Free Video Editing learning guide
Advanced color grading in DaVinci Resolve
Advanced color grading in DaVinci Resolve — a free advanced-level guide covering advanced color grading in davinci resolve. Learn with clear...
What you will learn
- Advanced Node Architecture & Workflow
- Deep Dive into Scopes & Waveform Interpretation
- Advanced Secondary Corrections & Power Windows
- Custom LUT Development & Integration
- HDR Grading & ACES Color Management
- Film Emulation & Look Development with ResolveFX
- Advanced Tracking & Stabilization for Color Workflows
- RAW & Log Footage Color Space Management
- Collaborative Grading & Project Sharing in Resolve Studio
- Delivery Optimization: Color Space Conversion & Export Settings
1. Advanced Node Architecture & Workflow
When the Grade Becomes a Puzzle: A Real‑World Example You’ve just received a three‑camera shoot of a high‑contrast night street sequence. Camera A is a RED RAW clip with a strong teal‑orange look, Camera B is an ARRI Alexa in Log C, and Camera C is a Sony FS7 shot in S‑Log3. The director wants each camera to retain its “signature” feel while the overall scene should feel cohesive, with the neon signs popping without crushing the shadows. A single linear node chain will quickly become a tangled mess of competing corrections, and trying to layer everything in one node will force you to make compromises that degrade image quality. The solution? Design a multi‑node architecture that isolates each camera’s primary correction, runs parallel secondary passes for creative tweaks, and then stitches everything together with serial refinement. By treating each camera as its own sub‑tree and then merging the results, you retain flexibility, keep processing efficient, and preserve the full dynamic range for later grading stages. Below is a step‑by‑step exploration of how to build such a workflow, why you’d choose parallel versus serial structures in different scenarios, and how to embed custom scripts and macros to automate repetitive tasks. --- 1. Building Multi‑Node Trees for Isolated Adjustments 1.1. The Rationale for Isolation When grading multiple sources with divergent color spaces, the first node (or node group) should act as a “clean‑room” where you bring each clip into a common working space. This isolation accomplishes three things: 1. Preserves headroom – By performing only the essential primary corrections (gamma, exposure, white‑balance) early, you keep the full 16‑bit precision for later passes. 2. Prevents cross‑contamination – Adjustments made for one camera will not inadvertently affect another’s tonal balance. 3. Facilitates parallel processing – Once each source sits on a neutral baseline, you can branch off into parallel nodes that operate independently. 1.2. Constructing the Primary Sub‑Tree 1. Node 1 – Input Transform Set the Color Space and Gamma to match the source (e.g., RED Film → Rec.709). Use the Input LUT if you have a camera‑specific conversion (e.g., ARRI Log C → Linear). 2. Node 2 – Primary Lift/Gamma/Gain Apply a Lift to bring the shadows up just enough to avoid clipping. Adjust Gamma to set the mid‑tone reference (often using a mid‑gray patch). Use Gain to bring the highlights to a target peak (e.g., 95 % for Rec.709). 3. Node 3 – Primary Saturation & Hue A modest Saturation boost compensates for the desaturation that often occurs during Log → Linear conversion. If the camera’s hue response is skewed (e.g., Sony S‑Log3’s slight green bias), apply a Hue vs Hue curve here. 4. Node 4 – Noise Reduction (Optional) Place …
2. Deep Dive into Scopes & Waveform Interpretation
When a Broadcast Deadline Looms, the Scopes Become Your Stopwatch A live‑to‑air documentary crew has just handed you a 4 K log‑encoded night‑city sequence. The director wants the gritty neon look, but the network’s compliance team has already flagged the clip for “potential broadcast‑limit violations.” The waveform shows spikes flirting with the 100 IRE ceiling, the vectorscope is a kaleidoscope of magenta‑orange, and the histogram is a long tail of saturated reds. You have five minutes before the final render – every adjustment must be data‑driven, precise, and reversible. In the following sections we’ll turn that pressure into an advantage. By mastering the interplay of vectorscopes, waveforms, and