Dolby Vision Mastering in DaVinci Resolve: The Complete Workflow

Introduction: The Dolby Vision Imperative

In today's premium content delivery landscape, Dolby Vision has become absolutely essential. Whether you are delivering to Netflix, Disney+, Apple TV+, or for theatrical exhibition, Dolby Vision is the gold standard for High Dynamic Range (HDR) content. It guarantees that the creative intent forged in the color suite is preserved across an incredibly diverse ecosystem of consumer displays.

The core difference between base HDR10 and Dolby Vision lies in the metadata. HDR10 utilizes static metadata (MaxFALL and MaxCLL), meaning a single set of brightness values is applied globally to the entire film. If a movie has one incredibly bright explosion but is otherwise a dark thriller, the static metadata compromises the entire presentation. Dolby Vision, however, utilizes dynamic metadata, providing frame-by-frame or shot-by-shot instructions to the display.

The primary challenge and massive advantage of this workflow is the "one master to rule them all" philosophy. You create a single HDR "hero" master, and through the power of dynamic metadata, automatic derivatives (like SDR 100 nits or HDR 600 nits) are generated flawlessly. While this workflow is technically complex and requires strict adherence to color science protocols, it is a necessary evolution that saves enormous amounts of time and resources compared to doing manual, separate grades for every single delivery format.

Why a Specific Workflow

Mastering in HDR is only half the battle. Delivering an image that accurately translates your creative intent across thousands of different consumer displays—from high-end 2000-nit OLEDs to budget 300-nit LCDs—requires intelligent tone mapping. Dolby Vision uses dynamic, shot-by-shot metadata to instruct the consumer display precisely how to map the mastered image to its specific hardware capabilities. DaVinci Resolve provides a complete, natively integrated toolkit to generate, refine, and export this metadata.

The iCMU Remapping: How It Really Works

The internal Content Mapping Unit (iCMU) inside DaVinci Resolve acts as the brain of the Dolby Vision pipeline. During the analysis phase, the iCMU scans the image and calculates three crucial data points per shot: minimum luminance, average luminance, and maximum luminance (this forms the Level 1 or L1 metadata).

Consider a concrete example: a shot features deep shadows at 0.1 nits, actors' faces (midtones) at 100 nits, and a bright practical lamp at 1000 nits. If the iCMU needs to remap this 1000-nit master for a standard 100-nit SDR television, it uses those three L1 values to mathematically compress the dynamic range. It will roll off the 1000-nit highlight to fit within 100 nits while trying to preserve the contrast of the 100-nit faces. Because this mathematical compression can sometimes alter the perceived contrast or saturation in unintended ways, artistic trims are necessary afterward to allow the colorist to restore the exact creative intent.

Static vs Dynamic Metadata: The Dolby Vision Advantage

The ultimate advantage of Dolby Vision is efficiency through versatility. A prominent use case is modern studio delivery: the exact same 1000-nit HDR image sequence, paired with its Dolby Vision XML, can be ingested by Netflix to stream to HDR TVs, tone-mapped by Disney+ for legacy SDR devices, or sent to a theatrical facility. This drastically reduces the time and resources required in post-production, eliminating the need to render and QC a half-dozen different discrete master files.

Prerequisites: The Reference Monitor

Before attempting a Dolby Vision grade, you must have a proper viewing environment. This means a reference HDR monitor (e.g., Flanders Scientific XVS-3114, Sony BVM-HX3110) capable of hitting at least 1000 nits, connected via a dedicated video I/O device (like a DeckLink or UltraStudio), not the GPU output. The monitor must be strictly calibrated to ST.2084 (PQ) and Rec.2020 or P3-D65 color spaces, meeting EBU Tech 3320 Grade 1 standards.

Why These Specs Exactly?

