mid_gen's Launch a Game Studio Thread

Soldato
Joined
20 Dec 2004
Posts
17,490
So, people reading my posts on here for a while might have seen I'd been considering this for a while, waffle incoming :

I've been a professional game developer at AAA studio the past 12 years or so, with another 10 years software engineering and business analysis experience prior to that. I have been Lead AI Programmer for most of that time in games, and started on the Settlers franchise at Blue Byte. Fun fact, of the 5 game projects I have worked on....one was an already live service game, one launched (Avatar : Frontiers Of Pandora), one was cancelled recently (RIP Everwild), one is very likely going to be cancelled imminently (Unannounced), and another is shelved indefinitely. One launch credit in 12 years, FML.

Anyway, people probably know the game jobs market is a bit of a disaster at the moment. There are jobs out there (particularly in AI), but I own a house and my partner likes her job, so I'm very reluctant to relocate in this environment where any game studio could go bust at short notice as investors are pulling investment left right and center.

So, I've had this idea in mind that I'd like to start my own studio some day, god knows I've learnt enough about how not to run game projects, right? For the last 9 months or so I've been working full time for a fintech startup run by a former boss of mine from my days working in investment banking. I'm building agentic AI systems which is pretty interesting and I'm sure I could find work doing that, but I sank a *lot* of time and effort in switching careers to games, and as much as AAA dev has been pretty demoralising at times, I still love making games and want to do it in future.

So, I've decided to take the plunge and get my own studio started. I registered my limited company and got that set up. From the beginning of 2026 I'll continue providing services for this startup, but on a contract basis, with the money mostly staying in the business to build myself a runway for the studio.

I've built a good network over the years and have formed a kinda cooperative group with some other ex-AAA developers, the core members being me (programmer) plus an animator and an environment artist. We've all got our own game ideas, so the idea is to collaborate, and do bits of work for each other's studios, rather than try to pool together as one, which I think is a pretty sound model.

Ultimately, I want to go full time making my own titles, but with this fintech contract work paying reasonably well, I think I need to make hay while the sun shines and build up as big a runway as I can. I'll have to be pragmatic about things as I can't really afford to blow through all my savings when retirement is looming ever closer in the distance. I've had quite a few approaches already for games industry contract work, so may end up doing bits of work for other developers here and there to keep the coffers filled.

I have done some work at one of the main games degree courses earlier in the year, so I've toyed with the idea of trying to get a title launched while I'm still full time on other stuff, doing the programming in my spare time, and hiring some of the talented graduates to do the art side of things.

Why the thread? Well a big part of launching a successful game is building an audience....I want to try being quite open about the process, I already run a hobby Youtube channel that's pretty popular, and I'd plan to do the same with the studio, but also involve the OcUK community in the development efforts for feedback and bouncing ideas off. I'm a bit away from finalising what game I want to start making just yet...but there's plenty to get on with.

Already Done :
I've got my company set up, and business bank account

In Progress:

I'm in the process of building my tech pipelines. This is a pretty big chunk of work that I need to get sorted and critical to get right....particularly if I want to get investment in future. I need to make sure I've not only got a solid efficient pipeline for myself, but one that will also scale up as and when I bring other people onto the project. Some things I'm already settled on :

I'm going to self-host rather than start incurring monthly cloud costs. I do have a lot of exposure to AWS infrastructure in my current job so I'm confident in migrating workloads up to cloud infrastructure if I need to.

Engine - Unreal, I know it well, lots of artists and other devs out there will be familiar. (don't worry I know how to address the hitching ;))

Source Control - I'm evaluating a few options at the moment.
  • Perforce is industry standard, but it gets pretty expensive as you scale. It is a requirement for some Epic tools like UGS/Horde. Free up to 5 users then $35/user/month after
  • I'm currently evaluating Ark VCS which is a new system being developed. Very attractive licensing terms, but early days.
  • Subversion is actually a pretty solid solution for games, good value licensing options. Also evaluating at the moment.
  • Git just isn't viable for game development
I'm leaning towards Ark as the developer is really active and engaged and I'd rather support another small dev than funnel money to a big corp.

CI - I'm using TeamCity. I have a good working setup at the moment....happy with this, free for the scale I will need it for. Used for automated builds and testing.

Documentation - WikiJS. I'm running this in a docker container on my home server at the moment. Can easily be punted up to Amazon ECS if I need to in future.

Secure Remote Access - The ability to give external developers controlled access to parts of my self-host infrastructure....I'm using Tailscale for this....what an absolutely bloody brilliant bit of software. Love it.

So, whichever decision I go with for source control, I'm still going to be operating at £0/month up to 5 users. If I go with Ark then I'm still paying nothing for source control up to $500k revenue. Main cost will be Tailscale at $6 a month per user as I scale.

That's the core of the pipeline for now. I will need to look into Xbox/Playstation devkits and the like in future, but that's a future me problem.

I have a windows server running this infrastructure right now.

Up Next :
  • Create a reliable backup routine to ensure all my home server data is backed up to my NAS and I have a disaster recovery process that works.
  • Then create a nightly off-site backup, probably Backblaze, from my initial research.
  • Setup my accounting software and make sure I'm all ready to start invoicing in January
  • The big one......what do I call my studio? I've been batting around ideas for weeks and haven't really settled on anything yet. The limited company is named, but that was never going to be the game studio trading name. Ideas?
  • Over the Christmas break start documenting all the prototype ideas I've had floating around and putting some thought into what I want to start with
I really need to get the studio name sorted so I can buy the domain and pick up the social channels etc. I do want to do a socials-first documentary type coverage as I build the studio to help grow the audience, will probably stream once I start development on the title.