histograms, you’ll be able to: Read each scope at a glance for color balance and exposure. Set and verify legal broadcast limits without sacrificing creative intent. Deploy custom scopes to isolate hue or luminance bands, feeding directly into a node‑based workflow. Diagnose and correct exposure and color problems using a systematic, scope‑first approach. All of this builds on the node architecture introduced earlier—particularly the parallel node groups that keep adjustments isolated and preserve headroom. --- 1. The Three Pillars of Scope Literacy 1.1 Vectorscope: Hue & Saturation in Polar Coordinates | Feature | What to Look For | Advanced Tip | |---|---|---| | Hue Position | Angle around the circle corresponds to dominant hue (e.g., 0° = red, 120° = green). | Use the “Hue vs Hue” curve from Node 2 to align skin‑tone vectors (≈ 30°– 45°) across multiple shots. | | Saturation Radius | Distance from center indicates chroma intensity. | When the radius exceeds the broadcast‑safe “saturation limit” line (often a dotted circle), consider desaturating in a parallel node to retain local contrast. | | Cross‑Talk | Overlap of multiple hue clusters can indicate color casts or over‑mixing. | Deploy a Custom Vectorscope (Scopes Custom Vectorscope) filtered by a hue qualifier to isolate the offending cluster. | Nuance: In a log‑to‑Rec.709 pipeline, the vectorscope will initially show a compressed chroma range. After the Input Transform (Node 1) and primary Lift/Gamma/Gain (Node 2), the hue positions should stabilize, while saturation expands. If saturation “blows out” before you reach the legal limit, the issue lies in the gain stage—pull back the Gain or add a parallel node with a subtle Saturation reduction limited to the offending hue band. 1.2 Waveform Monitor: Luminance Distribution & RGB Balance | Feature | What to Look For | Advanced Tip | |---|---|---| | RGB Overlay | Separate red, green, blue traces reveal color balance across the image. | Align the three traces using Node 2 – Primary Lift/Gamma/Gain; any persistent offset indicates a color bias that should be corrected with a Lift (for shadows) or …
3. Advanced Secondary Corrections & Power Windows
When a Single Mask Isn’t Enough Imagine a nighttime cityscape: a lead actor walks under a flickering streetlamp while a neon billboard glows behind him. The director wants the actor’s skin to pop, the billboard’s magenta to stay vivid, but the surrounding shadows to retain their moody texture. A single power window can isolate the actor, but it will inevitably include part of the billboard, and a simple qualifier on skin tones will spill onto the magenta sign. Solving this requires layered qualifiers, composite power windows, and meticulous feathering—the exact tools this chapter explores. --- 1. Crafting Complex Qualifiers 1.1 Multi‑Dimensional HSL Selections A basic qualifier isolates a hue band (e.g., “skin tones”). In advanced work, you’ll often need to slice the color space along two or three axes simultaneously: 1. Hue vs Hue – isolates a narrow hue range while excluding nearby hues that share similar luminance. 2. Saturation vs Saturation – useful for separating a highly saturated neon sign from a desaturated background. 3. Luminance vs Luminance – isolates bright highlights without pulling in mid‑tone skin. Tip: Use the Qualifier Curves (the graph under the qualifier tab) to draw custom response curves for each axis. This lets you create a “soft‑shoulder” selection that mimics the way the human eye perceives color boundaries. 1.2 Stacking Qualifiers When a single qualifier cannot cleanly separate the target, stack multiple qualifiers in series: - Why stack? Each qualifier narrows the selection, reducing spill. - How to manage: Place each qualifier in its own parallel node (see “Advanced Node Architecture & Workflow”) to keep the selection logic independent and avoid cumulative rounding errors. 1.3 Using the “Hue vs Hue” Curve to Combat Spill A common edge case is color spill from a bright source (e.g., the magenta billboard) onto the actor’s skin. The “Hue vs Hue” curve can be set to suppress the unwanted hue while preserving the target hue: - Create a narrow notch at the billboard’s magenta hue (≈ 300°). - Expand the curve gently around the skin hue (≈ 20–30°). - Result: The qualifier retains skin tones but rejects the magenta spill, even if luminance values overlap. 1.4 Qualifier “Key Output” for Precise Masking The Key Output button converts any qualifier into a mask that can be fed into a Power Window or used as a Node Mask: - Workflow: Qualifier → Key Output → Add as a Mask in the Power Window tab. - This bridges the gap between color‑based selections and spatial masks, enabling hybrid selections that are otherwise impossible with either tool alone. --- 2. Power Windows: Beyond Simple Shapes 2.1 Composite Window Operations Resolve’s Power Window panel offers Boolean operations to combine multiple shapes: | Operation …
