Post Production Studying

CB

-Contrast (luma) then Balance (chroma)!

-Contrast is adjusted first because it has a profound impact on balance rather than the other way around

-Ask: how does this relate or connect to CB? when we talk about and read and learn stuff in class

-Mess with luma and then chroma!

-Luminance affects chrominance but chrominance don’t really affect luminance

-Start with luma then go into chroma — you can go back and forth a little eventually but you need to start the process like this

Contrast

-Disparity between the darkest and lightest points

-Contrast increases/decreases the saturation of colors when it is adjusted

Luma/Luminance

-Contrast! luma is contrast. Remember CB — mess with luma then chroma

-It is the tonal range of bright and dark values

-Black is defined as 0% and white is 100%

-Called the “Y” or “L” channel — interchangeable

-Bigger difference or disparity between the light and dark of the frame means higher contrast and vice versa

-Easy to see when you put the image in black and white

-There are different ways to build luminance

-It is a fundamental building block

-Think YRGB parade in resolve, it adjusts luminance or contrast with the Y section because Y stands for the luma channel

Tone/Value

-Refers exclusively to contrast/luminance ranges so don’t say “earth tones” because it isn’t describing the color it’s describing contrast ranges

Chroma/Chrominance

-Color! The balance part of CB! luma then chroma!

-3 channels for chroma: RGB—red green and blue channels

-Color is defined as RGB combined which makes what we’re looking at

-Opposing colors are red and cyan, green and magenta, and blue and yellow

-All of the colors together make white

-Chroma is color but more specifically it is the RGB channels

-Think YRGB parade in resolve, it adjusts the RGB channels

-Hue and saturation are how we articulate chrominance

Opposing Colors

-Red and cyan

-Green and magenta

-Blue and yellow

Waveform/Scopes

-The overarching terms for the ways in which we measure color correcting properties and aspects

-An exact measurement, not just guessing by eye

-How we measure luminance, Y, chrominance, and RGB channels — these four things

-EX: YRGB parade, vectorscope, histogram, etc. — these are like units of measurements so like using both a ruler and a tape measure because they both measure/are useful for different things

YRGB Parade

-Type of waveform/scope, the main one we’ll be using this semester like 80% of the time

-Measures luminance, Y, chrominance, and RGB channels

-Think of Looney Tunes Wile E. Coyote who gets smushed flat like a pancake, parades show us a flattened version of the image

-If colors are out of alignment, it means there’s a color cast going towards that direction (EX: if blue level is higher, the image has a blue color cast)

-If RGB are lined up in evenly in any luminance (contrast) range then it is colorless and neutral! This is how you balance an image

-The dots on it are called the trace, “where trace sits on the parade”

-Each pixel in the frame is represented by one tiny dot on the parade — the y or up and down shows the intensity from top to bottom and the x or left and right shows the horizontal placement of picture elements with no regard to vertical placement

-Shows value from 0-100 of every single pixel

-In the EX of bridge frame from class: spikes at the top that are hard to see are like lights on a bridge or the brightest parts,t he weird stuff at the bottom are the darker parts, the big curve is the sky — it’s up and down!! So remember that!

-YRGB is your friend so use it like 80% of the time but then go into vectorscope and see if there are any issues with drift like skewing magenta (pixels going in one direction) – two different things that these tools do

-Parade monitors luma and how chroma interacts with specific luminance ranges, vectorscope is exclusively a chroma tool

-Measures luminance in red green and blue channels (chromanents) but it’s hard to measure specific shades of colors — you can’t do this based on the parade so we need vectorscope for this

SMPTE Bars

-Academic way to read scopes

-0% means black and 100% means white

Curve

-A concept — there are different types of curves like a log e curve

-Composed of shoulder at top, toe at bottom, and linear center that the angle of that is called the gamma

F-Stops

-A unit of measurement that measures how much light is coming in — it is a way to concretely quantify amounts of light that are being let in

-Each stop is double the brightness

-Each chip in a parade is like one f-stop of light

Log E Curve/S Curve

-Type of curve that stands for Logarithmic Exposure Curve

-This curve is in the shape of an S, also called an S Curve

-The human brain applies a contrast curve to the image we’re seeing with our eyes in real life — this curve mimics what the human eye sees

-This curve is what makes images look good, without the curve the image looks funky and flat

-3 parts of an S Curve: the shoulder, toe, and linear center

-If the curve isn’t a curve but is a flat line, it’s in log so if there’s no S Curve

-Straight line is more of a harsh or snappy or crunchy toe and one that goes out further is softer

-Crunchier toe makes it look sharper and in focus and more contrast

Shoulder

-One of the 3 parts of an S Curve

-The shoulder is the bending part at the top of the curve — it represents what happens as you go toward 100%

-Shows how luminance compresses and goes up toward the peak value

-Maximum tonality allowed is 100% white

-Shoulder is how the linear center ramps up and compresses to go to white

-Called knee in other fields

-This is the inverse of the toe

-Shoulder is as we go from the end of the linear center and we start approaching 100% it starts to bend and compress – same with toe but other way

Toe

-One of the 3 parts of an S Curve

-The toe is the bending part at the bottom of the curve — it represents what happens as you go toward 0%

-Shows what happens as you approach 0% which is black or lowest value allowable

-Toe is how the linear center goes down and compresses to black

-Called pedestal in other fields

-This is the inverse of the shoulder

-Straight line is more of a harsh or snappy or crunchy toe and one that goes out further is softer

-Crunchier toe makes it look sharper and in focus and more contrast

Linear Center

-One of the 3 parts of an S Curve

-Also called Gamma!