Told you it was a lot of waffle :P Subscribe for updates! Studio name ideas welcome. Pitch me your game ideas too, why not :cool:
 
Watching this thread, as an SRE with an infrastructure fetish this is relevant to my interests.

Out of interest why is Git not suitable for game development?
 
Watching this thread, as an SRE with an infrastructure fetish this is relevant to my interests.
Cool stuff. One thing I haven't mentioned yet is the use of AI tools. That ones a bit of a hot potato I think I'll make a separate thread about somewhere!
Out of interest why is Git not suitable for game development?
Game development typically involves a very large amount of binary assets that cannot be merged, so typical game dev workflows revolve around locking or checking out assets. Perforce, SVN etc. have strong Unreal integrations so when you're in the Unreal editor and you need to edit something, you check it out, and everyone else working on the project sees that you have it checked out and knows not to (or physically cannot) edit it at the same time.

Game dev typically operates 'trunk/main based development' where everyone always works on the same branch through most of production, moving to release branches later on to manage stability if needed.

This is kinda antithetical to the whole workflow of git and it's branching methodology. There's no native support for locking assets with Git as your source provider in Unreal, so you have to maintain lists of "who has edited which binary asset" in spreadsheets or other such horrible workarounds.

Game repositories also get absolutely massive after a while, particularly if art sources are stored in it (a lot of modern art workflows work revolve around doing extremely high res sculpts, and exporting lower-res into the engine depending on target platforms and other variables). Git just isn't designed for this sort of scale....it's wonderful for handling pure text assets that can always be merged, it just falls apart when lots of binary files are involved.

Some people have migrated to hybrid workflows where Git manages the source code, and binary assets are stored in Perforce....which can work but involves a whole lot of overheard managing how you synchronise code and data.
 
I'm wondering what sort of size of game you're thinking. Something you can make in a year, or bigger?
Small as possible the first one really, there'll be a lot to learn with just getting a launch out on Steam (especially if I self publish). Let alone if I try to launch on consoles.

I've got some quite ambitious ideas in my head, but they are going on the back burner. Priority will be something small, focused, that leans into my strengths as an AI programmer, and can realistically be delivered in less than two years.
 
Studio name suggestion:

Opera of Forces

Opera being Latin for "work"
 
Last edited:
This is fantastic, I'll be following this thread with interest.
 
Git just isn't viable for game development
Agreed, I'm a web developer and Git is more suited for web development rather than games. Website and web app repositories are generally smaller too, so it makes sense.

Fantastic stuff, I'm looking forward to following your journey :)
 
Best of luck this sounds awesome.

Maybe you could tie in your hobby into the naming, Rock face Productions, Free solo Studios. Abyss Games. As you can probably tell I know nothing about rock climbing!
 
This is fantastic, I'll be following this thread with interest.
Agreed, I'm a web developer and Git is more suited for web development rather than games. Website and web app repositories are generally smaller too, so it makes sense.

Fantastic stuff, I'm looking forward to following your journey :)
I believe that this is how Rebellion got started.

The very best of luck to you.
Cheers!

Managed to tick off a few things today, got my accounting software sorted out. Was in Nottingham earlier and also went and checked out a co-working space there (Arc Spaces) which looks decent. Only £15 a day or £125 for a monthly pass....I live quite out in the sticks and it does my head in being at home all the time....would be nice to get out and network a bit and have a change of scenery.

Looking at backup stuff now, looks like FreeFileSync is going to be a good best for getting my self-host stuff onto my NAS.

I need to have another brainstorming session for the studio name this evening....! Tis a big one, so many considerations.....domain name availability, trademarks, logo, the overall 'vibe' of the studio/name/logo, merch opportunities.
 
Was in Nottingham earlier and also went and checked out a co-working space there (Arc Spaces) which looks decent. Only £15 a day or £125 for a monthly pass....I live quite out in the sticks and it does my head in being at home all the time....would be nice to get out and network a bit and have a change of scenery.
I feel you! I also like separating work from home if possible.

@roccles suggestion of your rock climbing hobby sounds like a cool name idea.

Fingerlock Labs?
 
I feel you! I also like separating work from home if possible.

@roccles suggestion of your rock climbing hobby sounds like a cool name idea.

Fingerlock Labs?
Hah, not sure I want 'finger' in the name :P

My main YouTube channel is rock climbing stuff so already got my brand for that (Pulp Friction Climbing)....and the domain, but I don't think I'd ever make a living off it!

Was faffing around with FreeFileSync for a bit this evening then got fed up with it and just used Claude to write a powershell script to run robocopy and keep a second rolling copy of the previous, and got task scheduler running that nightly.

One more job done, just trying to chip away at things whenever I have time. Maybe in a few weeks I can actually get started on a game idea :)
 
No idea what you're talking about with all the technical stuff, but good luck.

As for game ideas though, I still hanker after a follow up to From Dust, a quirky little god game puzzler, which to the layman seems to be the type of thing a smaller studio could take on.

 
Last edited:
No idea what you're talking about with all the technical stuff, but good luck.

As for game ideas though, I still hanker after a follow up to From Dust, a quirky little god game puzzler, which to the layman seems to be the type of thing a smaller studio could take on.

Not a bad shout, I did play it back in the day and have it in my magic Ubisoft account.

Those are the types of games I’m likely to make. My days of grinding out long sessions in twitch shooters are over :P

I’m likely to build something quite systemic and replayable, AI focused. I’ve got a few ideas I’ll share soon.

Quite interested in some of the content tools around, like the Rokoko motion capture setup. Quite appealing indie pricing to be able to produce all my own animations. So many possibilities!
 
Back
Top Bottom