4. Custom LUT Development & Integration
A Real‑World Prompt: The Signature Look of a Mini‑Series A director of a four‑episode drama has commissioned a “signature look” that must appear in every scene—muted shadows, a warm mid‑tone push, and a subtle teal‑to‑orange tilt. The colorist has already built a sophisticated node tree that includes: Node 1 – Input Transform (log → Rec.709) Node 2 – Primary Lift/Gamma/Gain with a custom hue‑vs‑hue curve Node 3 – A Power Window isolating the background, with a secondary saturation tweak Node 4 – A ResolveFX film‑grain effect (optional) The production schedule now demands that the same look be applied quickly to dozens of new clips, while still allowing per‑clip tweaks. The most efficient solution is to freeze the core aesthetic into a custom 3‑D LUT, import it back into Resolve, and then layer it selectively in the node graph. The steps below walk through that workflow, exposing the nuances that separate a “good enough” LUT from a production‑ready one. --- 1. Exporting a 3‑D LUT Directly from a Node Tree 1.1. When to Export Export a LUT only after the node tree contains static, deterministic operations. Anything that varies per frame (temporal NR, motion‑based ResolveFX) cannot be baked into a LUT and will be lost or cause artifacts. 1.2. The Export Procedure 1. Isolate the “look” node chain Create a Node Group (right‑click → Create Node Group) containing only the nodes that define the visual style. Ensure the group starts at the output of the Input Transform and ends just before any optional effects (e.g., grain). 2. Set the correct color space In the Color page, open the Color Management panel. Verify that the Timeline color space matches the intended delivery (usually Rec.709‑ST 2084 for SDR). 3. Generate the LUT With the node group selected, go to File Export 3D LUT (or right‑click the node group → Generate 3D LUT). Choose CUBE format – the industry standard for 3‑D LUTs. 4. Select LUT resolution 33³ (35 KB) is adequate for modest contrast tweaks. 65³ (500 KB) retains finer detail and reduces banding, especially important when the LUT will be applied to high‑dynamic‑range footage. 129³ is rarely needed and can balloon file size without perceptible gain. 5. Save with a version‑controlled name Example: SigLookv03Rec70965.cube 1.3. Edge Cases and Gotchas | Situation | Why It Matters | Remedy | |-----------|----------------|--------| | Non‑linear nodes (e.g., Log curves) inside the group | LUT generation linearises the output; hidden log curves can cause clipping or loss of headroom. | Convert all log‑based adjustments to linear space before exporting, or keep them outside the exported group. | | ResolveFX that cannot be baked (e.g., Film Grain, Glow) | These effects are frame‑dependent and will be stripped from …
5. HDR Grading & ACES Color Management
HDR Project Foundations: From Log to Master A feature‑length sci‑fi drama was shot on a Sony VENICE in S‑Log3, destined for both a theatrical release (PQ‑based HDR10) and a live‑stream broadcast (HLH). The DIT delivered the RAW files with a ACES‑cct input transform, but the colorist must now juggle two HDR standards while preserving the cinematic intent. This scenario illustrates why a disciplined HDR‑ACES workflow is essential: a single grade must translate cleanly across divergent transfer functions, peak luminances, and display pipelines. 