Let gravity do the boring bit
RealTransforms targets a familiar scene-dressing problem. A prop looks grounded from one camera, floats from another, and clips through its neighbour after the next adjustment. The package adds physics-assisted Move, Rotate, Drop and Arrange operations directly to the Unity Scene view.
The workflow keeps transform editing inside the editor. Artists can select props, toggle physics assistance and move them while collisions make objects slide, stop and settle against scene surfaces. Drop uses gravity. Rotate keeps contact in the calculation. Arrange handles groups rather than forcing the artist to repeat the same correction object by object.
Familiar hands, fewer nudges
The default control scheme follows the editor’s existing habits. Caps Lock toggles Physics Assist. W, E and R map to Move, Rotate and Arrange. Space acts as the primary action. The developer says every binding can be reassigned, and recommends putting the toggle on a mouse button for quicker access.
The primary action changes with the active mode. It can drop props, rotate them randomly, turn them forward or arrange them in a grid. The developer describes this action as useful even beyond the physics functions.

There is no separate loading step in the described workflow. The user toggles the mode and continues in the Scene view. That choice matters for a placement utility because every extra panel, modal dialogue or context switch competes with the tiny edit the tool is meant to accelerate.
Multiple selections can be clumped, spread, spaced, levelled or given random rotation. That moves the package beyond a gravity button. It also addresses repetition, which usually consumes more time than the final artistic adjustment.

Physics stays in the editor
The package uses PhysX through the engine’s built-in 3D physics setup. Its placement workflow runs in the editor rather than in Play Mode. The developer added logic that prevents it from operating during gameplay, specifically to avoid unexpected changes inside a running project.
That boundary makes the current use case clear. It is an authoring tool for building scenes, not a runtime object-placement system. Generated placements remain part of the authored scene, while the interactive physics assistance stays out of the shipped game logic.

Colliders meet the awkward shelf
Physics placement depends on useful collision geometry. A MeshCollider can represent complex surfaces, but a collider used with a Rigidbody generally needs a convex shape. A single convex hull cannot preserve inward spaces or hollow forms such as open crates and shelves.
Those shapes need compound collision, built from several convex pieces. One piece can represent the floor of a crate, others its sides and rails. The result leaves the opening usable while still giving the solver surfaces to contact.

This is the technical snag behind a deceptively simple interaction. Dropping a bottle onto a table is easy when both objects already have sensible colliders. Dropping it into a crate demands geometry that represents the cavity rather than sealing it with one hull.
The package includes a collider-generation approach for these cases. The developer built a voxel-based system after licensing conflicts prevented the inclusion of V-HACD and CoACD. Both are approximate convex decomposition methods that split difficult meshes into convex parts.

Compatibility is pleasantly dull
Version 1.0 was released on June 9, 2026. The listing identifies 6000.3.11 as the original editor version and reports compatibility with the Built-in Render Pipeline, the Universal Render Pipeline and the High Definition Render Pipeline under version 6000.3.11f1.
The download is 7.5 MB. The licence type is Extension Asset under the Standard Asset Store EULA. Extension Assets require seats for users who access the extension, so teams should match seat count to actual users rather than treating one purchase as a studio-wide utility licence.
An Unreal Engine port is in development. No release date, price or supported version has been specified. The developer has asked for feedback and says updates will address reported problems. That is a development intention, not a service-level guarantee. The version number, sparse rating data and the developer’s own language all point to a young tool.
Price before the props fall over
The store lists the package at 13.75 US dollars before tax, reduced by 50 percent from 27.50 US dollars at press time. Taxes and VAT are calculated at checkout.
At that price, the practical question is not whether gravity is clever. It is whether the tool saves enough repetitive placement time on the scenes a team actually builds. A shelf-heavy interior and a sparse exterior will produce very different answers.