8-bit AI video to 16-bit cinema quality

Powered by

Welcome to VP Land! AI tools are maturing into professional territory while legacy players exit stage left. From cinema workflows to model testing platforms, it's a week of advancement and transition.

Last week’s poll showed you were split on whether studios will trust Sora’s IP opt-in model — some expect selective use of older or less valuable IP, while others see it as too risky despite the granular controls. Check out today’s poll below.

In today's edition:

ETC’s Topaz Workflow Brings AI Footage to Cinema Standards
TiVo Ends DVR Manufacturing After 25 Years
fal Launches Sandbox for Side-by-Side AI Model Comparison
Ireland Raises VFX Tax Credit to 40%

From 8-Bit AI to 16-Bit Cinema Files

USC's Entertainment Technology Center has developed a systematic three-step workflow using Topaz Video AI to transform compressed AI-generated footage into cinema-ready 16-bit HDR files.

The approach, demonstrated on their AI film The Bends and presented at Infinity Festival, addresses a fundamental challenge: while traditional filmmaking starts with high-fidelity source material and reduces quality through post, AI workflows must build fidelity up from low-quality 8-bit SDR output.

The Technical Challenge: As USC's Sahil Lulla explained, "We started with an 8-bit SDR, which is almost like your proxy. You have to raise fidelity through each downstream step." This inverted pipeline requires building quality rather than managing it down, fundamentally different from traditional camera workflows.

The Three-Step Topaz Pipeline

Step 1 - Repair with Nyx Model: Remove compression artifacts and banding at source resolution before upscaling. USC found Topaz's Nyx model, designed for denoising, more effective at debanding than dedicated debanding tools when configured properly.
Step 2 - Upscale with Gaia Model: Scale from 720p/1080p to 4K while maintaining natural textures. After testing all available Topaz models, they chose Gaia for its strength with organic content and motion consistency.
Step 3 - HDR Conversion with Hyperion Model: Transform SDR to true HDR using inverse tone mapping, expanding from Rec. 709 to Rec. 2020 color space, and converting to 16-bit EXR format suitable for professional color grading.

Results and Limitations: The workflow produces files that colorists can actually push and pull without falling apart, though they acknowledge the 16-bit EXR output isn't equivalent to native camera capture. The process remains labor-intensive, requiring shot-by-shot parameter tuning and sequence-level consistency work.

SPONSOR MESSAGE

Vu Studio: Create at the Speed of Thought

Vu Studio transforms virtual production into an accessible, all-in-one platform for filmmakers of all levels. This comprehensive software suite brings together storyboarding, environment building, previsualization, and shooting capabilities in a single intuitive interface – something never before available in one place.

The platform integrates with industry-standard tools like Frame.io, streamlining collaboration and feedback across your production team. Whether you're creating storyboards, building virtual environments, or controlling multiple display surfaces, Vu Studio puts everything at your fingertips.

Ready to create content at the speed of thought? Start your Vu Studio journey today at vu.network/vu-studio.

Mention "VP Land" when subscribing for an exclusive 10% discount on annual subscriptions.

Guillermo del Toro, Netflix Launch Stop-Motion Studio

Guillermo del Toro and Netflix are co-founding a new stop-motion animation studio that will function as both a production facility and educational center. The announcement came during the Venice Film Festival, where the pair were promoting del Toro's latest film.

Dual purpose approach: The facility will give artists opportunities to learn stop-motion techniques while working on actual productions, with students potentially joining projects as apprentices
Next project in development: Del Toro and Netflix are already working on a second stop-motion film based on Kazuo Ishiguro's novel The Buried Giant, following their successful collaboration on Pinocchio
International focus: Del Toro specifically wants to bridge European and Latin American creators, noting that stop-motion artists in Mexico "don't receive much support from the government"
AI resistance angle: Del Toro positioned stop-motion as uniquely "AI-proof" compared to other animation forms, calling it the equivalent of "slow food in the culinary industry"
Educational philosophy: Netflix's Ted Sarandos emphasized that stop-motion encompasses all filmmaking disciplines - costume design, set design, lighting, and camera work - making it particularly valuable for comprehensive film education

What matters: This represents a significant institutional investment in preserving traditional animation crafts. The underlying training model could genuinely help sustain specialized skills that are increasingly rare in the industry.

Framestore Contributes OpenQMC Renderer to ASF

Framestore has contributed OpenQMC, its quasi-Monte Carlo rendering library, to the Academy Software Foundation, according to the foundation's announcement. The C++ library promises to reduce rendering noise while cutting compute costs.

The tech addresses a fundamental challenge in VFX rendering: traditional Monte Carlo sampling creates grain and noise, especially with fewer samples. Getting clean images typically means longer render times, which limits creative iteration.

Key capabilities:

Quasi-Monte Carlo sampling that reduces noise compared to traditional random sampling
Cross-platform support for both CPU and GPU architectures
Production-ready integration through a straightforward API
Specialized tools for physically based lighting and volumetric effects

Framestore developed OpenQMC alongside their in-house renderer Freak over the past three years. The library has already been used on major productions including Paddington in Peru, Barbie, Wicked, and Thunderbolts.

Eddie AI Now Gives You Edit Notes

Eddie AI launched a new Feedback Mode that analyzes video edits up to 40 minutes long and provides timestamped, tactical suggestions for improvement. The AI reviews uploaded cuts and delivers specific feedback on pacing, story clarity, and rhythm—pinpointing exact moments that need attention rather than offering generic notes.

What Eddie AI analyzes and provides:

Pacing analysis with timestamped recommendations for tightening specific segments
Story clarity feedback, identifying moments wherethe narrative could be clearer
Rhythm improvements suggesting where to enhance the edit's flow
Exact timestamps for each suggested change, so editors can jump directly to problem areas
Actionable direction is integrated directly into the timeline workflow

The feedback feature is currently free during the introduction phase. Eddie AI also consolidated its previous ChatGPT, Social Media, and A-Roll Logging modes into a single Rough Cut Mode. Users can now import footage once and generate multiple edit versions—from long-form interviews to social media cuts—without re-importing.

The unified workflow supports direct integration with Premiere Pro, Resolve, and Final Cut Pro, letting editors pull sequences directly from their timeline.

ILM's Todd Vaziri breaks down various VFX shots.

Stories, projects, and links that caught our attention from around the web:

📺 The original DVR disruptor, TiVo, ends production of its iconic recording devices as the company pivots fully to software and licensing.

🖥️ fal launches Sandbox, letting users instantly run the same input across hundreds of AI models to directly compare speed, output quality, and cost.

🛑 Apple pulls the plug on its Clips video-editing app.

🇮🇪 Ireland raises VFX tax credit to 40%.

Joey and Addy break down Taylor Swift’s AI promo backlash, Sora’s shift to an opt-in IP model after creator outrage, leaked details of OpenAI and Jony Ive’s mysterious AI device, and more.

Read the show notes or watch the full episode.

Watch/Listen & Subscribe

Spotify | Apple Podcast | YouTube