• Peak luminance >=1000 nits: Required to accurately judge high-frequency highlights, specular reflections, and the true roll-off of your HDR signal.
• Black level <=0.001 nit: Essential for evaluating deep shadow details and ensuring blacks aren't artificially lifted or crushed.
• Contrast ratio >=1,000,000:1: Necessary to reproduce the massive dynamic range HDR demands without optical blooming or haloing.
• Gamut P3/BT.2020/BT.709: The monitor must physically be able to display the wide color gamuts used in modern cinema and broadcast.
• White point D65: The universal daylight reference (6504K) ensuring global color consistency.
• EOTF ST.2084 (PQ): The absolute perceptual curve designed specifically for HDR, mapping digital code values to exact nit levels.

Alternatives and Compromises

While "Tier 2" or prosumer monitors (like high-end OLED TVs) are cheaper, they lack the precision, stability, and uniformity of Tier 1 reference displays. The risks of using a Tier 2 display for mastering are high: incorrect calibration or automatic brightness limiters (ABL) can lead to a master that looks fine to you but is instantly rejected during studio Quality Control (QC). For example, a non-Tier 1 monitor might have raised black levels, causing the colorist to aggressively crush the shadows in the grade, ruining the image for properly calibrated consumer displays.

Viewing Environment

The physical room is just as important as the monitor. Ambient light must be strictly controlled, typically no brighter than 5 nits of neutral D65 (6500K) bias lighting behind the display. Why? The human eye constantly adapts to its surroundings. If the room is too bright, your pupils constrict, and the deep HDR blacks will appear as muddy grays, prompting you to incorrectly alter the contrast.

1. Prepare the Timeline

Your timeline needs to be clean. This means flattening multi-cam clips, removing unneeded tracks, and ensuring there are no hidden gaps. Dolby Vision metadata is generated on a per-clip basis; any nested timelines or compound clips can confuse the analysis engine. Make sure all edits are finalized—re-conforming after Dolby Vision metadata has been generated can be a severe headache.

Why V1 Only?

The Dolby Vision iCMU analysis engine inside Resolve fundamentally operates by looking at the lowest active video track—typically V1. In complex post-production use cases involving multi-cam edits, adjustment layers, or stacked VFX plates across V1, V2, and V3, the analysis can fail or yield unpredictable metadata. The standard solution is to select the stacked clips and create a Compound Clip. This flattens the visual data into a single analyzable element on V1. For instance, if you have a base plate on V1 and a greenscreen key on V2, nesting them into a Compound Clip ensures the iCMU analyzes the final composited image perfectly.

Managing Titles and Overlays

Graphic elements, lower thirds, and subtitles require special attention in HDR. Title luminance should generally sit between 100 and 400 nits, with 200 nits being the industry sweet spot. Why? A pure white title pushed to 1000 nits in HDR will be blindingly bright to the viewer and will almost certainly cause aggressive clipping and blooming when the iCMU maps it down to SDR.

Common pitfall: Leaving white text at maximum code value.
Best practice: Always test your titles by switching your Dolby Vision target display to Rec.709 (SDR) before running the final analysis. If the text looks blown out or causes the background image to dim drastically, lower the nit value of the graphics.

Blanking and Output Sizing

Blanking refers to the black bars around the active image area (pillarbox for 4:3, letterbox for 2.39:1 scope). The iCMU must ignore these black bars during analysis, otherwise, the L1 minimum luminance will always read as 0, ruining the metadata mapping. In Resolve, global blanking should be applied via Timeline > Output Blanking. If you need to adjust framing shot-by-shot, use the Sizing > Output Sizing palette.

Pitfall: Applying blanking using a node in the Color page and forgetting to exit Timeline mode, causing the blanking to be analyzed as part of the image.
Best practice: Strictly verify your blanking settings at the very end of your timeline conformation, utilizing the L5 metadata aspect ratio controls to define the active image area.

2. Project Settings

In Resolve's Project Settings under Color Management, you must use a color-managed workflow. While you can use Resolve Color Management (RCM), the industry standard for Dolby Vision is ACES. Set your ACES Output Transform to match your mastering monitor (e.g., P3-D65 ST2084 1000 nits). This absolute mathematical framework is crucial for the internal Content Mapping Unit (iCMU) to understand the exact light levels of your image.

Why PQ (ST.2084) and Not Something Else?

