Jump to content

Recommended Posts

Posted

Hello, im working years on Freya and C4 chronicles but yesterday i got aCis and before i even try do any imple on it or waste my time, i want ask people who already used it and the owner

whats the main problem of acis? 

don't tell me "customs" i don't give a fuck about customs, i hate "projects" that are not 100% retail and they have customs, i want take clean pack and add any custom i want even event engine.

one of the things i saw so far is that is a bit customized than the l2j i used to, i mean even the fact that it has no voicecommandhandlers. sure i already added the proper code in handler SayAll.java 

if (x.startwith".") bla bla and i added the voicecommands back but it was kinda weird to me to see no voice command handlers :))

But what would keep someone back to use acis?

Skills?Quests?

Posted (edited)

Well, for Interlude aCis is the only option I think. Havent touched new revs(already waiting for customer status), but now using free 340rev for open beta and it works like charm. 0 crashes, 0 errors, everything is working fine.

Skills works like it should, people tested it, almost 0 complaints. Buffs, debuffs, stun, stabs, bluffs and so on.. (few skills bugged, but its 340 problems I think).
I'm not very experienced, but I would recommend aCis for anyone, clean and stable project, with good community.

Edited by djterixx
Posted (edited)

15 missing quests, and few importants GBs (frintezza, zaken). Those will be added after L2OFF SpawnManager implementation. Both bosses scripts have been adapted since years, still, I wait SpawnManager.

 

Otherwise, there is less and less points about using anything else than aCis. That was maybe true 2y ago, when I decided to rework every single script (and so jython and all quests have been dropped), but the work is almost done. There is no stability issue at the time I'm speaking (at least for L2OvC, which holded something like 350 unique players, 500+ overall).

 

There are issues though, but that's true for any pack existing so far (even "professionnal" ones).

 

Voicedcommandhandler can be added in 2 different, more classical approaches :

- usercommand implementation

- NPC implementation (was the case for wedding system cleanup)

 

That's why it has been dropped.

 

Quests are added from time to time, by series of 2-3, while skills are already workbenched as the next big rework after SpawnManager implementation (skills already parsed, and a core engine already exists on private workbench).

 

The last cycle (350-359) fixed a lot of stability issues (notably the annoying stackoverflow plaguing movement since rev 238) and the focus was to decrease Bug report section reports, rather than optimize/cleanup (even if in this point, we did a lot too).

 

The incoming cycle will be about SpawnManager implementation, notably fixing 7s seal drops (retail is supposed to use 4 different spawnlist, L2J uses only one ; we parsed L2OFF spawns in a specific 7s timer, so we got only a single spawnlist, and a bad one ; even knowing that, I won't edit it until SpawnManager is out). The idea is to edit all hardcoded spawns : 4sep, castles initial spawns, events,... All will be controlled by territories.

 

You can also follow xxdem idea than basic gameplay from player PoV won't change from one pack to another, and use any random L2J IL pack.

 

I will let users say bad or good things. ;p

 

PS : I work by steps, I generally don't retouch to something until enough reports have been found. If a rework is in workbench, I also don't care about reports. Big reworks actually will be SpawnManager, skills, and items. All those workbenches are already under developement, in different % steps.

Edited by Tryskell
Posted

My own server is based acis latest but we have work a lot of this quests ai's grand bosses...im missing only quest 505 and its hard to find it :P a lot of work can be with acis a perfect pack..im trusted acis developers and they are give help with a lot of questions...there is someone to answer...

Posted

L2AEPvP still uses the ancient < 80 rev of aCis, and its going great. Heavily reworked ofc, most of the engines are custom and others are dropped since its custom gameplay

Posted (edited)

L2AEPvP still uses the ancient < 80 rev of aCis, and its going great. Heavily reworked ofc, most of the engines are custom and others are dropped since its custom gameplay

Well afc when you make something you put a huge effort to rework it. Take ex L2j (cleary the best project cause everything is base on it)

the fact that aCis is so far the best interlude doesnt mean l2j is useless. They saved tryskell lots of time to translate packets and create the whole code structure

still l2j without rework is like Piece of shit :D

But yea i see acis as only option.

Edited by AccessDenied
Posted

15 missing quests, and few importants GBs (frintezza, zaken). Those will be added after L2OFF SpawnManager implementation. Both bosses scripts have been adapted since years, still, I wait SpawnManager.

 

Otherwise, there is less and less points about using anything else than aCis. That was maybe true 2y ago, when I decided to rework every single script (and so jython and all quests have been dropped), but the work is almost done. There is no stability issue at the time I'm speaking (at least for L2OvC, which holded something like 350 unique players, 500+ overall).

 

There are issues though, but that's true for any pack existing so far (even "professionnal" ones).

 

Voicedcommandhandler can be added in 2 different, more classical approaches :

- usercommand implementation

- NPC implementation (was the case for wedding system cleanup)

 

That's why it has been dropped.

