Mobius

As it should be. So I won't have to deal with shenanigans from pieces of shit like you. As mentioned above, I've been doing this for a long time.

Manual coding? Shameless...

Performance was never the goal. Lineage 2 isn't a generic 3D game. Retail movement and pathing follow geodata the way the official server does and players notice when monsters path differently from retail. A navmesh can be fast and still give the shitty results. Paths retail would never take, mobs reaching spots they shouldn't, behaviour nobody who plays retail recognises. Beating a benchmark with the wrong logic is idiotic, like yourself. So no bet. There's nothing to race. My goal is to match retail, not to destroy it. Subscriptions aren't a prize either, and access isn't given to people who spend their time bashing the project.

Prompting is easy. Knowing whether the output is right is the hard part. Your share is living proof of that. Free shares never cost me a customer. I do release my full sources, yet people still choose to subscribe instead of begging for scraps like you do, hiding behing alt accounts.

Congratulations, for another useless share. You have no idea what I was talking about. Keep wasting your tokens on useless shit.

Shame there is no pathnode button though... oh wait, there's no pathnode anything. Your overnight prompt rebuilt the rule-based approach we moved past since the last leak we had. You even shamelessly copied the idea of Doors and Water generation. All good though, no rush. You're definitely skilled enough to make it work properly. Because any bozo with AI can do miracles, right @Elfo?

L2jMobius MapExtractor After three months of dedicated development, the MapExtractor tool is finally ready! Once again, L2jMobius is leading the way, becoming the first project to develop and share such a tool with the community, with a completely FREE public release coming this Christmas! 🎄 For years, similar tools have been sold for €400–€800, often locked to a specific game client. Others have charged outrageous prices of €350–€700 for geodata for each individual protocol. We believe this is simply unacceptable. Geodata is an essential part of any Lineage 2 emulator, not a luxury feature that should cost hundreds of euros extra. It should be provided alongside the emulator itself, and frankly, every project should have adopted this approach long ago. At L2jMobius, we believe in developing real solutions, making essential tools accessible, and giving back to the community rather than exploiting its needs. Three months of work. Hundreds of euros in potential costs eliminated. Available to everyone, for free. This is what L2jMobius stands for. 💪

This didn't start in this thread. He tends to come to the L2J Mobius Discord to advertise his projects, which aren't tested against Mobius (he openly says he uses aCis), and then argues with the rules the server runs on. That's what I reacted to. I could have said it more plainly, but that's what my reply was about, and he knows it. My community is not a billboard for people who don't use my project and won't even respect the rules there like everyone else. You knew the rules, you chose to argue with them so you could advertise, and now you're here framing my reaction as unprovoked while leaving out what caused it. As I see it, I'm the one ignoring you.

L2jMobius has data oriented pathfinder since May. Already have "less memory" used for geodata since 3 weeks ago. Avoiding 1 GB less memory in geodata on latest clients. Not to mention rolling with an actual geodata generator since 3 weeks ago again. We also did actual benchmarks, tests with actual players, and decided to not use AI slop navmesh. I guess working only on a 10 year old client gives you a lot of free time and need to kill your boredom somehow... Need to show some progress somehow... Kinda get it... And unlike aCis we do release our full work every 4 months. So expect to actually see our work on this Christmas release.

Game works as it is meant to work. I trust NCsoft more than someone hyping closed source AI slop. You on the other hand, feel free to flex your 5090, running Interlude and be happy for 200 FPS.

I already have an Interlude client that can run on a potato. No bugs introduced by Elfobug.

https://www.youtube.com/watch?v=4V-jCTiknI4 Ο τζάμπας ζει.

Every commit to the L2J Mobius project since 2015 visualized as comets impacting the Mobius planet. 10+ years of development, thousands of commits, hundreds of contributors, all brought to life.

Spreading FUD disguised as "experience," mocking the project publicly, then acting shocked when called out. That's the conduct.

The ban was for the conduct, not the critique. As for the technical assessment, we'll agree to disagree. Anyone with eyes can see on his own.

And this is gentlement what I had to deal with, and still pay the price for it.

I told you, you where on my shit list a year ago. Ignoring your previous crap doesn not mean I forget. Says the guy that pops like a fart on every chance he can spill poison. Disputing that banning him was without reason. When I told the same years ago I was flamed. Having a project with actual progress, seems to piss a lot of people. Thank you.

I talk about this. https://www.youtube.com/watch?v=7Smpjz-0lEk The character creation is cool as well.

How about a complete interface rework for C4 / C6 to look like Classic? A nice retail Classic, not the existing Essence like with build in bot and automations. Add to that the login from above, or more spesifically the blue sky one, and you have what people ask for. Specially after I stoped the L2jMobius Classic Interlude.

Enthusiasm about everything functioning perfectly, until those files became public. These files taught me the difference between making things work in a private, commercial environment versus releasing your work publicly and facing intense scrutiny over even the smallest details.

To anyone reading this after 2011... Do yourself a favor and download my release of Epilogue. It is only been worked for at least 10 years after L2DC died.

What the fuck are you talking about on 2025? Did you ever played on any L2Gold server? Do you know where it was hosted and what files it run? It's particularly grating when people who haven't logged into L2 in years, or worst, someone actively attacking servers, pontificate about L2J vs L2OFF differences. Debating theoretical technical points while actual players just want stable, fun servers. I am aware of well known servers that ran successfully on L2J as far back as 2010.

Don't get me wrong, I have no hate for AI. The potential for LLMs in gaming, especially for creating dynamic NPC dialogues or complex, adaptive game masters, is immense. The point of my previous post wasn't to dismiss AI as a whole, but to question its practical application for this specific, solved problem. However, the example given (basic farming) is the worst possible use case to demonstrate this potential. It's the equivalent of using a fusion reactor to power a desk lamp. The overhead is astronomical compared to the task. The core of my argument is about efficiency and the right tool for the job. For the predictable, loop-based behavior of auto-farming, a state machine is not just adequate; it is superior. It is lightweight, incredibly fast, reliable, and consumes negligible resources. To prove it's not about the volume of code but the efficiency of execution, here is the entirety of the auto-play logic for my server: AutoPlayTaskManager 400 lines of code: AutoUseTaskManager 470 lines of code: This code provides full, retail-like auto-play support for all classes, including offline play. It runs on any standard VPS without a dedicated GPU, using a tiny fraction of CPU cycles. An LLM-based solution for this same task, even a "weaker" one, would: -Introduce significant latency (response time) for each decision. -Require expensive GPU hardware to run locally or incur API costs for cloud services. -Add immense complexity for parsing natural language responses back into game actions. -Be inherently less reliable than a simple "if mob dead -> loot" check. So, while I agree the research is "quite cool" as a proof-of-concept, championing it as a practical solution for auto-farming is where the "AI hype" label fits. The real innovation would be applying that LLM power to a problem a state machine can't easily solve, not one it already solves perfectly.

AI hype from dumb AF devs at it's best. The video shows a character doing the most predictable loop imaginable attack, loot, repeat. You don't need a "reasoning" LLM for that. A lightweight task scheduler or even a basic script can handle it with a fraction of the resources and zero GPU overhead. Spinning up a big language model to decide "hit the mob and pick up the drop" isn’t clever, it is wasteful. It's like hiring a rocket scientist to flip a light switch: impressive hardware, pointless job.