Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

Green & magenta highlight fringe - why?

Karl Roeder

Well-known member
Joined
Dec 10, 2007
Messages
79
Reaction score
0
Points
6
Age
61
Location
Ottawa, Canada
I have attached two stills captured from RedCine. I am concerned with the colored edging along the highlighted woman's cheek. Image one shows a green fringe. Image two (one second later) shows a magenta fringe. Has anyone seen this before? All current release builds were used.

- Karl
 
Yes.
Looks like high contrast edge CA. What lens were you using and at what aperture?


Thanks Jeff... We used a Zeiss super speed 50mm at ƒ 2.0, clean no filters.
Why do you think it would shift from green to magenta?

- k
 
Yes, it is chromatic aberration from the Super Speed lens not the camera. I've used a few different sets and there was always CA present to a greater or lesser degree- more so the wider the aperture.
 
Yes, it is chromatic aberration from the Super Speed lens not the camera. I've used a few different sets and there was always CA present to a greater or lesser degree- more so the wider the aperture.


Thanks for the responses everyone...
Any suggestions on the best primes available to minimize this (at wide apertures)? Cookes?

- K
 
Any suggestions on the best primes available to minimize this (at wide apertures)? Cookes?

Yup. Cookes and Masterprimes.

The CA fringes are green when they are defocused in one direction and magenta when defocused the other way.

Photoshop's RAW importer does a surprisingly good job fixing this on stills. Anyone know of a tool to do the same for video?
 
Yup. Cookes and Masterprimes.
The CA fringes are green when they are defocused in one direction and magenta when defocused the other way.

So this explains why the fringing changed colour within the shot! As the actress moved about, the focus on her was inaccurate and the CA changed! Thank you so much for this explanation.. I have learned alot here (namely, to budget for the best lenses, especially for matte work).

- k
 
Anyone know of a specific test for chromatic aberration?
Anyone know how the Red zooms rate?
- k
 
Anyone know of a specific test for chromatic aberration?
Anyone know how the Red zooms rate?
- k

Aim the camera at a high-contrast edge, such as a window frame with a bright exterior and focus in and out. Build 17's 1:1 zoom mode will make this easier.

Note that in my experience some of the CA you see in the preview/LCD goes away in Redcine. I think they are doing some CA reduction automatically. For example we found the (original) Red 300mm showed a lot of fringing in camera, but was completely clean in Redcine.
 
Yup. Cookes and Masterprimes.

SuperSpeeds are not much worse than above mentioned lenses, at least I can speak for S4s and SuperSpeeds Mk2/3s.

IMO, all cine glass I worked with, Zeiss, Angenieux, Cooke are not badly prone to CA. Infact CA is a minor problem. There are many other aspects apart form focal lenghts which makes one lens more desirable over another. Such were: Barrel distortion, foucs throw, weight, size, speed, contrast, etc..

If you really want heavy CA use 2/3 B4 zooms.


Hans
 
Photoshop's RAW importer does a surprisingly good job fixing this on stills. Anyone know of a tool to do the same for video?

AE + some creative expression writing should do it... Although you may end up with edge strobing or jitter. Haven't tried it. Could probably clean it up in some of the better grading apps with good edge tracking tools.
 
Note that in my experience some of the CA you see in the preview/LCD goes away in Redcine. I think they are doing some CA reduction automatically. For example we found the (original) Red 300mm showed a lot of fringing in camera, but was completely clean in Redcine.

Craig, I'm not dead sure but I guess it has to do with the in camera debayering and the highlights handling of RedSpace and Rec709. Do you have that with RAWview? Surely Red is not WYSIWYG.

Anyway, I totally agree that for CA evaluation you have to develope the footage in RC, best would be 4K and RedLog. Many CA is gone then.

Hans
 
Anyone know of a specific test for chromatic aberration?

Projecting lenses shows CA pretty clearly. The Cooke S4s are not with out their flaws. I projected s4s next to Ekran speeds, and the Ekrans had less CA than the Cookes on some focal lengths.

Next time you are at a rental house, ask them to project some lenses for you. Even different lenses by the same manufacturer will have different traits.

Cheers,

Nick
 
Anyone know of a specific test for chromatic aberration?
Anyone know how the Red zooms rate?
- k

A really good test for CA is to shoot a back-lit chart or detailed dark object with sharp edges in front of a bright light. Don't blow out the back-lighting. You can then shift focus on the chart or object to see the CA and how it behaves with the edges in sharp focus or slightly out in either direction. CA is typically more apparent at wider apertures.

The RED zooms are actually pretty good for CA. Very minimal on the RED zooms. I haven't worked with the 50-150 enough to know just how much it produces, but it seems very minimal. The 18-50 produces almost no CA in most situations. People like to talk down on the 18-50 and I used to be one who complained about its mechanics. However, after I picked up a used one, I find that I use it all the time. Great lens, excellent size / weight, close-focus, sharp picture, just all around good lens for the money. Mechanics are a bit gritty, but not usually any more so than most rental lenses that have been around the block a dozen times. My biggest complaint with the 18-50 is the breathing. It breathes like snoring dragon.

Here's some backlit chart images - shot 4K 16:9, 100% crops of lower left corner.

Here's a backlit chart image from the RED 18-50 @ T3 18mm
535_1233076988.jpg


Here's the same from a Cooke 18-100 @ T3 18mm
535_1233077164.jpg
 
Back
Top