TL;DR
Get business pricing on tools and workshop supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
A maker says Google Gemini helped generate OpenSCAD code for components of a Fallout-inspired Halloween prop he is building with his son. The AI output required repeated prompting and manual edits, and the project is still in progress, with a rough first version planned before Halloween.
A maker is using Google Gemini to help design 3D-printed parts for a Fallout-inspired Halloween prop he is building with his son, after deciding that a full DIY power-armor costume was beyond their available time. The project shows one practical use for AI in a small creative build: generating adjustable OpenSCAD code to help turn an early idea into printable components, though the maker says the output needed repeated prompts and corrections.
The maker initially wanted to create a costume connected to his son’s interest in Fallout 4. Rather than attempt a complete suit of power armor, they settled on a more manageable concept: a character wearing a decorated chest piece with a wasteland-survivor look. A delayed or apparently lost order for a motorcycle chest protector pushed the maker toward using foam and craft materials for the costume, while he concentrated on designing props.
For one component, he used Gemini to produce OpenSCAD code for a shape based on two connected truncated cones. He says the code let him alter parameters and test dimensions, and that an initial print was useful enough to guide another version. The next iteration is planned to be larger, with mounting holes so it can attach to a wooden box that will form the prop’s main structure.
The maker also used AI to generate a model for a handle, but says the results did not follow some instructions: the side arms were not flush, pill-shaped holes appeared in the wrong places, and unwanted chamfers were added. He changed some code parameters himself, including the number of mounting-plate holes and the dimensions of slots. A low-density print was underway when he wrote the report. He may later redesign the handle in Fusion or similar software for more direct control.
AI as a Workshop Design Aid
The report offers a grounded example of AI assisting with an intermediate step in a maker project rather than producing a finished object. For someone who finds it difficult to translate an idea into a multi-part model, generated code can provide a starting point that can be adjusted, tested, and printed. In this case, the maker says AI helped him visualize geometry before committing to a more detailed design.
The limits are part of the result. The model did not reliably follow dimensional and placement instructions, and the maker had to inspect the code, revise parameters, and decide whether conventional modeling software would be better for later changes. That makes the workflow less like automatic design and more like iterative prototyping: the person remains responsible for checking whether the generated part fits the intended assembly.
The project also reflects a common practical constraint in hobby builds: time. The maker plans to attach separately designed pieces to a small wooden box, potentially using screws, glue, or other fasteners, rather than build a fully integrated assembly. The approach may be less cohesive, he says, but could allow the pair to finish a rough version for Halloween and refine it later.
beginner FDM 3D printer with large build volume
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
From Costume Idea to Prop Parts
The maker’s original plan was to create a recognizable Fallout-themed costume for Halloween, but he concluded that full power armor was not realistic within the time available. A simpler costume, using a hard hat and a chest piece with faction-inspired decoration, left room to make the prop a more prominent part of the design. He also considered a scavenged-looking flashlight attached to the helmet.
He had experience modeling parts he already understood, but says new designs are harder when he has not settled on the shape in his head. He tried asking Gemini to break down a fantasy tool shown in Fallout-related reference material into printable subassemblies. The proposed breakdown was not especially useful, according to the maker, though it supported his earlier idea of using mounting pins for glued assembly. He and his son are instead developing an original prop, which the child can name once it is built.
Other parts of the build include green air fittings modeled from McMaster-Carr 3D files. The maker printed them to check how they might fit with quarter-inch inner-diameter air hoses, and says this helped him decide whether to order metal fittings without buying an unnecessary variety. He notes that the printed fittings are for visualization, not pressurizable use.
“The AI made a lot of mistakes, and I’m also learning how to better constrain dimensions in my prompts.”
— The maker, in the ToolGuyd report
The Prop Is Still a Prototype
The final design, finished materials, and exact assembly method have not been settled. The maker says the prop’s concept still largely exists in his head, and that the separate parts may be screwed or glued to a small wooden box, with hot glue or epoxy among the possible options. He has not said whether the current printed components will all make it into the final version.
It is also unclear whether the low-density handle print will fit the planned assembly or whether the maker will redraw it in Fusion or another modeling program. He describes the AI-generated subassembly breakdown as not very helpful, and says Gemini once timed out and dropped him to a lower AI tier until the limit reset. No independent evaluation of the designs or software output is included in the report.
A Rough Build Before Halloween
The maker’s stated next step is to assemble a rough first version of the prop in time for Halloween, then refine the design if time allows. That work includes printing and assessing the handle, enlarging the cone component, adding mounting holes, and fitting the parts to the wooden box. The maker may print the final components in black, but says that decision depends on how the prototypes work.
He plans to build the prop with his son, who can decide what the finished device is called. Further details on the foam costume pieces or the final build were not provided in the report. The outcome will show whether the AI-assisted parts can be assembled into a usable prop, but the maker has not claimed that the design is complete or that the project will meet its Halloween deadline.
Key Questions
What is the maker building?
He is developing an original, Fallout-inspired Halloween prop with his son, using separately designed components attached to a small wooden box. It is intended to accompany a simpler wasteland-style costume rather than a full power-armor suit.
How is Gemini being used?
The maker used Google Gemini to generate OpenSCAD code for shapes and parts, including connected truncated cones and a handle. He could adjust parameters in the code and use prints to test the designs.
Did the AI produce a finished design?
No. The maker says the output required repeated prompting and manual edits, and some instructions were ignored. The parts and overall assembly remain in development.
What happens to the printed parts next?
The maker plans to assess the prototypes, revise dimensions and mounting features, and attach selected parts to a wooden box. He hopes to make a rough version before Halloween, with further refinement possible afterward.
Source: rss
Fall yard work Picks
leaf blowers
As an affiliate, we earn on qualifying purchases.
