Welcome to our community

Be a part of something great, join today!

Log in
  • 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 :)

Extreme Purple Fringing on Dragon

D

Dana Rice

Well-known member
Joined
May 12, 2011
Messages
170
Reaction score
0
Points
16
Location
Los Angeles, CA
Yesterday I helped a friend shoot a music video... I had my Epic MX and he had another guy there with a Scarlet Dragon. I took the footage home to transcode and upon looking at some of the Dragon footage I noticed an extreme amount of purple fringing that is not present in any of the MX footage. While this only occurs in a few of the shots, as a whole the MX footage seems to look better in just about every shot... cleaner, less artifacts, etc. I'm a little concerned because this is now the third time I have shot Dragon and MX side by side... the other 2 times were Epic Dragons, and all 3 times the MX has outperformed it. I have a Scarlet-W on preorder, cus I like the new body, but I'm fearful the image quality might end up being a big downgrade from my current MX.

Here is a frame from the Dragon

d2.jpg


Also, look at her hair... some other weird stuff going on there that never occurs on MX

Now is here is a frame from MX

mx3.jpg
 
I'm on Redcine build 40... I don't see a Draconcolor1 option... only Dragoncolor2, RedColor4, REC 709, REC 2020 and sRGB... I tried all 5 of those and it's still heavily present in all. Also I changed the MX footage to dragoncolor2... no fringing there regardless of setting.
 
Edit>Preferences>General

Check "Show Advanced Controls". Then restart REDCINE-X Pro.

DRAGONcolor2 is tuned a bit more blue than DRAGONcolor and will exaggerate some of the violet spectrum of chromatic aberrations created by some lenses.

Once you explore DRAGONcolor, make sure you are exporting "full debayer", not 1/2 or 1/4 if possible.

In the future there's a few things you can do to minimize this. Specifically stopping down the lens or avoiding clipping by extreme levels if possible. UV filters in some cases help in front of the lens. Just depends on the lenses you used.

If you have a second, please do let us know which glass you shot this with.
 
I was shooting the MX so I wasn't even aware of the issue till I took the footage home to transcode. I've never seen anything like it on MX and I've been shooting that almost daily for the last 5 years or so. In regards to lenses... I had a Sigma 18-35 on the MX... I believe the Dragon had the Canon EF 1.4 on.

Also what about the back of her hair, near the veil? Looks like some weird stuff going on there to.

I'm going to look into the Dragoncolor1 in a moment and will post the result once I do.
 
Dana Rice said:
Also what about the back of her hair, near the veil? Looks like some weird stuff going on there to.
Click to expand...

The Dragon sensor has more Total Captured Dynamic Range than Mysterium-X and due to this it can potentially "see more".

In terms of wrangling chromatic aberrations in high contrast shooting situations you ideally would stop the lens down 2-3 stops and you will see a noticeable improvement. If you are shooting at f/1.4 in this particular scene you will 100% have CA if that overexposed/hot window is behind a darker object.

DRAGONcolor will help minimize this, but the overexposure is the general spawning point of this chaos.
 
So I applied Dragoncolor1... it certainly helped, but the fringing is still heavily present. The one thing I'm confused about is that the Dragon is supposed to be superior to MX... especially in the highlights. In this particular case, that Dragon had the Skin Tone OLPF... we were both shooting at ISO 800... he was at a 1.4... I was a 1.8... however looking at the metadata... we were both shooting 60FPS, yet he was at a 180 degree shutter while I was shooting 360, so we essentially had about the same exposure. Yet the MX footage looks perfectly fine IMO, while even aside from the fringing, CA, etc. the Dragon footage looks much worse overall.

Also, you state the solution as stopping down 2-3 stops... while I don't doubt this would work to eliminate the fringing... in looking at the footage, it doesn't seem that would've been possible without dropping him down almost into a silhouette. I'm not 100% but I think going to 400 ISO is 1 stop down, at that point there's still a decent exposure and that would've been doable on the lens, but at 2 stops (100 ISO) he's almost a silhouette. Obviously I know doing this with the ISO isn't going to get rid of the fringing, but I just wanted to see at what point there was a usable exposure on his face and it seems 1 stop down was about the limit. If the fringing still would've occured at 1 stop, how do you get a shot like this on Dragon without doing something like bringing in a big HMI or NDing the windows or something to that effect?