1. Configuring Resolve for HDR10, HLG, and PQ | Setting | HDR10 (PQ) | HLG | |---------|------------|-----| | Timeline Color Space | Rec.2100 (ITU‑BT.2100) | Rec.2100 | | Gamma / Transfer | ST 2084 (PQ) | HLG | | Color Primaries | Rec.2020 | Rec.2020 | | Target Luminance | 1000 cd/m² (typical) | 1000 cd/m² (but roll‑off differs) | | ACES Working Space | ACEScct (or ACES2065‑1) | ACEScct (or ACES2065‑1) | 1. Open Project Settings → Color Management - Set Color Science to ACES. - Choose ACES Input Transform matching the camera’s log (e.g., Sony S‑Log3 → ACEScct). - For the Timeline Color Space, select Rec.2100; this gives you a neutral HDR canvas that can be mapped to either PQ or HLG later. 2. HDR Output Settings - In Color Management → Output, set Output Device Transform (ODT) to Rec.2100‑PQ for HDR10 master, or Rec.2100‑HLG for broadcast. - Keep Output Color Space at Rec.2100; Resolve will apply the chosen ODT at render time. 3. Display Configuration - Under Preferences → Video I/O, enable External HDR Monitor and select the appropriate Display Color Space (PQ or HLG). This ensures the scopes and viewer are driven by the correct transfer function. Tip: Preserve headroom early by grading in a linear ACES space (node 1) before any non‑linear transforms. This prevents “cross‑contamination” when you later branch into parallel PQ/HLG paths. 2. The ACES Node Chain in an HDR Context The ACES workflow maps neatly onto Resolve’s node architecture. A typical HDR chain looks like: 2.1 Input Transform (Node 1) - Purpose: Convert camera‑specific log to the ACES working space. - Best practice: Keep this node isolated (no additional grades) to avoid contaminating the linear data. 2.2 Primary Lift/Gamma/Gain (Node 2) - Apply the core creative look using the familiar Lift/Gamma/Gain controls. - Because you’re still in a linear ACES space, the Gamma knob behaves like a log‑style exposure rather than a traditional gamma curve—this is why node architecture from the previous chapter remains critical. 2.3 HDR Tone‑Mapping (Node 3) - Insert a HDR Tone‑Mapping node (or use the HDR Wheel in the Color page). - Key controls: - Highlight Roll‑off: Determines how the brightest values compress into the display’s …
6. Film Emulation & Look Development with ResolveFX
A Real‑World Challenge: Re‑creating Kodak 5207 on a Modern DSLR Imagine you’ve just wrapped a two‑day shoot for a period drama. The footage is clean, shot in Log C, and you’ve already balanced the primary lift/gamma/gain in Node 2 using the workflow from Advanced Node Architecture & Workflow. The director now wants the final picture to “feel like it was shot on Kodak 5207 film – you know, that warm‑golden skin tone, subtle grain, and a soft halo around highlights.” Your task is to layer film grain, halation, and diffusion on top of the already‑crafted primary grade, while preserving headroom and avoiding cross‑contamination with the underlying color correction. This is precisely where ResolveFX shines, and where the techniques in this chapter will give you a repeatable, project‑wide workflow. --- 1. Positioning ResolveFX in the Node Graph 1.1 Keep Primary Corrections Isolated - Node 1 – Input Transform (color‑space, gamma, input LUT) remains untouched. - Node 2 – Primary Lift/Gamma/Gain contains your main tonal balance. - Node 3 – Look Development becomes the sandbox for film‑emulation FX. Tip: Use a Node Group for the entire look (grain + halation + diffusion). This lets you toggle the whole aesthetic on/off without disturbing the primary grade. 1.2 Parallel Nodes for Independent Tweaks When you need to adjust grain density without affecting halation intensity, add Parallel Nodes inside the group: Because they are parallel, each module receives the same input (the primary‑corrected image) and writes its output back to the group’s composite node, preserving independence and avoiding cross‑contamination. 1.3 Order of Operations Matters 1. Film Grain – applies before any blur, preserving the original grain structure. 2. Film Blur (Diffusion) – softens detail, mimicking the scattering of light in emulsion. 3. Glow/Halation – adds the characteristic bloom around highlights after diffusion, ensuring the halo follows the softened edges rather than the original sharp peaks. --- 2. Building a Layered Look with ResolveFX 2.1 Film Grain – The Backbone of Emulation ResolveFX → Film Grain offers: | Parameter | Typical Range for Kodak 5207 | Typical Range for Fuji 3513 | |-----------|------------------------------|-----------------------------| | Grain Size | 0.35 – 0.45 (medium‑fine) | 0.20 – 0.30 (fine) | | Grain Strength | 0.15 – 0.25 (subtle) | 0.10 – 0.18 (lighter) | | Color Tint | Warm (R + 0.02, B – 0.01) | Neutral‑cool (R – 0.01, B + 0.02) | - Why not full intensity? Over‑grain can mask detail and cause compression artifacts, especially when delivering in 8‑bit formats. Keep the grain strength low enough to survive a 10‑bit to 8‑bit down‑convert. - Fine‑tuning: Use the Strength slider while watching the Noise scope. Aim for a noise floor that sits just above the Temporal NR …
7. Advanced Tracking & Stabilization for Color Workflows
The Challenge: A Moving Subject, Shifting Light, and a Grain‑Rich Film Look Imagine a late‑night street scene shot on a high‑speed 4K sensor. A motorbike darts across the frame, its headlights flickering as it passes under a series of neon signs. The director wants the bike to stay in crisp focus, the neon glow to retain its saturated punch, and the overall image to preserve the gritty film grain that was added in post‑production. To meet these creative demands you must track the bike’s motion, stabilize the footage without smearing the grain, and drive color corrections that respond to the bike’s changing exposure as it moves through varying light. All of this must sit cleanly within the node architecture you built in earlier chapters, respecting the primary‑secondary hierarchy, preserving headroom, and avoiding cross‑contamination of adjustments. The following sections walk through a production‑ready workflow that accomplishes exactly that, while highlighting the nuances, trade‑offs, and edge cases that separate a competent grade from a cinematic one. --- 1. Choosing the Right Tracker for the Job DaVinci Resolve offers three core tracker types, each with distinct strengths and limitations. Selecting the appropriate tracker early prevents re‑work and keeps the node graph tidy. | Tracker | Best Use | Limitations | |---------|----------|--------------| | Point Tracker | Small, high‑contrast features (e.g., a logo, a lamp post) that stay in view. | Fails when the feature is occluded or moves out of frame. | | Curve Tracker | Objects that follow a predictable path (e.g., a vehicle on a road) where a single point cannot capture rotation or scaling. | More computationally intensive; requires careful keyframe refinement. | | Planar Tracker | Flat surfaces that undergo perspective changes (e.g., a billboard, a moving screen). | Struggles with non‑planar geometry or heavy distortion. | 1.1. Practical Decision Tree 1. Is the target a single high‑contrast point? → Point Tracker. 2. Does the target move along a smooth trajectory but rotate/scale? → Curve Tracker. 3. Is the target a flat plane subject to perspective warps? → Planar Tracker. If the scene contains multiple moving elements (e.g., the bike and a passing car), you may need a combination of trackers, each feeding its own node branch. --- 2. Setting Up Trackers in the Color Page 2.1. Preparing the Clip 1. Create a dedicated tracking node - Add a Serial Node (Node 5) after your primary correction (Node 2). - Name it “Tracker – Bike” for clarity. 2. Isolate the subject with a temporary qualifier - Use a Hue vs Hue qualifier to pull out the bike’s dominant color (e.g., bright orange). - This reduces noise in the tracker’s analysis window. Tip: Keep the qualifier’s Softness low during tracking; you …
8. RAW & Log Footage Color Space Management
From Sensor to Timeline: Capturing the Full Story Imagine a high‑end cinema camera on a night shoot, the sensor flooded with the faint glow of streetlights. The raw files hold 15‑plus stops of dynamic range, while the log‑encoded footage from a DSLR only preserves the highlight detail. The difference between a usable grade and a washed‑out mess hinges on how you import, interpret, and convert those signals inside Resolve. This chapter walks through the exact steps—raw‑specific controls, log‑to‑working‑space transforms, and the management of gamma, gamut, and primaries—so you can keep that headroom intact and avoid the subtle “color shift” that often creeps in when the wrong conversion is applied. --- 1. Importing RAW: The First Line of Defense 1.1 Choose the Correct RAW Settings at In‑Clip Level When you drag a RAW clip onto the Media Pool, Resolve automatically creates a Camera RAW tab in the Clip Attributes. The three levers you must set before