Kit Bodega - Gameoverse - Glitch Productions

Update - My incredible niece painted the model, so I am using it as the new main picture for it! Thanks Addie!

I hope you like this model! After my kids roasted me about my first version of this, I reworked multiple elements to get this one I am now sharing with you (improved goggles, stronger tail, smaller gun hand, better teeth, improved eyes and adjustments to the base). I hope they did the hard work for you on this one and you can just enjoy it. It prints up easily in PLA or PETG, and the supports are pretty simple but you should be cautious with the tail. 

I use AI to help me make these models, but wanted to give a little more context on my process, since it can be a sensitive topic for many. I start by watching the show and getting a sense of the characters. I then look at clips from the show as well as official character images from the company. In the past I have added more of my own creativity, but I have learned that most users really just want something as close to the character in the show as possible (so my creativity is usually more practical than exploratory in nature). When I have decided on an image I like, I then use various AI image generation tools to create multiple angles of the original 2d image. I usually will have it create them in a more 3d version with shading and shadows. This is important because it answers a lot of questions that a 2d drawing simply doesn't fully explore. When I have consistent good images from 4 angles, I load it into Hitem3d and generate my model. In this case I actually generated 13 full 3d versions before finding one I was mostly satisfied with. I am almost never 100% satisfied with the generated stl file so I then prepare it for adjustment in onshape. Getting the output of a 3d generative tool to be usable in onshape is not trivial, I have ended up actually creating a program (with the help of cursor) that cleans up the output of these AI tools and solves geometry and manifold issues that can be a nightmare. This tool essentially outputs a clean 2mb version I can edit easily in onshape. When I get the model into onshape, I then begin fixing all the things that bug me. For this specific model it was: the hands (I pretty much always need to fix the hands), the eyes, the teeth, the size of the gun, the thickness of various parts like arms and tail, and create a new set of goggles. I fix any modesty issues (overly defined butt cracks usually) and other oddities. This process usually takes a few hours. When I have something I like, I send it to bamboo studio and tweak the settings until I have something that I think will print up nicely. Then I print it. My first version is pretty much always a throw away version. I learn where the print fails, weak spots, and get brutal feedback from my kids. By version 3 or 4, I usually have something that is good enough to publish. In all, I probably put 6+ hours into each model and end up with multiple versions sitting on my desk. I know I have something good, if my youngest puts it in his backpack and takes it to school to show his friends. But inevitably I end up with a desk full of models that I try to give away to neighbors, nieces or nephews.  Once a model is live, I watch far to anxiously to see if anyone cares. Positive comments and pictures absolutely make my day and keep me motivated to do more. Negative feedback is usually helpful, especially when it is detailed. However, sometimes reworking  old models feels like a chore. That explanation was longer than I had planned, but I hope someone finds it interesting and maybe inspired to create your own models!
 

You can check out my other designs at my homepage here: https://spencermann.com

It is a work in progress, but I hope to provide more information about my models, and some tips if you want to create your own.