 

Quests are added from time to time, by series of 2-3, while skills are already workbenched as the next big rework after SpawnManager implementation (skills already parsed, and a core engine already exists on private workbench).

 

The last cycle (350-359) fixed a lot of stability issues (notably the annoying stackoverflow plaguing movement since rev 238) and the focus was to decrease Bug report section reports, rather than optimize/cleanup (even if in this point, we did a lot too).

 

The incoming cycle will be about SpawnManager implementation, notably fixing 7s seal drops (retail is supposed to use 4 different spawnlist, L2J uses only one ; we parsed L2OFF spawns in a specific 7s timer, so we got only a single spawnlist, and a bad one ; even knowing that, I won't edit it until SpawnManager is out). The idea is to edit all hardcoded spawns : 4sep, castles initial spawns, events,... All will be controlled by territories.

 

You can also follow xxdem idea than basic gameplay from player PoV won't change from one pack to another, and use any random L2J IL pack.

 

I will let users say bad or good things. ;p

 

PS : I work by steps, I generally don't retouch to something until enough reports have been found. If a rework is in workbench, I also don't care about reports. Big reworks actually will be SpawnManager, skills, and items. All those workbenches are already under developement, in different % steps.

Well i find VC useful somehow even tho the handler is overrided in any npc so i re-added it but so far congrats for your pack. 

i just wish u had it less "customized" cause i love to rework on retail things but i see u pretty much touched everything even 

announcements xD

Posted (edited)

Well i find VC useful somehow even tho the handler is overrided in any npc so i re-added it but so far congrats for your pack. 

i just wish u had it less "customized" cause i love to rework on retail things but i see u pretty much touched everything even 

announcements xD

 

aCis is less customized than L2J itself (far lesser configs), so I don't get your point... If by custom you mean edited, than you should know I don't edit things if it doesn't need an edit.

 

My point about customs is, the time you use adding or fixing custom is time you could use to fix retail. And retail behavior needs shitload amount of time. So I don't busy about customs.

 

I reworked aCis to be more stable, less exploitable, easier to read (comments, javadoc, writting style), and generate easy to use and reusable content.

 

And yes, I edited almost everything, from Duel (rework from zero), to CW, to partymatching, to Announcements, to AI systems, to buylist, to multisell,... Reworked, verified and tested 380 scripts. Added a new geoengine, with its own format (L2D). Whatever can say ppl about aCis, I and my community do what few people did.

 

I should compile added/removed/edited lines over the last 5y, that should be funny.

 

Anyway, use whatever you want, and you can see latest aCis on L2OvC (which I use as internal test server for experimental things, due to the decent population).

Edited by Tryskell
Posted

aCis is less customized than L2J itself (far lesser configs), so I don't get your point... If by custom you mean edited, than you should know I don't edit things if it doesn't need an edit.

 

My point about customs is, the time you use adding or fixing custom is time you could use to fix retail. And retail behavior needs shitload amount of time. So I don't busy about customs.

 

I reworked aCis to be more stable, less exploitable, easier to read (comments, javadoc, writting style), and generate easy to use and reusable content.

 

And yes, I edited almost everything, from Duel (rework from zero), to CW, to partymatching, to Announcements, to AI systems, to buylist, to multisell,... Reworked, verified and tested 380 scripts. Added a new geoengine, with its own format (L2D). Whatever can say ppl about aCis, I and my community do what few people did.

 

I should compile added/removed/edited lines over the last 5y, that should be funny.

 

Anyway, use whatever you want, and you can see latest aCis on L2OvC (which I use as internal test server for experimental things, due to the decent population).

well congrats then )

Posted

well congrats then )

 

It's a community work, and there is a lot to do, even after 5y of fixes and edit. I think the biggest problems would be solved in around 30-50 revs, if I successfully follow the idea one big rework / cycle. Once SpawnManager, skills and items are reparsed/fixed, and scripts ended, then the leftover is just bonus.

Posted

Have fun with skills.

 

If you do it properly that's targeting, effects, conditions, cast and the whole stats system to be done.

Posted

i find that the bug with stutter stepping still occurs with geodata but not as often as before

 

stutter stepping ? If you mean on target approach during skill casting then it got nothing related to geoengine, and the problem and fix are already known.

