The Final Countdown (Patreon)

This might be the most bittersweet announcement I can make on this kind of holiday, but with the end of a year comes the beginning of something new. We close the door on the events that came before, and prepare for something new. So, even if this is going to be one of the most difficult announcements I have had to make, I feel that now is the best time to do this. It has to be this way.

Long story short; I’m officially announcing that 2024 will be my last year of support for SFM, and the last year I will be focusing on SFM output as the primary focus of future projects.
From January 1st, 2025 forward, I will be focusing on designing models primarily for use directly in Blender 3D. Any involvement from me in SFM modeling, porting, and rigging will be halted, and Source Engine specifications will no longer be a focus in my primary output.

It’s a distressing situation to be in. But you all know I am not a man of few words, so I will explain the situation and my reasoning as thoroughly as I can. I made a similar speech back in 2018 for Garry’s Mod, but this one is a much more emotional load to bear. This will be for everybody who is concerned or worried about what will happen going forward.
If you wish to skip to the questions section, please feel free. But for those who want the full in-depth story, here we go.

Why am I making this transition? There are 2 explanations for this; a personal one, and a technical one.

12 JAN 2015 was the last time SFM had gotten an update. At the time, I was just barely able to get deeper into Garry’s Mod porting as I was finishing school. I was still smashing my head against the screen, cursing blender day in and day, with only 2 hours a week to access an internet connection. In that time, tons of things in the world were changing and advancing; personally, globally, and technologically. It was the technological advancements that pushed me to transition from garry’s mod to SFM.

I spent the next 7 years learning SFM, mastering its limitations. Scouring the internet and art communities for advice, suggestions, and information, trying to learn what people had discovered during the SFM community’s peak. In time, an amazing community of artists formed around my work, full of passionate people who loved using SFM as the primary tool for their art. I found more support and love in this community than I could have ever dreamed of, and I owe that amazing experience and my prosperity to this software, its users, and its community. Due to the unique purpose and flexibility of the software at the time, an entire wave of people were exploring the joys of using SFM to explore the process of making 3d art. Experiencing this dynamic motivated me to learn the many “kung-fu” production tricks needed to push models through its restrictive specifications.

In 7 years, I went from creating models with rough topology, old Gmod flexes, HL2 skeletons and clothing models, painted-on genitals, and other issues created by limited knowledge of porting to making complex, modular monster models. The latest generation of models have maxed out the 128 controller limitless, maxed out the 32 material limit, required multiple split revisions to dance around memory limits, and use a library of production tricks and workflows to push the engine as far as it is willing to be pushed.
Due to the nature of the NSFW community and how it can motivate people, the skill levels of artists can raise quickly, along with the standards of quality. This brings more artists in, getting them interested in the hobby and expanding the community. So when I make models, I try to design them not only as tools, but also as toys to play with in a virtual world. They should be flexible and powerful enough for a professional artist while also being fun to work and learn with as a beginner.

All of this is why I defend SFM with the passion I do. It’s an amazing tool for anyone who wants to enter the world of CGI, and I try to support that with my models as best I can while scraping against the limits of the software.

All of that said……12 Jan 2015 was the last time SFM got an update.

It’s been 9 years to the day since SFM has received any patches or improvements. It’s been 20+ years since the version of Source engine that it runs on was developed. And in the 7 years I spent making 30+ character models, I’ve begun to hit limitations with the software that have caused me to cancel entire projects due to bugs, missing features, outdated code, and most of all the 32-bit limit on memory that causes the software to die when too many models are loaded.

This year, 90% of my joy this year came from making scenes and animations directly in Blender. 60% of my frustrations came from trying to plan out more and more complex models, and dealing with the gauntlet of compiling errors that result from that complexity. Back when I gave up on Gmod, I was spending 10 minutes for each compile and another 10 minutes just trying to open the model on my computer; a process that ate tremendous amounts of time when it had to be repeated. SFM solved that process at the time with its less frustrating compiling process and relatively fast ability to test models. But now, I have DOZENS of models to keep track of for every project, which has wiped out those gains in my workflow speed. At this point, the only thing I enjoy in the porting process is assigning materials.
And it’s even more depressing because now I even feel tired of just opening SFM at all. When I open a model to make a release poster, I’ll often find a bug that delays the final release even further. It’s gotten to the point where I prefer to make my posters in Blender, or have to ask friends and testers if they can provide posters to assist me.

On top of all of this, SFM has severely limited the design decisions I’m able to make while building a model. SFM imposes heavy restrictions on polycount, weight counts, textures, and materials that limit my ability to make models that look and function well in Blender. With Blender becoming a de-facto industry standard, I need to start learning how to design around its potential instead of limiting myself to a 20 year old engine.

This is a point of no return for my career. I’m 28 years old and I depend on this software and this work to sustain my existence. I need to start thinking about improving my skills and preparing myself to find work in the 3D CGI industry and take on bigger projects. And honestly, I need to start moving forward for the sake of my own mental health. I have been trying to stick with this engine for the sake of all of the people who still rely on it, but I can’t keep supporting a depreciated engine indefinitely and still feel like I have a future in this profession.

I would also suggest to those of you who have the hardware for it; try dipping your toes in the blender pool sometime. It is a very complex piece of software, but there is a tremendous amount of information, tutorials, addons, and assets out there to help you pick it up. Blender is constantly being supported and improved. Functionality is always getting updated and refined, and GPU implementation is getting better every year as the deep learning rendering tech continues to reduce render time. The software is capable of producing so many different styles and assets, and there are already many fur artists migrating and learning how to take advantage of it that can help you learn the basics.

As the community grows, and more newcomers in the future use Blender as their starting point, the community will become more and more welcoming for rookie blender users. I think giving people a clear deadline of support is the healthiest way I can make the transition at this point, and my hope is that this year will be enough time for people to prepare for that transition. In the meantime, I will continue making SFM models as I always have, while slowly dedicating time to practicing for the 2025 blender transition.


The first Blender model will be a full remake of Alilkira. She was always the “test bed” character for these kinds of transitions. Her major design characteristics have already been developed in her most recent version, so it will be a nice way to recap and compare the results between the old and new software constraints. This should help motivate me to take on more lore-focused projects in the future.

I do not have much else to add at this point; I dropped the idea of a whole Q&A section because I prefer to answer them in the comments, so if you have any questions, please do not hesitate to ask.

On a quick note, most aspects of my work not related to SFM will be kept almost identical; VRC Ports, Model edits, porting to other games, and other use cases will still be considered with the addition of now allowing other people to port the models to SFM themselves if they want.

I will look for someone to do the port process officially but it's not a guarantee yet. The focus right now will be to prepare for the transition and to keep the last few SFM releases consistent with the level of detail I always try to give.

I have no shortage of words for how much I owe to SFM, and I want to be respectful of it and the people who advocate for it. But this is my limit, and I need to move on. I encourage you to join me in this transition, and I will be happy to answer any questions in the comments that you have. As always, thank you so much comrades! and happy 2024! -WM OUT