any node work begins are: | Setting | Typical Options | When to Prefer | |---------|----------------|----------------| | Decode Quality | Full Res / Half Res / Quarter Res | Full Res for final grade; Half/Quarter for proxy or memory‑constrained work. | | Debayering Method | Bayer (default) / X‑Trans / Foveon | Match the sensor’s color filter array; X‑Trans for Fujifilm, Foveon for Sigma. | | Color Space / Gamma | Camera‑native (e.g., Arri LogC, REDlogFilm, Blackmagic Film) | Keep the native log curve; only convert later in a controlled node. | Tip: Set Decode Quality to Full Res even if you later create proxies. Resolve will automatically generate proxies while preserving the original data for the final render, ensuring you never lose fidelity. 1.2 Using the Input Transform Node The Input Transform node (Node 1 in the “Advanced Node Architecture & Workflow” paradigm) is where you apply the Camera RAW settings. Rather than relying on the clip‑level defaults, create a dedicated parallel node that houses the raw decode parameters. This isolates the conversion from any downstream primary corrections, preserving the “preserves headroom” principle. - Why a parallel node? It prevents cross‑contamination with the primary lift/gamma/gain adjustments that follow, and it makes the raw‑to‑working‑space conversion repeatable across clips from the same camera. 1.3 Custom RAW Presets If you frequently shoot the same camera, save your raw settings as a Color Space Preset: 1. Open a clip’s Camera RAW tab. 2. Adjust Decode Quality, Debayering, Color Space, and Gamma. 3. Click the Save Preset button (the floppy‑disk icon) and name it descriptively (e.g., “ARRI Alexa‑LogC‑FullRes”). Now, any new clip can be right‑clicked → Apply Preset, instantly aligning it with your workflow. --- 2. Converting Log to Working Spaces Log footage is already a linear‑ish representation of …
9. Collaborative Grading & Project Sharing in Resolve Studio
Setting the Stage: A Multi‑City Color Team in Action A post‑production house is handling a 12‑episode drama that was shot in raw ARRIRAW and LogC. The lead colorist in Los Angeles builds the primary look using the node architecture described in Advanced Node Architecture & Workflow. Simultaneously, two assistants in Toronto and London must refine secondary corrections, apply custom LUTs, and lock down HDR grading. All three need to work on the same project without stepping on each other’s toes, and the final grade must travel between machines without any loss of node fidelity or LUT integrity. --- 1. Building a Shared Resolve Database 1.1 Choosing the Right Database Engine - PostgreSQL (recommended for large teams) – supports concurrent connections, granular permissions, and native transaction logging. - MySQL – viable if your infrastructure already runs MySQL; ensure UTF‑8 encoding for node‑graph XML. - SQLite – only for solo or very small crews; it cannot handle simultaneous writes. Tip: Resolve Studio ships with a bundled PostgreSQL installer. Use the same version across all workstations to avoid protocol mismatches. 1.2 Deploying the Database 1. Install PostgreSQL on a dedicated server (or a VM with enough RAM for the expected project size). 2. Create a database named resolveprojects. 3. Add a service account (e.g., resolveuser) with a strong password and grant it CONNECT, SELECT, INSERT, UPDATE, DELETE on the database. 4. Configure pghba.conf to allow connections only from trusted IP ranges (your office subnets and VPN). 5. Enable SSL for encrypted traffic—critical when collaborators are remote. 1.3 Pointing Resolve to the Shared Instance - In DaVinci Resolve → Project Settings → General → Project Database, select PostgreSQL and enter the server address, port (default 5432), database name, and credentials. - Verify connectivity by creating a test project and checking that the Project Manager shows the new shared database. 1.4 Permissions and Role‑Based Access | Role | Typical Permissions | Use‑Case | |------|---------------------|----------| | Lead Colorist | CREATE, DROP, UPDATE on all tables; can modify project‑level settings | Sets primary node structure, manages ACES/ HDR pipelines | | Assistant | SELECT, UPDATE on timeline and grade tables; cannot delete projects | Refines secondary corrections, adds custom LUTs | | Intern | SELECT only on mediapool | Reviews media organization, adds markers | Use PostgreSQL’s GRANT statements to enforce these roles. Resolve respects the underlying DB permissions, preventing accidental project deletions. --- 2. Multi‑User Access Patterns 2.1 Optimistic vs. Pessimistic Concurrency - Optimistic (default): Resolve assumes conflicts are rare; it writes changes and resolves conflicts at save time. Good for distributed teams with high latency. - Pessimistic (timeline locking): Prevents simultaneous edits on the same timeline segment, eliminating merge conflicts at the cost of reduced …
