Dragon Non-Linear Color Response?

[Gavin Greenwalt]

So the motive here is what... strictly punative, or are we really interested in good color? Are some of us looking for capitulation or color science?

I think we're all looking for a solution. But I'm also saying that if you want definitive answers about shortcomings--it's not going to come from the biased self-interested business.

 

[Scott Crawley]

I think we're all looking for a solution. But I'm also saying that if you want definitive answers about shortcomings--it's not going to come from the biased self-interested business.

True.

For his part Graeme just offers solutions and only when he has them. No lip service. :smiley:

 

[Marc Wielage]

But now I am wondering... does the RedMote weigh the same as a duck?

Well, it floats in water.

 

[Scott Crawley]

Well, it floats in water.

OK, that's a ROFL!

Did you intentionally put one there?

 

[Eric Haase]

OK, that's a ROFL!

Did you intentionally put one there?

I can say that I was very close to throwing my Redmote in the ocean out of frustration while shooting on the beach. I can't say if it floats or not. I DO know it stays in a little case inside another case and never comes out though.

 

[Marc Wielage]

OK, that's a ROFL! Did you intentionally put one there?

No, but it was too good a one-liner not to use. (I worked in sitcoms for 25 years, so it's a habit.)

I've been on sets where the Redmotes were perfect and flawless in every way, and I've worked on sets where the camera crew cursed and fumed and went nuts for 10 hours. I chalk it up to WiFi weirdness.

Wearing my colorist hat, I've never seen color from Red or Arri or Sony or Canon look bad except when I was dealing with 8-bit material. With 10-bit material, as far as I'm concerned, the faces were generally fine, and I can't recall ever seeing this "blotchiness" problem that people are talking about. I don't dispute that it could be there under certain makeup and lighting conditions, but I bet this would be relatively easy to see in testing.

I've definitely seen all kinds of nastiness when people grab secondaries and push them too hard, particularly when the key isn't qualified well or if the actor is under weird lighting conditions. But I also saw that in the film days, too. Film didn't deliver 100% perfect linear color response, either.

 

[Matt Ryan]

Just watched this video again and noticed the green splotches. Could have been amplified by the color. Shot on Epic X.

 

[Paris Remillard]

I'm glad that this issue is finally being openly discussed. I feel like I've been ignored or shut down every time I've tried to mention it over the years. To me it's not a question of why it happens or whether it happens. We all seem to know pretty well the conditions under which it rears it's head. And the specific color range where it falls.

The question is, why does it happen so much more frequently and prominently on Red?

Is the remedy for it something that can be built into the raw development? I know that Graeme mentioned something about that earlier in this thread.

And I of course know that using different make up and different lights are possible options. Though not always or not entirely or not always entirely. But obviously other camera and software manufacturers have found some way to alleviate or fully eliminate the problem in their development and processing. Like Eric, I've had side by sides of raw stills from my 5D and frames from my Epic and the 5D stills were free of the color casts. The Epic full of them.

I mentioned this link, http://fstoppers.com/hasselblad-vs-red , a couple of years ago when it was posted but no one seemed to agree with my assessment that the Hasselblad still had nice even skin tones, while the Epic still is green on one half of the face and magenta on the other with not so pleasant transitions between the two. Same lighting, same make up. So obviously Hasselblad or Capture One or Lightroom have figured out in their debayers how to correct for it.

Perhaps Red should open up their debayer to third parties who have successfully dealt with this and let them have a go.

And I've read all of these post and know that it does show up at times on other cameras, so I really don't want to rehash any of that. I just don't want it to be an issue on the Red when it's not an issue on another camera in the same situation. Which has happened frequently in my experience. In fact I'm dealing with it on two recent jobs as I'm typing this post.

 

[Ivan Kovax]

I'm glad that this issue is finally being openly discussed. I feel like I've been ignored or shut down every time I've tried to mention it over the years. To me it's not a question of why it happens or whether it happens. We all seem to know pretty well the conditions under which it rears it's head. And the specific color range where it falls.

The question is, why does it happen so much more frequently and prominently on Red?

Is the remedy for it something that can be built into the raw development? I know that Graeme mentioned something about that earlier in this thread.

And I of course know that using different make up and different lights are possible options. Though not always or not entirely or not always entirely. But obviously other camera and software manufacturers have found some way to alleviate or fully eliminate the problem in their development and processing. Like Eric, I've had side by sides of raw stills from my 5D and frames from my Epic and the 5D stills were free of the color casts. The Epic full of them.

I mentioned this link, http://fstoppers.com/hasselblad-vs-red , a couple of years ago when it was posted but no one seemed to agree with my assessment that the Hasselblad still had nice even skin tones, while the Epic still is green on one half of the face and magenta on the other with not so pleasant transitions between the two. Same lighting, same make up. So obviously Hasselblad or Capture One or Lightroom have figured out in their debayers how to correct for it.

Perhaps Red should open up their debayer to third parties who have successfully dealt with this and let them have a go.

And I've read all of these post and know that it does show up at times on other cameras, so I really don't want to rehash any of that. I just don't want it to be an issue on the Red when it's not an issue on another camera in the same situation. Which has happened frequently in my experience. In fact I'm dealing with it on two recent jobs as I'm typing this post.

I don't think people ignored that. The original MX sensor had magenta green shift issues. It seemed like the matrix was skewed. Dragon with new OLPF I haven't noticed any issues as of yet.

 

[Ivan McCullough]

I don't think people ignored that. The original MX sensor had magenta green shift issues. It seemed like the matrix was skewed. Dragon with new OLPF I haven't noticed any issues as of yet.

umm, the example that this thread exists for is not one?

 

[Ivan McCullough]

I'm glad that this issue is finally being openly discussed.....
And I've read all of these post and know that it does show up at times on other cameras, so I really don't want to rehash any of that. I just don't want it to be an issue on the Red when it's not an issue on another camera in the same situation. Which has happened frequently in my experience. In fact I'm dealing with it on two recent jobs as I'm typing this post.

I agree, no point in just ignoring it and hoping it will go away.

And a few people on this thread did try to just dismiss it as a RED based issue.

I do know that Graeme is aware of it and knows what those who have seen it are seeing.

Its just years later and into new generation sensor tech I have doubts about whether it can be solved using RED methods.

 

[Scott Crawley]

Has the OP image been processed in the new color science yet? I would be interested to see what it looks like otherwise unmodified. Maybe I will do that for giggles.

 

[Paris Remillard]

Its just years later and into new generation sensor tech I have doubts about whether it can be solved using RED methods.

Exactly...