Dolby Vision relies fundamentally on the PQ (Perceptual Quantizer, SMPTE ST.2084) electro-optical transfer function. Unlike traditional gamma, PQ is an absolute curve. This means that a specific digital code value always maps to a precise, absolute nit level on the display. This absolute precision is the bedrock of Dolby Vision. However, its disadvantage is that it is highly sensitive to ambient light, hence the strict 5-nit surround requirement. In comparison, HLG (Hybrid Log-Gamma) is a relative system that adapts to the screen's peak brightness automatically, which is better for live broadcast but lacks the frame-by-frame precision required for premium cinematic Dolby Vision mastering.

P3 for Working, Rec.2020 for Delivery

While the delivery container for Dolby Vision is almost exclusively BT.2020, colorists typically limit their working color volume to P3. Why work in P3? Because no current consumer or mastering display can fully reproduce the massive BT.2020 gamut. P3 is smaller, fully coverable by Tier 1 monitors, and much easier to accurately control. Resolve handles the conversion automatically via color management, placing your P3 grade safely inside a BT.2020 container.

Pitfall: Exporting the final master with P3 color primaries flagged in the file header instead of BT.2020. This will result in an immediate QC rejection from platforms like Netflix.
Best practice: Always verify that your final output color space and metadata tags are explicitly set to BT.2020.

203 Nits Reference and P3-D65

In HDR, diffuse white (the brightness of a well-lit white piece of paper or a bright Caucasian face) is strictly defined by the ITU at 203 nits. This provides a physiological anchor point for the viewer, ensuring that midtones don't look artificially bright, leaving the remaining 800+ nits entirely for spectacular highlights. The P3-D65 color space simply refers to the P3 color gamut anchored to the D65 (6504K) daylight white point, yielding richer, more saturated colors than traditional Rec.709 SDR.

3. Enable Dolby Vision

Still in Project Settings, check the "Enable Dolby Vision" box. You will be prompted to select a version (v2.9 or v4.0/v5.0). Version 4.0 and above are the modern standard, offering much better interpolation and secondary trim controls. Choose your mastering display peak brightness, and set your target display (usually Rec.709 100 nits for the SDR trim pass).

Analysis Tuning: Balanced vs Highlight Detail

When configuring Dolby Vision v4.0+, you will notice an "Analysis Tuning" dropdown. The default setting is "Balanced," which provides an excellent compromise for 90% of cinematic shots. However, if your scene has complex highlight dynamics, you can adjust this.

Use cases: Use "Most Highlight Detail" for a shot dominated by a bright sunset where preserving cloud textures is critical. Conversely, use "Least Highlight Detail" for a dark, brooding scene where you want to prioritize shadow information over tiny, specular practical lights. Avoid using "Legacy CM4 L1" as it utilizes older, less efficient algorithms.
Best practice: Always start with Balanced. Only change the tuning on a per-shot basis if the default analysis causes unacceptable tone mapping.

Mastering Display: Choose the Right One

You must precisely declare the monitor you used to create the HDR grade. An example selection would be "1000-nit, P3, D65, ST.2084, Full". As noted earlier, P3 is chosen over BT.2020 because it reflects the actual physical capabilities of the monitor. The "Full" vs "Video" data levels distinction rarely impacts the internal math of the iCMU, but should match your video I/O settings.

Pitfall: Choosing a 4000-nit mastering display in the software while grading on a physical 1000-nit monitor. The iCMU will generate highly inaccurate metadata.
Best practice: Triple-check that the software mastering display matches your physical hardware specs exactly.

4. The iCMU Analysis

Once your HDR grade is completely locked, switch to the Color page and open the Dolby Vision palette. Select all clips in your timeline and click "Analyze." Resolve's iCMU will scan every frame, calculating the minimum, maximum, and average luminance levels (L1 metadata). This mathematical pass defines the boundaries of your image for the tone-mapping algorithm.

How Min/Average/Max Analysis Works

The L1 metadata represents the core of the dynamic nature of Dolby Vision. For every single shot, the algorithm identifies the darkest pixel (Min), the brightest pixel (Max), and the geometric mean of the entire image (Average).