-This is the area that you are able to perceive most clearly

Gamma

-The linear center on an S Curve

-This is the area that you are able to perceive most clearly

-Angle of the linear center

-A relative concept, some photographers don’t like it because of this like Ansel Adam

More is Less Philosophy

-Don’t do a little bit of everything — it is a big giveaway that you’re not paying attention if you adjust every part/aspect in Revolve when you’re showing your work to the class

-Sugar cookie EX: like you don’t need to add cinnamon, nutmeg, cumin, mustard seed, unnecessary spices to a sugar cookie because it will taste bad and be doing too much

-Need a plan of attack — need to know what you’re going to adjust rather than just moving stuff around

-You can do so much with just primary corrections, you don’t always have to be doing too much and keying and whatevering targeted adjustments

-Do stuff in passes, do primary and balancing and getting things in the ballpark first

Unity/Disunity

-When RGB are aligned and with the white, it means there is no color

-When RGB aren’t aligned with the white, it means there is color

-If we don’t want color pollution, we’d put all of the colors in unity in that specific luminance (contrast) range

-If we do want color pollution like in the image of the man in class we’d bring the blue channel up and take them out of unity

LOG (color space)

-Log is a color space

-It is how we get big amounts of information into a small file — it compresses the information

-It is a flat file

-We will be working with a lot of Log encoded images

-If it was filmed on filmed it can’t be log

-We’re working in a color space when we say we’re working in log – we’re compressing it into one space

-Use LUT to put it back out

-Log isn’t raw and raw footage isn’t log

-We work with log in exercises – “looks flat”

RAW (format)

-Log is a format

-All of the in-camera decisions are meta-data

-Can change the decisions that the DP made when filming which can be disappointing for them but also helpful sometimes

-Huge file sizes, it can go wrong — it is not compressed like how Log is compressed

LUT

-A lut is like a template that you can put on an image so it adjusts the aspects of color correcting for you, like a filter

-LUT stands for look up table

-It is a series of Log E Curves that affect luminance (contrast) and chrominance (color)

-The LUTs we are using are ones that were designed by the camera manufacturer – this is how they were monitoring the image on set, start from the set mindset and build outwards

Vector

-Vector measures just chrominance or color

Zones

-Zones are stops of exposure — each zone is at 10% intervals in terms of YRGB parades

-Textural range allows us to define the relation from one area to another, this is clearly visible in the image

-People color on thermonuclear levels (when a mac is all the way bright), everything will be dark when you look at it on a monitor because the computer was at a crazy brightness level when they were editing

-Living human flesh is within a very small range of zones

Gamut

-The gamut is the range of displayable colors within the color space we’re working in

-We’re working in gamma 2.4 (defines luma) and rec 709 (defines chroma)

-It is the range of allowable values — the values relate to luma/chroma/tonality

-The allowable lum and chrom values – we’re working within this

-Working in a smaller gamut is okay bc the colors will reproduce when you look at a monitor in the same gamut

The Color Space We’re Working In

-We’re working in Gamma 2.4 (defines luma or contrast)

-We’re working in REC 709 (defines chroma or color)

Clipping

-Luminance (contrast) clips when a straight line forms at the top of a parade

-The luminance is going beyond the gamut or what’s allowable in the color space so it turns either white or black depending on which extreme it’s exceeding

-Clipping destroys the detail of that area too

-When something exceeds the maximum allowable in that one area

-Legal means it’s not at that point yet

-Can clip the blacks, whites, and chromanents

-Rule of thumb = clipping is bad don’t do it in this class but can be an option if intentional

-Blue channel clipping for EX is when u bring that channel all the way up and you lose detail like how you would for black and white — this is chroma channel clipping (be careful in underwater section of project)

Vectorscope

-YRGB is your friend so use it like 80% of the time but then go into vectorscope and see if there are any issues with drift like skewing magenta (pixels going in one direction) – two different things that these tools do

-Parade monitors luma and how chroma interacts with specific luminance ranges, vectorscope is exclusively a chroma tool

-Parade measures luminance in red green and blue channels (chromanents) but can’t measure specific shades of colors – so we need we need the vectorscope to measure this

-Similar to squishing into flat pancake like YRGB

-If the image is at black and white/unity, it’s just a dot in the center of the vectorscope

-Once a color gets on the frame, its distance from the center outward shows how saturated that specific color is

-R - red, m - magenta, b - blue, c - cyan, g - green, y - yellow

-Across from each other (opposing colors/complementary), red with cyan, magenta with green, blue with yellow

-Adding saturation makes the trace extend out – displaying the chrominance of all of the pictures that is defined as the vector of it – the hue

-The distance from the center defines the saturation

-Hue and saturation are how we articulate chrominance

-Saturation is the distance from the center and hue is defined by the compass thing — for EX it’s going toward the red part in that image from class

-Trees actually skew quite yellow in real life and the light makes it look green, sky isn’t blue as it skews closer to cyan

-Luminance-wise they can’t be very bright bc they’ll lose saturation – like Dorothy’s shows in Wizard of Oz

-Don’t want to see the straight tall lines in our first projects — means clipping like the line goes the other direction it means don’t do more because you’re outside the gamut

-Colors might look fine but they could be looking wonky in the graphs so always check

-These are global corrections like primary it applies to the frame equally so we’ll get into secondary corrections later

Textural Range

-It’s okay to go out of textural range if it’s intentional

-Anything that is critical or clearly needs to seen needs to be in the textural range

-Clipping makes that area not have texture