenses are notorious for breathing a lot (i.e. Angenieux 10-150, anyone!)

Zooms (as a general rule, but not always) breath more than primes. Longer focal lengths more than wider focal lengths. For those who are thinking of drafting older zoom lenses into service for narrative production, you may want to check how much those old zoom lenses breath in comparison to shooting with modern zoom/prime lenses. Comparing old to new, the older zoom lenses generally look particularly egregious, as far as breathing is concerned.

DP (and all-around cool guy) Randy Sellars shot a breathing test during the Portland Lens Test last week. None of the names of the lenses are listed yet. Lens names names go up in about a week. I'm intimately familiar with many of the sets that were part of the test, and if you watch "part 3" I'm certain you will see several sets of lenses under $50K with very little breathing http://vimeo.com/33690160

In the spirit of full disclosure, Hot Rod Cameras is an authorized dealer of Zeiss, Cooke and Schneider cine lenses, and I was also personally present at the Portland lens test. However, I have not yet watched the test results, nor did I have anything to do with the execution of the tests. I'm just as interested as everyone else to see the final version.

Maybe next fall we can do the Portland 2012 zoom lens test, that would be fun!

 

[Alexander Ibrahim]

I can tell you from my 15 year as a flame artist, trying to figuring out even the zoom factor from "focus pull breathing" using tracking dots of a moving camera shot might be doable, but it's freaking hard, involves the higher level of 3D tracking. Breathing is not only "zoom" it's a "fisheye" warp thing going on also so to create a perfect flat source with only focus moving involves warping swell, do not think you have it in resolve. But do not know for sure do not use the program.

No ... you can't correct geometric distortions in Resolve. Or at least I don't know how.

I know Nuke better than Flame, and neither particularly well. (I'm still a Shake guy essentially.) I think that Nuke has more tools for dealing with crap lenses and cameras than Flame though. That may influence how easy I think this is.

In any case I wasn't suggesting the level of correction you are talking about. Its another step way above and beyond what I think is Les was suggesting.

To be clear the context here is that Les wants to make software that take footage from cheap plastic lenses and automatically turn them into the equivalent of Master Primes.

I'm saying that isn't going to happen for a variety of reasons.

Having a perfectly flat source however is improving 3D tracking and 3D integration by a lot so getting the numbers out of the canon lenses and have it feed into the meta data is huge for post...

...

Well meta data lets you do that.

Reliable meta data would let you do that.

One of the issues I've been carrying on about is that the meta data from Canon and Nikon lenses is not accurate or reliable in my experience.

This would be far more workable a suggestion for LDS or /i lens data systems than it is for SLR lens data systems.

So i we are talking about using software to turn Ultra Primes into Master Primes, then yeah, this is a no brainer project.

We can probably turn cheap plastic lenses, into better SLR lenses, but I'd argue that Canon et. al are already building this into their cameras.

Maybe we can let red output the black and white mosaic that the camera is actually shooting then...LOL.
But personally I like to have at least some sort of representation of a half decent debeayered image on my touch screen... And I actually also like, for client purposes, to have the meta data grading in camera, to be able to see how the picture could look in post, having the lens data there aswell (which we will soon) would be great if it could be used but would not need to be burned in as an actual pixel change.

Wait ... That is exactly what Red does by recording RAW.

The various displays and such are just a representation of what the image looks like ... you aren't really bringing post into the camera. You are offering visualization tools in the camera though.

This is different than eliminating breathing and other lens defects in camera, which is actually destructive.

Doing this in post under an artists control is probably a good idea. I think its happening already in important regards.

The highly automated tools for "eliminating breathing" Les is talking about don't see to be the way to go to me.

Some more accessible tools intended for use on better SLR and low end cine glass ... that would be welcome.

Ultimately Les and I differ on the valid range of applications for such software.