If a shot features deep space (0.001 nit Min), a bright starship hull (100 nit Average), and a thruster engine flare (1000 nit Max), the iCMU maps these three parameters to the target display's volume. If this analysis captures an anomaly—like a single dead pixel or a brief compression artifact—the metadata will be skewed.
Pitfall: Assuming automatic analysis is always perfect. Bad analysis equals bad remapping.
Best practice: Scrub through the timeline and visually verify the L1 boundaries in the Dolby Vision palette before proceeding to trims.

Home and Cinema Targets: When to Use

You will be mapping your master to different target displays depending on the deliverable.
- Home SDR: Target Rec.709 100 nits (for standard legacy televisions).
- Home HDR: Target P3/Rec.2020 600 nits (for mid-range HDR consumer TVs).
- Cinema SDR: Target Gamma P3 48 nits (for standard DCI theatrical projectors).
- Cinema HDR: Target Dolby Cinema HDR PQ P3 108 nits (for premium Dolby theatrical projectors).

Use case: For Netflix delivery, you will heavily scrutinize the Home HDR and Home SDR targets. For theatrical releases, the Cinema targets take priority.
Best practice: Test your grade against both your lowest target (100 nits SDR) and a mid-tier target (600 nits HDR) before final delivery to ensure metadata robustness.

Analyze All vs Selected vs Blend vs Frame

Resolve offers several analysis modes. "Analyze All" processes the entire timeline (usually faster than real-time). "Analyze Selected" is used to re-analyze specific shots after you've made a late color correction. "Blend" allows you to link the metadata of several similar shots within a scene to prevent jarring luminance jumps during cuts. "Frame" analyzes only the current frame and applies that static metadata across the entire clip length (useful for shots with strobe lights or rapid flashes that confuse the temporal analyzer).

Pitfall: Modifying a color node and forgetting to re-analyze the clip, resulting in stale metadata that does not match the actual video pixels.
Best practice: Always run a final "Analyze All" immediately before export.

5. The Trims: Manual Remapping Correction

Turn on the target display output in the Dolby Vision palette to preview the SDR tone-mapped version. The automated iCMU analysis is mathematically correct but not always artistically pleasing—highlights might roll off too aggressively, or shadows might crush. Use the Trim controls (Lift, Gamma, Gain, Saturation, etc.) to manually adjust the SDR mapping on a shot-by-shot basis. These adjustments are saved as dynamic metadata, leaving your HDR hero grade completely untouched.

Why Trims Are Necessary

While the iCMU is highly advanced, it is ultimately a mathematical algorithm. It does not understand narrative intent. For instance, in a high-contrast shot where a character stands against a bright window, the iCMU might aggressively compress the highlights to save the window details, inadvertently darkening the character's face in the SDR translation. Trims allow the colorist to manually intervene, perhaps using "Highlight Clipping" to sacrifice the window detail in order to lift the character's face back to a pleasing level in SDR.

Pitfall: Over-using trims can lead to an unnatural, unstable image that breaks the perceptual link to the HDR master.
Best practice: Apply subtle trims. The goal is translation, not a complete re-grade. Less is always more.

The 21 Primary Trims: Usage Guide

Dolby Vision v4.0 provides a sophisticated suite of 21 trim controls. Key tools include:
- Lift/Gamma/Gain: Similar to standard color wheels, but they operate specifically on the tone-mapped signal. Use Lift to rescue crushed shadow detail, and Gain to restore punch to compressed highlights.
- Saturation Gain & Chroma Weight: Manage color volume compression. If the SDR map looks desaturated, gently boost Saturation Gain.
- Tone Detail: Injects micro-contrast back into the image, highly useful when the SDR compression makes textures look flat or washed out.
- Mid Contrast Bias & Mid Tone Offset: Powerful tools to shift the overall perceived exposure of the midtones (skin tones) without breaking the shadow/highlight anchors.

