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 :)

Purple Fringing WORSE on Dragon - PHOTOS

K

Kenneth C Merrill

Well-known member
Joined
Aug 9, 2015
Messages
506
Reaction score
0
Points
0
Location
Los Angeles
I was able to test the Raven against my Scarlet-MX this afternoon, and I hope to post the results later this week, but in the meantime here is something that has bothered me since I saw it first, and the response to the issue has bothered me even more. Every time someone brings up the issue of purple fringing on the Dragon sensor, they are shouted down by people saying it's merely a lens characteristic, a form of CA, it can happen to any camera, and it's not the fault of the Dragon sensor itself. I will admit that all of those things are true, but because purple fringing (at least the kind we see here) is a digital artifact, it makes sense that different sensor designs would be more or less prone to it. I think the Dragon sensor is much more prone to purple fringing than most cinema cameras and unfortunately even the MX sensor.

Here is a frame from the Raven this afternoon. Sigma 35mm ART, f1.4 (the fringing began to show at f2.5)



And here is the same frame from the Scarlet MX. Same lens/settings




Admittedly, I was able to produce purple fringing on the MX by pushing the focus just a few inches away from the subject, but it is very slight, and I was barely able to make it happen (as opposed to the Raven, which I didn't have to try very hard...)




I guess the point of all this is I wished Dragon users would recognize that this sensor is more prone to it; it's a problem that most people will want/need to avoid; and provide thoughtful solutions, rather than trying to pass the problem off to the user's lenses. Thoughts?
 
PatrickFaith said:
Red said their will be a fix in the next patch. I'd submit these r3d just to help them on their testing.

http://www.reduser.net/forum/showthread.php?145907-Raven-black-hole-sun-issue&p=1637728#post1637728
Click to expand...

From what I've seen, this is probably a different issue. The Raven is the first Dragon Sensor camera to show this purple/black hole son artifact, but it is not the first Dragon sensor to exhibit purple fringing--that suggests they are different things.

I could be completely wrong though! I would love to hear more expert opinions on this.
 
On a shoot all day today, but I will try this when I get home. If the fix is really that simple, then I stand sheepishly corrected!
 
Jacobo Martinez said:
Graeme mentioned that in DC2 chromatic aberration is out of color gamut, so it produces that over saturated purple . So if you want to avoid it, go with DC1.
Click to expand...

This info should pop up anytime DC2 is selected in camera. How many lenses have been bashed and ridiculed for this dragon-enhanced CA?
 
Kenneth, stoked that you're bringing this up. If people are to intimidated to be critical then we're in a sad place. Good you're speaking up. That way Red also has a chance to fix it. How would they otherwise ever become aware of these issues...
 
OK, so I played around with DragonColor (not 2). Here are the results:

Raven, DragonColor


Raven, DragonColor 2


Scarlet MX


DragonColor really does take the edge off the purple fringe, making it fairly tolerable. Obviously it's still worse than the MX. It does seem to be a color science issue, though, and not necessarily a sensor issue. Is it not possible to tweak the color science to reinterpret the extra-visible spectrum light or whatever is causing this? I feel like if more noise had been made about this (instead of pretending that every other sensor/color science behaves the exact same way), we might have a better solution by now. Thoughts?

For what it's worth, I really loved playing with the Raven. Even just looking at the frames above, besides the purple fringe, it clearly has the better image. I can honestly say I was floored by how well the camera performs. I'm excited to get my own; I just wish this purple fringe thing weren't so common; as I frequently find myself in high-contrast situations.
 
The Scarlett MX frame needs the DRX slider applied. That will eliminate the "purpleness" of the white. Anytime you blow out pixels this much you are going to have problems.

It's really best to understand how much of the image lies below clipping with RED. It's amazing how much you can put back when you protect the highlights. When you expose for shadows (which are easily pulled out) without clipping highlights you can end up with some pretty stunning stuff.

one thing that happens with RED (and I have been spoiled by this for being with them from the begining) is that it is much more forgiving that many of the systems out there. You still have to nail exposure, but from the start the hard clip of the top end (which doesn't exist nearly as much anymore) was always something to protect against knowing through experience that the processing of that image is just fine with a nice wide DR and all the things we love.

It's important to test and to not do too much by WYSIWYG because there is A LOT more in a RED negative than you might imagine.

When those white pixels are blown out, they really are gone, but if you protect the highlights a bit, then all the rest falls into place.

It's good that you try to break things, but all you rally have to do is see that the stoplights are lit up beside the histogram to know that they are broken.

David





David
 
You must log in or register to reply here.
Back
Top