How To Analyze Your Footage In REDCINE-X PRO

Intro: REDCINE-X PRO

Unknown

When it comes to being a RED owner, REDCINE-X PRO is very important tool. This software, specifically designed to be used with R3Ds, is an extremely powerful instrument that truly allows the user to fully analyze his or her RED footage in order to improve his or her skills for future shoots. In this article, I’m going to teach you how to analyze and what to look for when inspecting your R3D files. 

_________________________________________________________________

I: The Basics of RED Footage In General

A001_C014_10056F.0000350F

                                                     Source: Frytg.com

Digital cinema cameras such as RED shoot RAW and are encoded with metadata; some of the data is used for organizational purposes in post production, while other data can be changed; this data includes: ISO, color temperature, gamma space and color space, which can all be manipulated throughout the process of color correction and color grading. What’s important to note is that the user must know the primary ISO and proper color temperature for his or her RED camera in order properly expose the camera and fully utilize the sensor.

Let’s quickly review each sensor’s natural ISO and their native color temperature. The Mysterium is ISO 100, while the Mysterium X is ISO 320, and the Dragon is ISO 250. Knowing these sensor-exclusive ISOs will allow you to accurately analyze your and others’ footage. So, if you shot at ISO 800 on your shoot, and the footage looks noisy to you, drop the footage down to its base exposure, which is not 800, and you’ll see what you actually captured. You’ll be punching yourself for days, but it’ll be a worthwhile learning curve.

All cameras are all 5,000 kelvin-based sensors, which means they operate best under daylight sources. Using tungsten on your RED cameras will under expose the chip by 1/3 of a stop

_________________________________________________________________

II: Let’s Talk Log Curves

freq_hue_lin

I’ve talked about this in my previous article, but I’m going to reiterate, so you can fully understand. The user captures data in a Linear fashion, and the camera internalizes that data Logarithmically. Then, the data is put back into the computer as a linear image; the user then breaks it back into a Log image and finally releases it as a linear image. Confusing?

What that means is that when the data is captured Linearly, the shadows have a floor (underexposure), and the highlights have a ceiling (overexposure). Now, when the data is internalized logarithmically, all of the important detail in your image is pushed toward the middle of the curve in an upward fashion; the highlights are a little underexposed, and the shadows are pulled toward middle gray. The reason for this is because the camera is trying to capture and preserve more detail.

_________________________________________________________________

III: Every RED Has its Limitations – How To Not Expose the Camera

I remember my teacher talking about how he was going to use the RED for a concert shoot. I said to him, “You should use the Alexa or the C300 because the RED is NOT a lowlight camera.” He responds, “Yeah, the rental house told me the same thing, but it’s a RED; it can do anything.”

Speaking of concert footage, let’s take a look at RED’s test footage to see where the cinematographer excelled or failed in this example. Here’s a concert from RED’s archive in REDCINE-X PRO.

Screen Shot 2016-07-12 at 3.19.41 PM

Let’s analyze the footage in the photo above. It was shot at a kelvin of 5600 (you can see that in the right tab under Pick WB); that’s good because the color temperature of the sensors is 5000 kelvin.  The crew shot with a 16mm lens at f2.8 at “ISO 800.”  the RGB histogram on the top left is slammed to the left, and the RGB parade on the top, middle right displays this as well. On the left and right of the RGB histogram, you can see the goalposts. The right is over exposure, and the left is underexposure. The left should NEVER be filled to the top or even halfway; heck, its bad to even fill it up 10 percent. It should never be full. The same goes for the highlights.  In the RGB histogram, the red and green photosites are starved for light; the green channel is two-thirds of the photosites. The one channel (blue) that is receiving light is very hollowed out. The blue channel is reading the ring light on top of the stage. Those histograms should be filled with information, and they’re not. They’re crushed and spread thin to accommodate for the lack of light information.

Remember, the camera’s main ISO is all that matters; it has only one ISO. If you expose for anything higher than what the camera’s ISO offers, you’ll be underexposing the chip. Here’s the same still with the metadata raised up to ISO 800, which is what the cameraman programed into the camera on the day.

Screen Shot 2016-07-12 at 3.20.27 PM

            ISO 800 (Image pushed up by 1 and 2/3 stops in post). 

Here’s what it looks like in its natural state/base ISO of 250.

Screen Shot 2016-07-12 at 3.21.05 PM

ISO 250 (Underexposed by 1 and 2/3 stops on set). This is what the camera actually saw.

The image is optimal at ISO 640. So to actually reach correct exposure, the team would’ve needed lenses that could open up to f 0.5. That’s never going to happen.

Screen Shot 2016-07-12 at 5.11.48 PM

ISO 640 (Image pushed up in post by 1 and 1/3 stops)

So, what happens exactly if you underexpose your footage? Basically, you’ll get nasty image noise when you try to push the exposure of the overall image. The reason for this is because the sensor is replacing the lack of signal (light) with noise to fill the exposure gap.

_________________________________________________________________

IV: Exposing Properly 

Now, let’s take a look at a well-exposed image that was shot on the RED One Mysterium X sensor; it’s an Audi A8 commercial.

Screen Shot 2016-07-13 at 2.37.07 PM

This image is near perfect; it was captured at the MX sensor’s native ISO of 320. The only flaw I see is that it was shot with Tungsten lighting at 3200; the lighting choice could’ve been to emphasize the cool, LED front lights. You can see in the RGB Parade (on the top right) that the midtones in the blue are a little bit weaker than the red and green signals, which are both really strong. They should all be equal in measurement, which would mean all of the channels are being filled properly. There is blowout from the car’s LED headlights, but that’s to be expected from a high-intensity car light.

Screen Shot 2016-07-13 at 9.18.46 PM

ISO 320 (Near Perfect exposure on set). This is basically what the camera’s supposed to see. It would’ve been perfect if shot with a Daylight-based source.

Just for fun, let’s see what the image looks like when it’s underexposed by a full stop and 1/3 in post.

Screen Shot 2016-07-13 at 9.17.05 PM

 ISO 125 (Underexposed in post by 1 and 1/3 stops).

And, finally, overexposing the image in post by 1 and 1/3 stops.

Screen Shot 2016-07-13 at 9.17.54 PM

ISO 800 (Overexposed by 1 and 1/3 stops in post).

 

This is just one of the many uses of REDCINE-X Pro. In a later tutorial about the software, I’ll go over how REDCINE-X Pro works, and how you can use it to your advantage when it comes to your workflow.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s