Pitfall: Using multiple opposing trims at once (e.g., heavily lifting shadows while aggressively dropping gamma) creates unstable mathematical curves.
Best practice: Address the primary issue with a single trim control first. Assess the impact before layering additional adjustments.

Metadata Levels L1-L9: Understanding the Hierarchy

Dolby Vision metadata is structured into hierarchical levels.
- L1: The fundamental min/avg/max analysis (Mandatory).
- L2: Trim adjustments targeting specific displays, like 100 nits (Highly recommended).
- L3: Additional offset adjustments (Rarely used in modern v4.0 workflows).
- L5: Defines aspect ratio and active image area / blanking (Crucial for proper mapping).
- L6: MaxCLL/MaxFALL values, serving as a static HDR10 fallback ceiling (Optional).
- L8: Extended CM v4 specific trims (Automatically generated when using modern trims).
- L9: Mastering display color primaries information (Optional but good practice).

Use case: For 90% of streaming deliveries, focusing exclusively on L1 analysis and L2 trims for a 100-nit target is completely sufficient.
Pitfall: Manually forcing unnecessary metadata levels can bloat the XML file and confuse downstream encoders.

Dolby Vision Trims License: Cost and Alternatives

To unlock the artistic trim controls inside DaVinci Resolve, you must purchase the Dolby Vision Trims License directly from Dolby. It costs approximately 1,000 USD. This is a one-time purchase, and the license is tied to your machine (or network dongle). This license unlocks the 21 primary trims as well as the advanced six-vector color secondary trims, applicable to both home and cinema targets.

Alternative: You can execute a Dolby Vision workflow in Resolve Studio for free, but you are limited to the automatic iCMU analysis (L1) and cannot apply any manual artistic trims.
Best practice: For small, budget-constrained projects, automated L1 mapping may suffice. For premium streaming or feature films, the 1000 USD license is absolutely mandatory and should be amortized across your project budgets.

6. Export the XML

After finalizing your trims, navigate to the Deliver page. You must export the Dolby Vision metadata as an XML file. This tiny sidecar file contains all the L1 (analysis) and L2/L3/L8 (trim) metadata mapped to your timeline timecode.

Why Export the XML?

The XML file is the absolute core of the Dolby Vision deliverable. Without this sidecar file, your massive, uncompressed HDR master video is effectively just a standard HDR10 file. The XML contains all the L1-L9 intelligence, telling downstream systems how to handle your image. For deliveries to Netflix, Disney+, or theatrical distributors, the XML travels alongside the video master.

Pitfall: Delivering the heavy video file but forgetting to export or attach the XML sidecar, rendering the entire Dolby Vision workflow useless.
Best practice: Build a habit of exporting the XML file to the exact same directory, at the exact same time, as your video master.

Canvas Aspect Ratio and Color Encoding Primaries

When exporting the XML, you must declare two critical parameters. First, the Canvas Aspect Ratio, which reflects the physical framing of your active image (e.g., 16:9, 2.39:1). This directly ties into the L5 blanking metadata. Second, the Color Encoding Primaries.

Crucial Rule: The Color Encoding Primaries MUST be set to Rec.2020, as this is the standard HDR delivery container.
Pitfall: Accidentally setting the XML export to P3 primaries because you worked in P3. This mismatch between the XML header and the BT.2020 video file will trigger an immediate, automated QC rejection at the studio level.
Best practice: Double-check the XML export dialog box. Canvas matches framing; Primaries match BT.2020.

7. Master Export (Archive and Mezzanine)

Export your HDR master video file. This is typically a 12-bit or 16-bit format like ProRes 4444 XQ or a 16-bit TIFF image sequence in the PQ transfer function. This file is your "mezzanine." It contains no metadata itself—it relies on the XML file exported in the previous step to function as a complete Dolby Vision package when delivered to a platform like Netflix.

Deep Archive vs Mezzanine: What's the Difference?

In high-end post-production, we distinguish between two types of masters. The Deep Archive master is the absolute highest quality rendering possible, typically a 16-bit uncompressed TIFF or EXR image sequence. Its purpose is long-term preservation. The Mezzanine file is slightly compressed (visually lossless) for distribution logistics. This is typically a 12-bit ProRes 4444 XQ or a JPEG2000 sequence.