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



  • Posts

    • I suggest you provide Test server for public and add a price.
    • Kroma — Lineage II Interlude on LU4 Client (Latest client)   I'm selling this project. Here's what it actually is and what's been done.   This is a few months of work, most of it on the reverse engineering side. I built it because I wanted to open my own Interlude server on it. That's not happening any more — I've moved on to other projects and don't have the free time left to take it the rest of the way, so I'd rather it went to someone who will.   The idea was to take the modern LU4 client and make it run an Interlude version of the game. Not a reskin of an old client — the actual current client, playing Interlude, against a server I wrote. That's what this is, and it works end to end: the client boots, authenticates, connects, and you play. There are six repositories. Roughly, it splits into three kinds of work: reverse engineering the client, writing the server, and rebuilding the client's own content so it matches Interlude.   Fully reverse engineered   The client's protocol wasn't documented anywhere, so it was derived from scratch off real traffic. I built a sniffer (bypassing anticheat packet encryption) that drops into the client's binaries folder and dumps live traffic to disk, then a separate Java toolchain that decrypts those captures — login and game traffic, both directions. Everything the server knows about the protocol came out of that loop: capture, decrypt, read, implement, test against the real client.   On top of that there's a native x64 hook and injector in C++. It points the client at your own servers and handles its authentication and session layer, plus a few client-side stability fixes that were needed to keep it from falling over once it's talking to something that isn't the official backend. The hooking is pattern-based rather than address-based, so a client patch doesn't automatically break it.   The reverse-engineering notes are in the repos — the patterns, the disassembly, the decompiled functions, what each one does and why it's hooked. That documentation is a real part of what's being sold; it's the difference between inheriting a working system and inheriting a black box.   The server   Interlude ruleset, aCis lineage, but substantially rewritten and extended to speak the LU5 client's protocol with modern features like Party matching, Offline shops or Mailbox. The geodata has been updated according to some layout changes in LU4 cities. There's also a separate login server handling the client's auth flow.   Customized client to adapt to Interlude   This is the part people underestimate. The retail client ships retail content — retail NPCs, retail items, retail zones, retail UI. All of that had to be reshaped. The whole client-side pipeline is scripted: extract the retail PAKs, apply every edit from data files, rebuild, drop the result back into the game folder. 23 purpose-built C# tools, ~6,700 lines, organised into three pipelines that are guaranteed not to write over each other. There's a documented checklist for regenerating everything from scratch when a new patch drops. The important thing is that none of it is hand-editing. Every change is a line in a data file. When the client updates, you re-run the scripts instead of redoing the work from memory.   All NPC / item / drop lists updated according to Interlude   The client's own data tables were rewritten from the real Interlude client data files — `npcname-e`, `itemname-e`, `skillname-e`, `systemmsg-e`, `npcgrp`. That means roughly 6,470 NPC and skill name/title updates and about 4,000 Drops/Spoils rows written directly into the client's `NPCDataTable` and `SkillTable`, so the client shows Interlude names, Interlude item text, Interlude system messages, and Interlude drop and spoil lists. There are also custom NPC rows on top of that, plus 304 game action rewrites.   Opened the catacombs   In the LU4 client the catacombs and necropolises are unusable — flooded, unlit, and with holes in the geometry. All three were fixed: Water removed. Water exclusion volumes were placed over all 12 of them: Rifts, Apostate, Patriots, Devotion, Dark Omens, Heretics, Pilgrims, Saints, Worship, Martyr, Forbidden Path and Witch. This was more involved than it sounds — the volumes' scale is ignored at runtime, so the tool picks the best-fitting native template per hole, and nearby lakes and rivers are separate water bodies that have to be excluded individually or you get leftover water borders. Lights added. About 28,900 light definitions imported into the client's level files. Missing walls fixed. Mesh actors replaced and spawned to close the gaps in the geometry.   Removed the modern "slop" features from the UI   The LU4 client is full of things that have no business in an Interlude server, and most of them are wired into the UI. Those are stripped out: The assistant manager — the auto-hunt / auto-play slot on the XP bar Vitality*and the XP lock button The fatigue system — updated the format to a regular VIP format The in-game browser and in-game radio buttons The "Learn Skill" tab in the player skills window Removals are scripted by path, so they survive a client patch and are trivially reversible if a buyer wants any of them back.   Reworked the UI toward an actual Lineage II feel   The original LU4 UI looks AI-generated. It doesn't feel like Lineage II. A good part of it has been reworked toward the classic look — the menu button row rebuilt as a wrapping layout with a proper background, the self status and target windows cleaned up, and 590 asset field edits covering textures, positions, colours, text and compiled Blueprint literals.   This part is not finished, and I want to be straight about that But the tooling to continue it is all there, and it's the good kind of tooling: you can target any widget property in any UI asset by dotted path, remove widgets by path without knowing their auto-generated slot names, and edit text that's dynamically bound through Blueprint bytecode rather than stored as a plain property. Continuing the UI work means adding lines to a data file, not reverse engineering anything new.   Other client-side work   A custom login scene, built in a real Unreal Engine project that's included Interlude music restored, plus ambient sound work (cities only for now) Custom splash and branding   What's included   All six git repositories, the Unreal Engine project and its custom assets, all the tooling — injector, hook, sniffer, decryptor, asset editor — every data set including geodata, XML, HTML and the SQL schema, the build and repack scripts, and the per-repo technical documentation including the reverse-engineering notes.   Happy to answer questions or show it running.   Gallery
  • Topics

×
×
  • Create New...

Important Information

This community uses essential cookies to function properly. Non-essential cookies and third-party services are used only with your consent. Read our Privacy Policy and We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue..