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 is using Google’s Gemini image-generation and coding tools to help design parts for a Fallout-inspired Halloween prop with his son. Gemini supplied OpenSCAD code that produced a usable model after revisions, but the project is unfinished and several AI-generated designs missed the maker’s specifications.
A maker is using Google Gemini to help design 3D-printed parts for a Fallout-inspired Halloween prop he is building with his son, reporting that AI-generated OpenSCAD code helped turn a rough idea into a usable model after several rounds of prompting and revisions. The project is still in progress, and the maker says some generated designs ignored instructions or needed further changes.
The project began after the maker’s son chose a costume inspired by Fallout 4. A full DIY power-armor costume was ruled out because of the available time. Instead, they are planning a character with a decorated chest piece and a handmade prop. A delayed motorcycle chest protector also pushed the maker toward foam and craft materials for the costume.
For one prop component, the maker asked Gemini for help modeling a truncated cone with a mounting ring. After multiple iterations, he settled on a design made from two connected truncated cones. Gemini produced OpenSCAD code, which let him adjust parameters such as dimensions before printing a test part. He described that print as good enough to move to the next stage, despite using filament that was too wet.
The maker also used AI to explore a handle and other parts. Some output did not follow the requested geometry: one design failed to keep side arms flush, placed pill-shaped holes in the wrong areas and added unwanted chamfers. He edited code to change dimensions and add mounting holes, and is printing a low-density version of a revised handle. The finished prop may combine printed parts with a small wooden box, screws, glue and other materials.
AI Helps Bridge Idea and Model
The account offers a specific example of how a maker is using generative AI as a design aid rather than a finished-product button. The tool helped create editable code for a component the maker had not fully visualized, giving him a way to test geometry in three dimensions before committing to a more finished build.
It also shows the limits of that workflow. The maker still had to identify mistakes, refine instructions, change code and decide how separate pieces would fit together. For hobbyists who can work with modeling tools but struggle to translate a mental image into a buildable design, AI-generated starting points may help. The report does not establish how broadly the approach works, or whether it saves time across other projects.
As an affiliate, we earn on qualifying purchases.
From Fallout Costume to Prop
The maker initially considered modeling a recognizable fantasy tool shown in Fallout reference material, but judged that design difficult to build and opted for an original prop instead. He is discussing a possible creature detector concept with his son, who will ultimately decide what the prop is called. The plan is to build a rough first version together and refine it if time allows before Halloween.
The project uses several methods alongside AI. The maker is considering a Milwaukee hard hat for headgear and a flashlight attached with duct tape for a scavenged appearance. He also printed green air fittings using McMaster-Carr 3D files to check how they might fit with quarter-inch inner-diameter hoses; he has not decided whether to use those or inexpensive metal fittings.
“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 Work in Progress
The report does not say whether the final prop will be completed by Halloween, what it will look like, or whether the current low-density print will work as intended. The maker has not settled on the prop’s theme, final materials, or assembly method. He is considering whether to remodel parts in Fusion or similar software, but says that may not happen for this project.
The maker’s experience is a personal account, not a controlled comparison of AI design tools. The report gives no measurements of time saved, cost, print success rate or performance against manual modeling. It also does not establish whether the same workflow would work as well for other users or more complex assemblies.
A Rough Version Before Halloween
The maker’s immediate goal is to assemble a rough first version from printed parts and a small wooden box, then judge what needs revision. The next cone version is intended to be larger and include mounting holes for attachment to the box. The handle is being printed at low density for testing; if it works, the maker may print it in a final color or remodel it manually for greater control.
He plans to build the prop with his son, who will choose its identity. Whether they have time to refine the design, replace the temporary assembly approach or produce a more cohesive final model remains open.
Key Questions
What is the maker building?
He and his son are developing a Fallout-inspired Halloween prop. The precise concept is undecided; the maker has suggested a detector for creatures from the game’s world.
How is Gemini being used?
The maker used Google Gemini to help generate OpenSCAD code for printable parts and to explore possible component designs. He then revised the code and geometry himself.
Did the AI produce a finished design?
No. The maker reports that some outputs did not follow instructions, including requests about arm placement and hole locations. The design is still being tested and assembled.
Will the prop be ready for Halloween?
That is not confirmed. The maker’s stated plan is to produce a rough first version and refine it if time allows.
Source: rss
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