10. Delivery Optimization: Color Space Conversion & Export Settings
A Real‑World Delivery Dilemma Imagine you’ve just finished a high‑contrast, HDR‑grade for a short‑form documentary shot on a Canon RAW sensor. The client wants three distinct masters: 1. A 4K YouTube upload (Rec. 709, 8‑bit, 60 fps). 2. A broadcast‑ready 1080 p file for a regional TV station (Rec. 2020‑HLG, 10‑bit, 59.94 fps). 3. A cinema DCP (DCI‑P3‑XYZ, 12‑bit, 24 fps). All three must preserve the artistic intent you built using the Advanced Node Architecture & Workflow and HDR Grading & ACES Color Management chapters. The challenge isn’t just hitting the right codec; it’s ensuring that the color space conversion, gamut mapping, and metadata survive the journey from timeline to final file without unwanted clipping or shift. The following sections walk through the decision‑making process, the technical steps inside DaVinci Resolve, and the validation methods that guarantee a clean hand‑off for every platform. --- Understanding Destination Color Spaces Web & Social Platforms | Platform | Standard | Gamma / Transfer | Typical Bit Depth | Common Pitfalls | |----------|----------|------------------|-------------------|-----------------| | YouTube, Vimeo, Facebook | Rec. 709 (sRGB is acceptable) | BT.1886 (gamma 2.4) | 8‑bit (sometimes 10‑bit) | Over‑compression of highlights, loss of HDR intent | | Instagram Stories | Rec. 709 | BT.1886 | 8‑bit | Severe chroma subsampling (4:2:0) | Why Rec. 709? Most browsers and mobile apps decode to the Rec. 709 primaries with a BT.1886 gamma curve. Even when a source is graded in HDR, the final web master must be tone‑mapped to SDR; otherwise, you risk clipping in the viewer’s display pipeline. Broadcast | Standard | Primaries | Transfer | HDR Options | Bit Depth | |----------|-----------|----------|------------|-----------| | ATSC 3.0 (US) | Rec. 2020 | PQ (ST‑2084) or HLG | PQ for premium HDR, HLG for live | 10‑bit (often 12‑bit) | | DVB‑T2 (EU) | Rec. 2020 | PQ/HLG | PQ for VOD, HLG for linear broadcast | 10‑bit | Broadcast often demands Rec. 2020 because it offers a wider gamut than Rec. 709, and the HDR transfer functions (PQ or HLG) preserve the dynamic range you crafted in the HDR Grading & ACES workflow. Selecting the right transfer function is critical: PQ is mastering‑oriented, HLG is broadcast‑friendly. Cinema | Specification | Primaries | Transfer | Bit Depth | Frame Rate | |---------------|-----------|----------|-----------|------------| | DCI‑P3 DCP | DCI‑P3 (≈ 0.68 R, 0.32 G, 0.15 B) | XYZ (linear) | 12‑bit (10‑bit per channel in practice) | 24 fps (or 25 / 30) | Cinema DCPs are built on ACES‑cct or ACES‑2065‑1 pipelines, which map the graded image to the DCI‑P3 gamut and then to XYZ for the projector’s color space. The final DCP must contain a mastering display metadata block that tells the playback …
Continue learning
- Advanced Color Grading for Cinematic Videos: Mastering Professional TechniquesAdvanced Color Grading for Cinematic Videos: Mastering Professional Techniques — a free advanced-level guide covering advanced color grading for...
- Advanced VFX and Compositing Mastery for ProfessionalsAdvanced VFX and Compositing Mastery for Professionals — a free advanced-level guide covering advanced vfx and compositing for beginners. Learn with...
- Learn Final Cut Pro: Beginner's Step-by-Step GuideLearn Final Cut Pro: Beginner's Step-by-Step Guide — a free beginner-level guide covering learn final cut pro for beginners. Learn with clear...
- How to Edit TikToks and Reels: Beginner's Step-by-Step GuideHow to Edit TikToks and Reels: Beginner's Step-by-Step Guide — a free beginner-level guide covering how to edit tiktoks and reels. Learn with clear...