Announcing that I have joined the founders team at Immens.dev as Chief Product Officer, with as goal to build the game engine of the future.
The week in games just did an interview with me as well - Epic Games' former Unreal Engine evangelist is making his own engine
Here is my thinking about where things are heading in the industry, and why we think creating a new engine is the right path:
1. Competition
First off, for the sake of the industry, we need competition and alternatives in the engine space. The entire industry is largely dependent on just two companies and their decisions. The one and only true alternative that has emerged in the last 15(!) or so years and that has gotten widespread traction has been Godot.
Over the last 10-15 years the standardization of engines and workflows has streamlined things, which in turn has helped provide the foundation that led to the growth of the industry and which in turn created opportunities for many, but the industry today is facing a whole new set of challenges due to saturation, player consolidation, formula driven generic experiences, team/budget sizes vs revenue, and so forth. As we have all seen the industry finds itself on shaky ground as it is trying to figure out where to go from here. While the standardization has helped the industry in the last decade, going forward I think we now need innovation to overcome the challenges the industry is seeing, rather than standardization.
The mission is to create an engine that can take a fresh approach to things, can operate at scale, for games of any size, and including all that entails to do that well. To operate an engine means to not just consider the technical aspect, but also the business side of it, the wider ecosystem, support, and so forth. That is the ambition.
2. A modern engine
I love Unreal Engine, as my track record shows, and Unity has powered many great games and teams too, but after decades of continuous development both engines are carrying ever more baggage and ever more legacy code, features, tools, workflows and so forth that they need to either maintain or support, or that they are constrained by. This is natural, and hard to avoid, but it is nonetheless complexity that spills over to the dev working on their game.
If I take a step back and reflect on things I think things have become very convoluted for both. Case in point, after 27 years of Unreal Engine, my genuine honest assessment for how much of Unreal Engine I now know is about 4% only...
So this makes me wonder, where is this heading? What would we do if we were to start fresh? What is possible in terms of things like performance or iteration times, extendibility, ease of use, and so forth? Especially if you also consider point 3 and 4 ->
3. Workflows of the future
Again I love the way we have all gotten used to of making games, I owe my career and everything I ever made to it, but here too I think we need to take a step back, ensure we don't get engrained in past perspectives, and really contemplate on where things are heading or what is possible in the future.
The way we have generally been working for decades now is that as we download a piece of software, that is typically developed by a set of people at a company somewhere, that works in a very established & set kind of ways (shortcuts, features, workflows) as designed by the developers of that software, and often saving proprietary formats that require said software. To work with it, you must learn and adapt to the way that program was made to be used. If you want something changed, you can request that change and if you are lucky a few months/years later it gets added in, or if you have the source code you could do it yourself but you will still need to go and deal with that change/recompile/maintaining your fork. This is true for almost all software, including game engines and most game dev tools.
But I don't think that model will survive the coming years. I think we are going to see a lot more bespoke tooling and software in the future. Software should adjust to you, not you to it.
I am an artist and designer and so like many of us I am often constrained by being able to execute on what it is I want to do because it relies on the tooling, or engine, or implementation of things. But we have now entered a world where we can create tools, workflows, prototype things, or connect things together, and it is going to dramatically speed up and open up things. You can create the exact tools with the exact UX, with the exact shortcuts, features and behaviour that you need for how you want to work and for your particular project. And if you can do that, so can others. It means we will go to a world where there is a lot more solutions and tools out there that people share or collaborate on, or can integrate with in turn.
The last time I felt like this was in 2013 with Blueprint, where suddenly it opened up a huge world of being able to create essentially entire games single-handedly, as many have done over the last 13 years. We are at such inflection point here again, and it is going to change how we work and what we expect out of software.
There is a ton more to say on this topic but for now, what it means for an engine: A rock solid and very efficient core engine that is highly flexible and extendible so that it can be used exactly as you require it to work. Easy to add or interchange any parts to it, seamlessly able to interact with a wide range of libraries/tools/DCCs/services/APIs, using text based or standard languages or formats wherever possible.
4. AI
What we are seeing today is a lot of agentic coding, but I think that is just the tip of the iceberg for where this can go. The possibilities are huge, and it requires an engine that is built with this in mind from day one. It requires agents to be able to fully comprehend the game and all the millions of pieces it is made up of, and how it all fits together, so that it can reason about its changes in the context of the entire game.
AI can be used in a positive and responsible way, to empower developers and allow them to iterate and experiment faster, which in turn will help creativity. We don't want prompt to game or other such low value use, but we want to find the valuable uses for it, that are compatible with how developers work. Iteration speed, tooling, prototyping, and so forth. That will help us forward as an industry, and I believe there the possibilities are near endless. It feels in a way like a return to the creative frontier of the 90s and 2000s, where workflows weren't set in stone and everything felt open for experimentation, which I find very exciting.


























































































































































