I wonder what the colors would look like if he black levels matched.

Posted by IEBA  on  09/21  at  07:59 PM

---

Thanks, Art.  I could leave a stupid comment, but to be honest, the only reason I’m posting this is to be alerted when other people comment on this article.  I enjoy the discussion.   

I’m hoping at some point you’ll tell me the reason I’m not satisfied with caucasian flesh color rendition on the RED.

Posted by Graham Futerfas  on  09/21  at  09:19 PM

---

I’m reasonably happy with it under daylight. Not so much under tungsten. We know that there’s blue in the greens under tungsten light, and there’s a lot of green in flesh tone, so maybe that’s part of the problem.

They do seem to have reigned in the reds a bit, but people do tend to look a bit ruddy. There’s one theory that says that’s how they’ve eliminated the blue noise under tungsten—by warming everything up—but that seems too simplistic. I’m guessing they desaturate the blue channel at some point. The noise is still there, but at least it’s not blue. It’s actually somewhat pleasant in small amounts.

As far as matching black levels, they are matched. What’s misleading is that the EX3 chart had flesh tones on it whereas I covered them up on the RED charts, so I punched the flesh tones out of the EX3 chart in Final Cut Pro. Those holes are a black that can’t be achieved in reality, so please ignore them. Just look at the colors.

Posted by Art Adams  on  09/21  at  09:56 PM

---

In addition to blue in the greens under tungsten, I’m also seeing red in the blues (lowest right hand color chip), and poor yellow reproduction (far upper left chip). Greens also seem more pastel.

What might be useful here in addition to the visual is a triple vectorscope display, overlaid in three colors so the variations are more easily quantified.

And Graham, I have a hard time liking Caucasian fleshtones on Red as well. People tend to look sallow, unwell, on footage I’ve seen.

Posted by Bob Kertesz  on  09/21  at  10:40 PM

---

Triple vectorscope display, eh? I can do that. Stay tuned…

Posted by Art Adams  on  09/21  at  11:08 PM

---

At first glance I would say the EX-3 looks more vibrant, knowing full well that CC can always be done.

Posted by Tom Daigon  on  09/22  at  10:44 AM

---

Colorists tell me it’s hard to get pure colors out of the RED, and now I can see why. BUT—keep in mind that very little of what’s shot these days involves rich, saturated colors. The cultural trend is towards desaturation, which works well for the RED.

On its own it looks pretty nice. It’s only when you compare it to something else that you start noticing that the color isn’t the cleanest thing in the world. I suspect RED is learning a lot from this experience and putting that knowledge toward picking a new sensor that will clean up some of these issues.

Then again… it’s a $17,500 camera so there’s a limit to how good a sensor you can put in it. The F35, for example, has an amazing sensor… for a $250,000 camera! The RED’s biggest strength is its affordability, and a 4k camera that’s this cheap has to cut corners.

Let’s hope that as technology gets less expensive there’ll be fewer corners that need cutting.

Posted by Art Adams  on  09/22  at  12:27 PM

---

makes me happy to have ex3’s

Posted by .(JavaScript must be enabled to view this email address)  on  10/04  at  05:05 PM

---

Makes me happy that I just bought an EX3.

Posted by Tom Daigon  on  10/04  at  06:06 PM

---

The thing here is we’re comparing a Bayer filtered single imager camera to a 3-chip camera with a dichroic prism. The latter is always going to produce truer and more vivid colors.

I’ve seen a couple of movies produced on RED One and the above charts serve to reinforce my impressions that the camera was lacking in the green spectrum and has an overal reddish cast to the whole image. I also noticed in movies shot on RED (viewing on Blu-ray via 30-bit projector on 154” screen) that while edge detail appears sharp, there is a general lack of texture in the fabric of clothing on the actors, and in the materials of building exteriors, etc. I just don’t much care for the look—it’s not true to life.

Posted by Mark Weiss  on  10/28  at  12:41 AM

---

The “lacking in green” and “overall reddish cast” probably has to do with shooting under tungsten light. It’s a very different shooting experience under daylight, which is why to date I’ve only shot with the RED once under tungsten—and even then I partially corrected to daylight with an 80D filter.

The RED is an odd beast. Once in a while it seems very “video,” as in the case of a shot in a recent project where some sheer curtains appeared to clip even though they weren’t anywhere near clipping. But then there are times when a shot comes up on the Da Vinci and everyone in the room comments on how “filmic” it looks.

It’s its own look, and for the price it’s not a bad one.

I’m shooting with one this weekend. All HMI’s in front of green screen. The overall color grade will end up being a blueish wash—and overall washes seem to work particularly well on the RED. I think we’ll be happy with the results.

Posted by Art Adams  on  10/28  at  09:28 AM

---

A “blueish wash” works well with the Red (see an episode of “Leverage” on TNT). It tends to hide the fact that it seems to have some problems making realistic pink fleshtones (like every other Bayer chip based camera I’ve seen).

Without some kind of cold color wash, fleshtones on the Red look to me like the people have been filled with embalming fluid - sallow, cold, lifeless.

Posted by Bob Kertesz  on  10/28  at  09:43 AM

---

Not entirely the lighting is to blame. The scenes I was referring to were outdoor scenes in natural daylight. I’ll second the opinions that this makes the skin tones look rather poor. It’s a look, but it’s not the look that I personally prefer.

Posted by Mark Weiss  on  10/28  at  01:04 PM

---

Name:

Email:

Location:

URL:

Smileys

Remember my personal information

Notify me of follow-up comments?

Submit the word you see below: