ProVideo Coalition.com: Stunning Good Looks by Art Adams

Film speed is determined by complex math that, frankly, I have no understanding of, but it revolves around a point at the bottom of the gamma curve where the emulsion starts to respond in a predictable way.

How, then, do we apply that same kind of thinking to an HD camera, especially a “raw” one? After looking at the data I decided that I would base my estimated EI around the point where noise started to noticeably corrupt both black and moderate highlights. In the case of the highlights I watched the vectorscope for the first indication that the white chip on the chart, at 2 stops above 18% gray, started being distorted by noise. Similarly, I watched the parade waveform for the point where the black chip on the chart, at 5 stops under 18% gray, noticeably lost definition in one of the color channels (typically blue). Based on those two factors, I determined that EI 320 puts Zones 6 and 7, where flesh tones fall, just above the level where they will start to become noticeably noisy, as well as putting black just above the noise floor. Brighter exposures show less noise but white and black retain consistent characteristics. Darker exposures show white and black devolving rapidly into noise.

This test was semi-scientific in that I did not shoot the DSC Labs chart full frame, but zoomed into it later—exaggerating the noise. This turned out to be a bit of a blessing, however, because when the RED becomes noisy it does so very quickly, and blowing up the chart revealed that threshold quite clearly.

The full frame was intended to show the effect of exposure and light color on flesh tone.

The RED reminds me of stories I heard about older film stocks whose shadows would turn green or red and be muddy unless some small amount of exposure was added to the blacks. It was common for a cinematographer to aim a small light into the shadows so that the toe of the emulsion received a little bit of light, which improved the quality of the blacks enormously. The RED strikes me as being similar, not that the shadows change color dramatically, but that it doesn’t yield a solid technical black at 0 IRE, always requiring some crushing of the blacks in post. It occurs to me that a little extra fill light might clean up some of the noise that otherwise haunts the shadows. Director of photography Geoff Boyle described the Vision Research Phantom similarly: he always light with a little extra fill light and crushes the blacks in post to make them richer and noise-free. I see no reason why that shouldn’t work with the RED.

I’m disappointed by the apparent contamination of the green channel by blue under tungsten light, but that seems to be an insurmountable hardware issue. The Rec 709 colorimetry appears vastly improved over earlier software builds. The tendency for the RED to exaggerate green or magenta seems to be gone. Over time I’ve come to prefer the Camera RGB color space for its subtle colorimetry, but Rec 709 seems usable now. The camera still renders color a lot better under daylight than tungsten, at least as far as the blue channel is concerned, but the exposure latitude under both kinds of light appears to be the same.

Here are my recommendations for under- and overexposure limits when using the RED ONE, build 20, at EI 320:

Like any film stock, you should vary the EI of this camera to reflect your personal tastes. Changing the RED’s EI does one thing, and one thing only: it changes where neutral gray falls on the RED’s 9.5-stop exposure scale. For example,

Art Adams is a DP who tries not to create a lot of noise. His web site is at www.artadams.net.

The Best of Stunning Good Looks

Art Adams | 08/30

A directory of my best articles, sorted by topic.

This entry is a guide to my best articles, sorted by topic. Enjoy!

NAB 2012: Assorted Snapshots

Adam Wilt | 05/08

A few cool things I saw at the show that didn’t fit into any other articles.

NAB is too big a show in too short a time to see more than a fraction of it. I’ve covered a few things in some depth (as have other PVC folks), but there’s plenty more that slips by without proper coverage. Here, I have a few photos…

NAB 2012: RED

Adam Wilt | 05/07

RED’s Ted Schilowitz discusses 2012’s products, and a photo gallery.

RED’s “Leader of the Rebellion” Ted Schilowitz held a press conference at NAB on Monday, describing the projects and products RED is working on. Rather than paraphrase him, I’ve got him on card (well, it’s not “on…

You must be registered to comment. This is an effort to reduce spam. Please REGISTER HERE.

Thanks, Art—really useful test!

How do you feel about setting exposure based on your spot meter at 320? Looking at your waveform captures at normal exposure, you hit about 40-45 IRE under daylight at 40IRE under tungsten. Could you open up 1/2 stop and bring grey closer to 50?

I ask because there’s been a lot of discussion in the past about where to rate the sensitivity of the RED, and even when I’ve used it and compared my light meters to the histograms on the camera, I’ve found it to be about 200 instead of 320, even when the camera’s set to 320.

Obviously, by setting 18% higher on the scale, you’ll get less over-exposure latitude and more under-exposure, but your mids might be more accurate.

What are your thoughts on this? I know there’s some debate as to where 18% grey should land on a waveform.

Great work writing this up—you make it look so easy!

Posted by Graham Futerfas on 08/15 at 04:43 PM

Great test Art! Been wanting to do one of these of my own since build 20 launched. I have noticed on quite a few non-scientific tests on my own camera the significance of the new build. I think color/kelvin is more accurate and the noise floor under tungsten light has had a significant increase in certain but not all footage. Its finally closer to a 9-10 stop DNR camera IMO as your tests show which RED and others boasted in the past but did not always prove true in real life shooting scenarios.
I feel most comfortable shooting between 250 and 320 ISO. That said, we rated the last feature I was on at 200 ISO. We were in a controlled interior environment and wanted to gain shadow latitude as we could control our highlights and the footage looks marvelous!
I have been experimenting in my own time shooting at 500-640 ISO on day exteriors to achieve greater overexposure latitude. If you expose correctly I have found the noise to be quite impressive in comparison to 320 ASA. Of course at the loss of some shadow latitude. But hey, first thing everyone does is crush the blacks in post anyhow Anyway, great stuff and thanks for the test results!

Posted by mikeburton on 08/15 at 05:06 PM

You’re most welcome, Mike. I agree with you that in most cases you’d want to cheat the ISO lower to reduce noise in the shadows, but 320 seems to be the point beyond which flesh tones and shadows get noticeably noisier, and very quickly, too.

I don’t like to play things so close to the edge, and in the past I’ve gone overboard to EI 160. I think I could do EI 200 (or 250 in a pinch) and feel pretty good about it, even in tungsten light.

What I’m discovering is that there’s a difference between pretty pictures and accurate pictures. The RED does pretty pictures, but not accurate ones. Unless you’re shooting food or fashion that’s probably not an issue.

There do seem to be a rash of fashion still photographers who think they can shoot with the RED and then pull perfect stills from it for print work, but they are completely high.

Posted by Art Adams on 08/15 at 07:48 PM

Hi Graham-

If you talk to serious video engineers they’ll typically tell you that middle gray is actually around 42 IRE, not 50, so in this case the RED seems to fall in the right spot—although that’s probably somewhat arbitrary when dealing with the RED. For example, in this case, outputting RedSpace gamma put all the picture info between about 10 and 80 IRE. I’m not sure what the other gamma settings will do.

I think you could shoot at EI 320 if you had to, but that’s a little too close to the edge for me. I don’t know that I’ve ever rated a 500 ASA film stock faster than 400, so my inclination would be to overexpose and push the noise floor down. I’m in the 200-250 camp, and it took these tests to convince me that I can back off from 160.

This article was anything but easy. I’m amazed I had the patience to finish it. The payoff was learning some very interesting stuff that they don’t tell you in the brochure.

Posted by Art Adams on 08/15 at 07:56 PM

“What’s odd here is that the red waveform seems to show that red is being detected in the column of color on the left side of the chart, where yellow and yellow-green appear. Those shouldn’t have any red in them at all.”

Au contraire—in additive RGB color, yellow is a mix of red and green. The RED’s red waveform is doing just what it ought to.

Posted by Adam Wilt on 08/15 at 08:05 PM

Aha! I feel an article correction coming on…

Posted by Art Adams on 08/15 at 08:21 PM

One questions.
WHen you talk about your EI are you talking about your lightmeter or RED´s ISO setting? DO they match in your tests?

Posted by .(JavaScript must be enabled to view this email address) on 08/15 at 08:22 PM

In the past I’ve left the camera set at EI 320 and then set my meter at EI 160 just to make sure I had enough lighting for an exposure. I’d then set the exposure by looking at all the telemetry in the viewfinder: zebras, histogram, raw data saturation sensor (barber pole), stop lights, etc. After having done this test I’d be curious to see how well I did by setting my meter and the camera to 200 and see how well the two lined up.

Posted by Art Adams on 08/15 at 08:59 PM

In the tests the camera remained set at 320, and Adam changed the exposure in .5 stop increments. I then compensated for that exposure change in RedCine by entering the appropriate ISO number (or, in the case of RedCine’s third stop increments vs. our half stop increments, to the nearest setting that matched the previous shot) and exported a ProRes HQ file with no further tweaking.

Red waveform mistake removed, blue waveform anomaly remains.

Posted by Art Adams on 08/15 at 09:02 PM

In the past with film cameras one was able to adjust the sensivity of the ‘sensor’ but the properties of the optical viewfinder was fixed. Nowadays it the otherway around. The properties of the sensor are fixed but the electric viewfinder is adjustable. For this reason digital cameras should not be tested like film cameras and this article creates bit a feeling as if cars are tested by driving them backwards.

The ISO setting does not affect the sensor and the gamma curve chosen in post has an effect to the level of mid grays. Consequently, IMO talking about overexposure is not that meaningful. Instead, one should exposure to the right. Red One gives the user a rather accurate idea where the clipping points of the channels really lie. For this one should set the raw view on and follow the histograms, trafic lights, and zebras to see what’s really going on. That’s how the last bits can be exploited and the distance to noise is maxmimized. More precisely, that’s the whole point of the raw view. Why would anybody want to underexposure unless one is forced to do so in lack of light?

Posted by .(JavaScript must be enabled to view this email address) on 08/16 at 01:27 PM

Art, RED Camera aside, I understand that tungsten’s typically run closer to 2900k than advertised 3200k but why are the daylight Kino’s so warm? Is it lamp age?

Posted by .(JavaScript must be enabled to view this email address) on 08/19 at 10:00 AM

Hi Art - Did you use any filtration in your tests? I would be curious as to how an IR cut filter would change these results. I have yet to shoot RED without either an IR cut filter or Pola filter and have had excellent results in color reproduction (build 17 though). My one color reproduction test with RED and no filtration was quite disappointing - hence my tendency to filter.

Posted by Red Razor Films on 08/23 at 10:57 AM

Name:

Email:

Location:

URL:

Smileys

Remember my personal information

Notify me of follow-up comments?

Submit the word you see below:

RED Build 20 Torture Tests

Conclusions