Pitfall: Confusing the two concepts, resulting in trying to upload a 5-terabyte uncompressed EXR sequence to a streaming platform over a slow connection.
Best practice: Generate the uncompressed Deep Archive for the studio vaults, and generate the 12-bit Mezzanine for actual platform delivery.

IMF (Interoperable Master Format): The Standard

While delivering a ProRes file + XML sidecar is common, the ultimate standard for modern VOD delivery is IMF (Interoperable Master Format). IMF is a standardized container that elegantly packages the high-quality 12-bit JPEG2000 video, multi-channel audio, and interleaves the Dolby Vision XML metadata directly into a single complex folder structure.

Use case: Netflix requires IMF App 2E delivery for Dolby Vision content.
Pitfall: IMF packaging is highly complex and strict. A single misconfigured metadata flag will break the package.
Best practice: Utilize DaVinci Resolve’s built-in IMF packaging presets (like the Netflix preset) to ensure the Dolby metadata is properly interleaved.

UHD vs HD: Where to Export

The vast majority of Dolby Vision HDR content is expected to be delivered in UHD 4K (3840x2160). While 1080p HD HDR is technically possible, it is commercially rare.

Crucial Step: To properly export UHD, you must set your timeline resolution in Project Settings > Master Settings. Changing the resolution solely on the Deliver page can lead to upscaling artifacts or metadata misalignment.
Pitfall: Working in an HD timeline for performance, and forgetting to switch the Master Settings to UHD before the final render, resulting in a rejected HD master.
Best practice: Always verify project-level resolution settings prior to initializing the final render queue.

8. Standard Export (Viewing and Social Media)

For clients, web review, or social media, you can render a "hard" SDR version. In the Deliver page, you can instruct Resolve to bake the Dolby Vision tone mapping directly into the rendered file (e.g., an H.264 Rec.709 file). This ensures the SDR version matches your exact trim decisions without requiring the playback device to understand dynamic metadata.

Native Dolby Vision Encoding: Mac vs PC

If you want to export a compressed MP4/H.265 file with embedded Dolby Vision metadata (for viewing on an iPhone or a compatible TV via USB), the platform matters. macOS natively supports Dolby Vision encoding via Apple's AVFoundation framework built into Apple Silicon chips. Windows PCs do not feature this native OS-level support; encoding true Dolby Vision H.265 on a PC requires a specialized Dolby encoding license or a specific hardware environment.

Pitfall: Attempting to generate a client-review Dolby Vision MP4 on a Windows machine without the proper licensing environment, resulting in a standard SDR or broken HDR file.
Best practice: Utilize macOS workstations for generating quick, compressed Dolby Vision review files.

Dolby Vision Profiles: 5.0 vs 8.1 vs 8.4

When encoding compressed Dolby Vision deliverables, you must select a "Profile."
- Profile 5: The classic, full Dolby Vision profile utilizing a proprietary IPTPQc2 color space. Used heavily by Netflix and Disney+. Unplayable on non-Dolby displays.
- Profile 8.1: Highly recommended for broad distribution. It uses a standard HDR10 (PQ/Rec.2020) base layer and layers the Dolby Vision dynamic metadata on top. If a TV doesn't support DV, it falls back to standard HDR10 smoothly.
- Profile 8.4: Utilizes an HLG base layer instead of PQ. Excellent for broadcast and iOS video capture.

Pitfall: Confusing the profiles, such as delivering Profile 5 to a client who only has standard HDR10 monitors, resulting in bizarre purple/green color shifts.
Best practice: Default to Profile 8.1 for maximum cross-platform compatibility.