Again, shots like this have never been a problem on MX, but it sounds like with Dragon the solution is to expose for the window and let your subject be a silhouette or involve expensive and/or time consuming lighting methods. Which in a perfect world might be the way to go, but unfortunately, many shoots don't have the time or budget for that.

Here is the shot with DC1 applied... better but still really bad

d3.jpg


Also look at this shot of him leaving the church

d4.jpg


Here it is on MX

mx4.jpg
 
Dana Rice said:
Also, you state the solution as stopping down 2-3 stops... while I don't doubt this would work to eliminate the fringing.
Click to expand...

Next time you have both cameras, setup both of them and shoot something similar with that Canon f/1.4 lens. You will see similar. The Sigma lens is a wildly different optical design with different coatings as well.

In terms of how lenses work, typically stopping down between 2-3 stops helps eliminate the inherit artifacts that comes with shooting wide open. I understand you are working with ambient lighting and that is indeed where the difficulty is in balancing your exposure.

Dragon and Mysterium-X are different and as I mention Dragon sees more, this includes more visibility in highlights and shadows.

A well contained exposure will remedy what you are experiencing. If this was a shoot I was on and couldn't bring lighting into the church, I would have simply added ND to the windows in the BG via stands or gaff.
 
Wow This is bad... Looks like my old FS700 :/ i thought cameras had evolve on this. I remember that recording with Odyssey 7Q The problem was fixed. I even dont get This on my a7rii.
 
This looks almost like blooming on old CCD sensors. Stopping down would likely solve the problem. but could this be reduced with filtering ? Read somewhere that a uv filter could reduce the purple. Never tried it though.
 
Was this fringing showing up as the shoot was ongoing - or is it showing up now?
 
What lens? That's the source of the problem. Good lenses won't do this
 
The problem is created by the lens. At least it plays the biggest part here.

And pointing it wide open into the direction of a bright source of light, with the contrasty subject right before that, won't help to reduce it. ;) It's where CA "shines". The thing in her hair could be some flaring. As Phil pointed out, there are some differences in sensors and color sciences, but basically you see the difference of those lenses.

Your SIGMA is great, it really is a bomb of a lens, not only because of sharpness. I highly suggest you to do some tests with that lens on MX and Dragon. Because seriously, i don't understand why your conclusion is that Dragon is worse, when both cameras have been used with a) totally different glass and b) different shots/settings.

You said something about a Canon f1.4 on the other cam, is it the 50? Well, that one isn't made to deliver best results wide open. I also didn't like the 35mm 1.4L, especially at 1.4 where it showed lots of CA. 24 1.4L was ok, but not perfect either.
 
This looks like the lens. You can't compare how two sensors look when you're using two totally different lenses. Stop down and you'll lose the CA.
 
It's not really fair to assume that because the mx sensor has little CA that the Dragon sensor is totally at fault too. Lens' have very different characteristics based on glass quality, coatings, etc. Also, you're wide open on the Canon lens, but what stop was the sigma landing on. As Phil says, the only way to have a true comparison is to set up an apples to apples test. Would love to see results if/when you've got some!
 
I have never seen anything like that. Ever. I don't see how that's not a problem with that particular camera... or that particular lens.
 
Peter Moretti said:
I have never seen anything like that. Ever. I don't see how that's not a problem with that particular camera... or that particular lens.
Click to expand...

Yeah that's one of the weird things about this issue. Dragon is supposed to offer more flexibility and image quality than anything else--but supposedly because it SO sensitive or has so much DR or whatever, it's ultra sensitive to CA, and so ironically you wind up having to work around things like this that you wouldn't have to work around on most other cameras (that are far less sensitive to this issue). In my mind, all the extra flexibility goes out the window because you're limited to a certain exposure ratio between objects in a frame--if you have something too bright on something too dark, you'll get purple fringing, in most cases where other cameras won't exhibit it.

So weird.
 
I've never seen it that bad either, and I shoot a lot
 
You must log in or register to reply here.
Back
Top