How far are we from AI recreating EQ99?

[BradZax]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

"Run in circles."

[You must be logged in to view images. Log in or Register.]

"You have 10 seconds to comply."

[BradZax]

That, and most of them are all in jail IRL.

[OriginalContentGuy]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

That, and most of them are all in jail IRL.

Yeah but with you that could mean they're crashing on your couch.

Prisons are for burning good night.

[Baugi]

Originally Posted by Dulu [You must be logged in to view images. Log in or Register.]

So the question is, how far are we from making a game like P99 in AI?

If we accept the dubious premise that LLMs can spit out a doom-sized game now with minimal intervention(human intervention cost probably scales more like exponential with project size) and oversimplify a lot -

Compute cost as a function of code-base size scales between O(n^2) and O(n), with traditional attention vs. various linearity hacks/Chinchilla

Doom was ~50k LOC

Everquest through Velious probably had between 500k and 1m LOC (EQMacEMU's server has ~2m for example)

Compute scales at something like 2x per 2-3 years according to Gelsinger vs. Huang's updating of Moore's law.

So maybe log2(500k/50k)*2 to 2*log2(1m/50k)*3, or ~7 to 26 years for the compute power.

You will almost certainly run out of training data long before that. We're basically out now.

I hope that helps. I need to get back to debugging the COBOL Cursor keeps inserting into my secure networking code.

Edit: for an interesting albeit bad comparison, Anthropic claims their C-compiler experiment was basically autonomous(still required expert developer designed test-cases and a CI pipeline that would be impossible for a game) and ran in 2 weeks for $20k, producing 100k LOC. It sometimes works.

[BradZax]

"ok lets start with step one, lets install godot and unity how do I do that idk what that is."

and just keep doing every step like that and in 2 years youll have eq [You must be logged in to view images. Log in or Register.]

[loramin]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

"ok lets start with step one, lets install godot and unity how do I do that idk what that is."

and just keep doing every step like that and in 2 years youll have eq [You must be logged in to view images. Log in or Register.]

Nope.

Building software with AI is like building a house: it can absolutely tell you how to hammer nails in, install electrical outlets, and so on, because it's good at construction.

But what AI won't do is tell you not to hammer a nail into a wall with an electrical outlet behind it (or not to run that electrical line there in the first place) ... because you never specifically asked it about that (and neither you nor the AI is an architect).

Now, if you asked Claude to come up with a course on becoming a (software) architect, it would probably do a pretty good job by stealing from a lot of existing online courses. Follow it for a couple years and you might become a programmer, and then you could properly use it to build software ...

... although you likely won't be that much fater than Nilbog ... who started building just the server way back in 2008 ... not the client, and even the server isn't from scratch (he leveraged the work of other devs who built EQEmu) ... and he still hasn't finished in 2026.

But keep believing that you and ChatGPT are going to make the next EQ tomorrow without a software background [You must be logged in to view images. Log in or Register.]

[Kich867]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

"ok lets start with step one, lets install godot and unity how do I do that idk what that is."

and just keep doing every step like that and in 2 years youll have eq [You must be logged in to view images. Log in or Register.]

Even if you did it every day for 2 years you would not have EQ home dawg. Not even close.

Teams of people who actually knew what they were doing wouldn't build EQ in 2 years.

Though I suppose, again, it depends on what your definition of "done" is. A contentless pile of shit with bad netcode and a billion easily exploitable problems? You can probably do that in 2 years yeah.

[BradZax]

I wasn't talking to you, loserman.

[loramin]

Originally Posted by BradZax [You must be logged in to view images. Log in or Register.]

I wasn't talking to you, loserman.

Perhaps you were unaware, but you're talking in a public forum.

[BradZax]

Originally Posted by loramin [You must be logged in to view images. Log in or Register.]

Perhaps you were unaware, but you're talking in a public forum.

Glad to know we agree that your name is a loserman.