Bitrate and Quality: 10-20 Mb/s

When rendering these H.265 review files, bitrate control is vital. For 1080p HD, aim for 10-15 Mb/s. For UHD 4K, 15-20 Mb/s provides an excellent balance between visual fidelity and file size. Furthermore, ensure your encoder is set to "Main10" profile, which guarantees the 10-bit color depth mandatory for preventing banding in HDR gradients.
Pitfall: Setting the bitrate too low causes macro-blocking in the shadows; setting it too high creates massive files that buffer endlessly on cloud review platforms.
Best practice: Conduct a short 30-second test encode to verify cloud playback performance on target devices before rendering the feature.

Tone Mapping: Convert to SDR

Tone Mapping in the Deliver page is entirely different from exporting a Dolby Vision file. Tone Mapping instructs Resolve to bake the iCMU analysis and your manual trims directly into the pixel data, outputting a standard 100-nit BT.709 BT.1886 SDR file. This file contains no metadata; it is a permanent SDR derivative.

Use case: Generating the SDR master for YouTube, social media marketing, or legacy broadcast delivery.
Pitfall: Mistaking a Tone Mapped SDR file for a Dolby Vision deliverable.
Best practice: Always generate a baked SDR master part of your final delivery package to ensure maximum distribution flexibility.

Preview Encodes on USB

A highly practical workflow is generating an H.265 Profile 8.1 MP4 file directly from Resolve onto a high-speed USB drive. You can plug this drive directly into a consumer LG or Sony OLED television to instantly verify how the Dolby Vision metadata behaves on actual consumer hardware.
Pitfall: Mistaking this highly compressed preview encode for a studio mezzanine master.
Best practice: Utilize USB previews strictly for client confidence checks in a living room environment.

9. Quality Control in Three Steps

Strict QC is mandatory.
1. Watch the HDR master from start to finish checking for digital clipping or gamut errors.
2. Switch the iCMU to your SDR target and watch the entire film again to verify your trims.
3. Re-import your exported mezzanine file and XML into a fresh Resolve project to ensure the metadata aligns perfectly with the video frames and timecode.

Step 1 - Metafier: Validate Metadata

Before submitting any files to a studio, you must run the XML through Metafier, a free command-line utility within the Dolby Vision Professional Tools. Using the terminal command ./metafier --validate your_file.xml, the tool rigorously checks the metadata version, mastering monitor specs, target display profiles, color encoding parameters, and shot-by-shot continuity of L1-L9 metadata.

Result: It will output a clear PASS or FAIL log.
Pitfall: Ignoring a Metafier warning. A FAIL flag means the studio's automated ingestion system will instantly reject your upload.
Best practice: Resolve ALL FAIL warnings before initiating the final upload.

Step 2 - MediaInfo: Verify Video Specs

MediaInfo is an indispensable free tool used to inspect the header flags of your exported video mezzanine. You must verify:
- Color range: Limited
- Color primaries: BT.2020
- Transfer characteristics: PQ
- Matrix coefficients: BT.2020 non-constant
- Mastering display color primaries: BT.2020

Use case: This guarantees that the ProRes or JPEG2000 file perfectly matches the expectations of the HDR container.
Pitfall: Assuming DaVinci Resolve flagged the file correctly without verifying. Software bugs happen.
Best practice: Make checking MediaInfo a mandatory step in your post-export checklist.

Step 3 - Visual Check: Watch the Image

Data validation is not enough; you must watch the final package. Create a brand new, empty Resolve project. In Project Settings, set Color Management to DaVinci YRGB (non-managed) and set the Timeline color space to match your file (e.g., Rec.2020 ST2084). Enable Dolby Vision and set the mastering nit level. Import your rendered video file. Go to the Color page, re-specify the blanking (which defaults to 1.77), and use the Dolby Palette Options to "Import data from XML" (or "Embedded data" if testing an IMF).

Use case: Play the timeline while toggling the target display output. This visually guarantees that the XML timecode perfectly syncs with the video frames.
Pitfall: Skipping this step because "it looked fine on the timeline." Export renders can occasionally drop frames or slip timecode, misaligning the metadata completely.
Best practice: Always perform a 100% real-time playback visual QC of the final assembled package.

The Dolby Vision Professional Tools Suite

For advanced workflows, studios utilize external Dolby tools like Metafier to validate, edit, and stitch XML metadata outside of Resolve. You can explore these enterprise solutions at Dolby's content creation portal. Access to certain software may require a Dolby Customer account.

