Mari 8.0 beta and Nuke 17.1: betas aimed squarely at production artists
Foundry’s Mari 8.0 beta and Nuke 17.1 compositing beta are coordinated updates to its 3D texture painting and post-production tools, designed to tighten feedback loops for professional VFX and 3D artists by modernizing node workflows, 3D viewers, and emerging Gaussian splat relighting techniques. The headline is not the version numbers; it is the intent. Foundry is telling production teams that iteration speed is now a first-class feature. Mari focuses on smarter material authoring in the node graph, while Nuke doubles down on 3D data—USD, splats, and Hydra 2.0 viewer improvements. Both releases arrive as open betas, inviting studios to stress-test these changes in real shots before they harden into a stable release. In an industry where one extra look-dev pass can make or break a sequence, these tools matter less as shiny upgrades and more as direct attacks on wasted time.
Mari 8.0: hex tiling, color remap, and A/B comparisons reshape texture workflows
Mari 8.0 beta is a clear statement that 3D texture painting pipelines should be math-driven, repeatable, and visually auditable. The new Hex Tile node minimizes visible repeats by tiling and blending textures on a hexagonal grid rather than a conventional rectangular pattern, matching techniques now seen in Houdini, Redshift, and RenderMan. That alone pushes Mari deeper into the same language as other key VFX tools, which matters for studios juggling mixed DCC stacks. The Compare node introduces A/B wipe comparisons between different node streams, making it easier to iterate on materials or create look variants without duplicating graphs. Color Remap centralizes hue, saturation, exposure, and gamma adjustments in a single node so artists can grade materials in one place instead of hunting through scattered adjustments. Add over 49 new math nodes—covering arithmetic, trigonometry, conditional logic, and advanced randomization—and Mari’s node graph starts looking less like a paint app and more like a programmable material lab.
Workflow and platform upgrades: small changes with real pipeline impact
What makes Mari 8.0 compelling is not only the headline nodes but the attention to everyday friction. Trackball rotation-style navigation for the virtual camera gives artists an alternative to the standard Orbit tool, reducing the micro-irritations of navigating complex asset turntables all day. Archived projects now show a clear visual indicator on thumbnails, a tiny but meaningful safeguard against loading or overwriting the wrong work in busy asset libraries. Under the hood, Mari moves to the CY2025 VFX Reference Platform spec, a step forward from a current stable release that is four versions behind. This is more than housekeeping: aligning with a newer platform spec makes Mari less of a compatibility risk in modern Linux-based render and pipeline environments. Subscription pricing remains rental-only: individual licenses at USD 86 (approx. RM395) per month or USD 689 (approx. RM3165) per year, with team subscriptions at USD 1289 (approx. RM5920) per year, which is up USD 60 (approx. RM275) since Mari 7.5. That increase is modest in context, but it raises the bar on how much value these workflow tweaks must deliver in real production.”
Nuke 17.1: Gaussian splat relighting and Hydra 2.0 viewer push 3D compositing
On the compositing side, Nuke 17.1 leans into 3D Gaussian splatting and USD workflows instead of treating them as fringe features. The GeoSequencer node can process sequences of USD, PLY, or SPLAT files and output a time-sampled USDC file that fits into Nuke’s existing toolset, making animated splats a practical option for environment and set-extension work. The updated SplatRender node now supports 2D rendering of Direct, Point, and Spot lights placed in the 3D system, enabling basic Gaussian splat relighting rather than locking artists into a single baked lighting solution. On top of that, the Hydra-based 3D viewer jumps to Hydra 2.0, opening new ways to display a scene with the current HdStorm render delegate, including previewing individual AOVs such as depth and position. Hydra 2.0 support is explicitly positioned to “pave the way” for third-party render delegates in future, which is a quiet but important hint that Foundry expects Nuke’s 3D viewer to keep pace with evolving USD render ecosystems.
Annotations, pricing, and why these betas matter now
Nuke Studio’s timeline tools show the same obsession with faster iteration that runs through Mari 8.0. The Annotations toolset gains a shapes system with rectangles and arrows, tuned for mouse or trackpad users with controls for color, width, and opacity. The Contact Sheet now works per sequence, with one-click loading of all visible tracks and the option to toggle between a grid view and a full-frame cell, which makes reviewing dailies and versions less of a manual chore. A new Compare Versions feature brings A/B comparisons into the timeline, supporting the same comparison modes as the Viewer so artists can judge shots in context rather than juggling separate views. Nuke 17.1 is in open beta on Windows 11, Rocky Linux 9.0, and macOS 15.0+, with no fixed stable release date but an expected window of three to six months. Annual rentals sit at USD 3839 (approx. RM17645) for Nuke, USD 5219 (approx. RM24000) for NukeX, USD 6379 (approx. RM29320) for Nuke Studio, and USD 462 (approx. RM2120) for Nuke Render, up USD 22 (approx. RM100). Nuke Indie subscriptions cost USD 499 (approx. RM2290) per year for artists earning under USD 100000 (approx. RM459000). In my view, these betas matter because they give studios a chance to pressure-test high-impact features—Hex Tile, Color Remap, Gaussian splat relighting, Hydra 2.0 viewer—on live shots, shaping tools that will define how quickly teams can experiment in 3D and comp for years to come.
