Neat! I am working on something similar and arriving at similar conclusions. eg sqlite local index. I am not ready to give up human authoring, though. How do tackle the quality gate problem and conformance? For programmatic checks like linting it’s reasonably clear but what about checks that require intelligence?
EHG is hiring! We're looking for a talented Senior Back-End Engineer
on Last Epoch. If you love ARPGs and want to work with a great team (and me!), check it out:
I've been using Claude to help write a complex prototype for game dev. Overall it's been a big productivity boost. However as the project has grown Claude has gotten much worse. I'm nearing 15k lines and it's borderline more trouble than it's worth. Even when it was helpful, it needed a _lot_ of guidance from me. Almost more helpful as a "rubber ducky" and for the fact that it kept me from deadlocking on analysis. That said, discussing problems and solutions with Claude often does keep things moving and sometimes reveals unexpected solutions.
If Claude could write the code directly unsupervised, it would go wild and produce a ton of garbage. At least if the code it writes in the browser is any indication. It's not that it's all bad, but it's like a very eager junior dev -- potentially dangerous!
Imagining a codebase that is one or two orders of magnitude larger, I think Claude would be useless. Imagining a non-expert driving the process, I think Claude would generate a very rickety proof of concept then fall over. All that said, I wish I had this tool when developing my previous game. Especially for a green field project, it feels like having access to the internet versus pulling reference manuals -- a big force multiplier.
EHG is hiring! We're looking for a talented game dev to help lead development on Last Epoch. If you love ARPGs and want to work with a great team (and me!), check it out:
I made a top-down 2D survival horror game with doors that could be opened, closed, locked, unlocked, and peeked through. Zombies could open them when unlocked. Sincerely hope to never make a game with doors that aren’t proximity activated in the future!
+1 to design documents. Would add another reason they're useful: reducing miscommunication & misunderstanding between stakeholders. In my opinion, a tech document's goal is to uncover missed specs, invalid assumptions, broad architectural errors, if we're actually prepared to build (i.e. can we speak credibly to what/how), and finally to reduce miscommunication between various teams/disciplines.
I'll add I think tech doc review is just as important as writing stuff down. In my experience it's often done poorly. Doc should be ready ahead of time, sent to relevant people, read ahead of time, and a list of open questions and comments seeded before the meeting. The meeting should have a notetaker. The meeting should be shorter rather than longer. Have fewer people rather than more. The meeting should not be about reading the document top-to-bottom, but about answering open questions and confirming everyone is on the same page.
Programmers always want to automate content creation. I get it as I’m a coder and it’s a fun problem space. Plus creating content is super time and skill intensive. It’s really hard to get right in my experience. Kind of like trying to automate writing a novel…
OK I read the article. I'm very skeptical of this approach. I doubt we can actually uncover fitness functions that reliably maps to "fun", and I believe it would require huge engineering effort to keep the game "simulable." Their examples aren't convincing. What would be convincing is a full, complex, and _fun_ game using these techniques.
Also the article seems like an ad for their AI solution.
I would love to hear more concrete recommendations or reading. It's an interesting hypothesis! That said I am a bit skeptical given the game industry has failed to sort this out for several decades... perhaps it's simply ignorance but occam's razor has me thinking there's another factor. Still, it would be interesting to hear more.
I'm in the trenches. Making games is amazing, but the business is brutal. Very hard to find a "geyser of money" and monopolize it for years let alone decades. Contrast with Google search or Microsoft Office, etc. The tech advances quickly, dating past work. One fad gives way to the next. There's a fixed amount of attention for an ever growing amount of content. etc. etc.
Meanwhile, the technical work is very sophisticated and game engineers can work elsewhere, so they justifiably want to be paid well or they will leave to adjacent industries. Plus life is expensive in America. Six figure salary ain't what it used to be. People overseas, including western Europe, will do comparable work for peanuts. Article basically gets it correct. Not sure the problem can be solved. Joe in Seattle wants $150,000. Jozef in Poland wants $50,000. Case closed.
Still, I'm grateful to get to make games with cool people every day. Carpe diem!
I mean if direct influence is the metric -- Fox has low millions according to Google. What do you think, 5 million people in a year under Murdoch's "direct" influence? Looks like Taylor Swift sold 4.35 million concert tickets in her latest tour. So who has more influence? Hard to say, in my opinion. We'd have to start unpacking the content of the information being propagated. How much of the information is novel? How receptive to new information is the audience? What is more impactful a news segment or a lyric? Seems impossible.
Or going about this another way, who was more influential: Hearst/Pulitzer or Freud? In my view, Freud flipped the world on its head meanwhile the other two are simply "yellow journalism."
Are we excluding religious founders? Even still, in no particular order and not exhaustive: Marx, Freud, Rand, Nietzsche, Skinner, Locke, Socrates, Darwin.
We're a growing company working on Last Epoch, a cool new action-RPG! We have a number of open positions including a Gameplay Engineer, Senior UI Engineer, Client Optimization Engineer, Backend Engineer, Technical Art Director, and Technical Director!
I'm also a fan of logs. If you have some more examples of how you typically log things to be most effective, I'd love to see 'em! I'm still finding my sense for when it's too much versus too little. Best way to incorporate runtime data. How to structure log messages to work well with other systems. Hearing from others and seeing battle tested examples would surely help. Or if you're down to chat a bit I can send you an email and continue the conversation. Will check out Pachyderm in the meantime~
I recently read "The evolutionary origins of modularity"[1] and it talks about NNs that are evolved based on task-performance vs task-performance plus "graph cost" and finds that the later greatly outperform the former. Graphs in the later group ended up being more "modular" which allowed them to evolve to meet changing requirements more readily.
Could we analyze computer programs and create graphs representing modules? And links between systems? A node would a method. An edge would be a call site. Graph-modules – clusters of highly interconnected nodes – would represent the real fault lines of the application, which could be juxtaposed against the developer’s conceptual model. The cost of the overall graph could be calculated and used as a metric – aiming for some lower bound.
Inflation: according to several calculators online, $100 in 2003 is roughly $170 (rounding up) in 2023 dollars. So 70% over 20 years. Meanwhile Case-Shiller has grown 123% over the same period.
Interest rate: in 2003 the interest rate was in the high 5's. Today it's in the mid 6's. So let's call it a wash.[1]
What I'm saying is, today someone is buying the same house, at roughly the same interest rate as someone twenty years ago except that today 1) the house is twenty years older, 2) it's also about 30% more expensive after adjusting for inflation.
I grant that means the person needs fewer than 2x the resources, but my point still stands.
Twenty years ago the Case-Shiller index was 140. Today it's 312.[1] The index tracks individual home sales over time -- that is a single home sold repeatedly. I interpret that to mean someone purchasing a home today must have roughly twice the resources to purchase the same home as someone in the past. Which leads me to conclude it's harder to own a home today than in the past all else being equal.