Metafier: Manage Metadata

Beyond validation, Metafier allows you to edit, merge, and correct XML metadata files via command line. If a studio requests a minor change to the global mastering display metadata, you can use Metafier to update the XML instantly without having to re-export from Resolve.
Pitfall: The command-line interface can be intimidating and syntax errors are common.
Best practice: Use Resolve for primary XML creation, and reserve Metafier strictly for validation and minor surgical edits.

CM Offline: Render Offline

CM Offline is a powerful utility designed to generate derivative versions (Rec.709 SDR, HDR10, HLG) directly from your Dolby Vision master file and XML, completely bypassing the NLE.
Use case: Automating the creation of multiple deliverable formats overnight on a dedicated server.
Pitfall: CM Offline is notoriously slow and computationally heavy. Rendering a feature film can take many hours.
Best practice: Offload CM Offline rendering tasks to a dedicated background server node to avoid bottlenecking your main grading suite.

Mezzinator: Create Mezzanines

Mezzinator is a specialized utility designed to ingest a ProRes file and its corresponding XML, and output a fully compliant IMF package ready for Netflix or theatrical delivery.
Use case: When your NLE fails to properly multiplex an IMF, Mezzinator acts as the reliable, Dolby-certified fallback.
Pitfall: Mezzinator requires a deep understanding of IMF folder structures and CPL (Composition Playlist) logic.
Best practice: Attempt IMF creation inside DaVinci Resolve first. Use Mezzinator for complex troubleshooting.

CM Analyze: Analyze Offline

CM Analyze processes existing HDR10 video files, running per-shot or long-play temporal analysis to generate L1 metadata entirely outside of a color grading application.
Use case: A studio provides an old, flattened HDR10 master that needs to be upgraded to Dolby Vision for a streaming re-release.
Pitfall: Like CM Offline, the analysis process is slow and computationally intensive.
Best practice: Utilize this tool specifically for catalog restoration projects rather than active post-production.

Dolby Vision Training and Certification

Professionals wanting to master this workflow can pursue official certification. Dolby offers detailed training courses, and earning the certification proves to clients and studios that you understand the strict technical parameters of HDR delivery. Visit Dolby Vision Cinema for global industry standards.

Dolby Creator Lab: Free Training

The Dolby Creator Lab offers comprehensive, free online training modules. These courses cover everything from fundamental color science to advanced DaVinci Resolve workflows, featuring videos, diagrams, and quizzes. The content is localized in English, French, Spanish, Japanese, Korean, and Chinese, and typically takes 20-30 hours to complete. Access at Dolby Creator Lab.
Use case: Ideal for beginners and colorists preparing for formal certification.
Pitfall: Completing the free training does not grant you the official certification badge.
Best practice: Complete the entire Creator Lab curriculum before attempting the paid certification exam.

Individual Certification: 250 USD

The official certification, accessible via Dolby Certification Portal, costs 250 USD. It comprises a rigorous online multiple-choice exam followed by a complex practical test. You must download a supplied HDR sequence, add specific titles and fades, correctly generate L1/L2 metadata, perform specific trims, and export a compliant IMF package. The exam takes roughly 2-3 hours.
Use case: Essential for professional colorists seeking to validate their technical expertise to major studios.
Pitfall: You only get two attempts to pass the exam before having to repurchase the voucher.
Best practice: Ensure your DaVinci Resolve setup is perfectly stable and you have set aside dedicated, uninterrupted time for the practical exam.

Certification Benefits

Passing the exam grants you the right to use the official "Dolby Vision Certified Professional" logo on your website and marketing materials. More importantly, your name and contact information are added to the official Dolby Professional Directory, making you discoverable by producers and studios looking for qualified talent.
Use case: A powerful, verifiable marketing argument for freelancers and boutique post houses.
Pitfall: Believing certification alone guarantees high-quality artistic grading.
Best practice: Combine your technical certification with a strong artistic showreel to win premium clients.