Tryskell

[fr]C'est l'excuse bidon ça, depuis le temps tu devrais savoir - si tu suis un peu le forum - que tout passe par un re-travail complet. A l'époque il y a eu les skills, après les items, maintenant les quêtes. Ca suit un cheminement parfaitement logique. Maintenant dis la même chose de Frozen, j'attends de voir ça :P. Frozen n'a pas de "#" à ses scripts, mais vla la pelletée de bugs que ce pack traîne dans son sillage. Et oui, 30% *seulement* des scripts sont actuellement réécrits. Mais ils ont le mérite d'être autant retail que possible. Je peux te citer une dizaine de quêtes foireuses, une autre dizaine de quêtes purement customs, les AIs incomplètes (notamment Baium) et j'en passe. Le syndrome L2J, "j'utilise beaucoup de trucs cassés plutôt que peu de trucs qui marchent correctement." Chacun voit midi à sa porte :P.[/fr] And sorry if you're not french, so far you should :P.

And when my quests and AI work, they work better (btw 120 quests/Ais atm) ;p. Not the subject anyway, I told you what it was.

If you want to learn java for l2j, you have to explore and see how are build files. Understand the structure of the software, the methods composing it, and you made 70% of the work. The 30% left are basically, programmation logic and java syntax. You can learn it everywhere, from any programmation book. For java syntax, well... If you're a little clever, you need only eclipse search tool and asking to yourself that mighty question : "where can I find an existing part of code to make my own code". That resolves 80% of issues.

Your DP is fucked, it's not stackable items.

setState refers to one of the possible enum states, aka State.CREATED, State.STARTED && State.COMPLETED. Following code is wrong : questItemIds = new int[] {_olymp3}; questItemIds = new int[] {_olymp5}; questItemIds = new int[] {_olymp10}; questItemIds = new int[] {_olympChest}; , must be written : questItemIds = new int[] {_olymp3, _olymp5, _olymp10, _olympChest}; Finally, if you have only one NPC in that script, the check about npcId are useless. The only use of that check is to differentiate behavior according of the NPC. Exemple : you got 1 NPC to start and speak, and 4 mobs registered on onKill. There's no need for npcId check on onTalk and onAdvEvent, but there's (perhaps, not even sure, if all mobs got same behavior it's pointless) surely a check on npcId. I say that, I say nothing :P.

dpbBryan, from the moment you know how to copy-paste, you can pretend to the title of "developer" on MxC. AbstractEnchantPacket.java for the answer, if I understood well. Edit getChance() method.

Rework AI.

I use Premium 9.1.8 cracked.

As I said a little higher, improving SQL crap it's like another project. On IL, there is enough work just speaking of undone features and core-side code lacks. If you begin to speak of HTMs (heritated from C3 ?) and SQL improvement... Well, you need more than one life for sure. Added to that, it needs many experiments, and to be honest I got no clue how to "profiling" a database performance (nor simple java core, btw if you know send links etc I'm interested...) :D.

Creates a skillhandler with a new SkillType ? If skillhandlers are restricted too, well you have to register a variable in L2Skill (let's say a boolean _isInstantActionSkill, with isInstantActionSkill() method and too a check on the "db" reading about it). Then on the check on the doCast or wherever it's made simply apply your new method. I'm lazy to explain all... Btw your ask is still kinda "vague". What sort of debuff do you want to cancel ? Any ? A certain type ?

Nowai, base stats influence many others factors. Weak CON means lowest chance to resist STUN/etc skills for exemple. Added to that in for exemple in case of destros, many ways add HPs to the NPC : template HPs got a big impact, but you mustn't forget passive boost HP skill too. One base stat doesn't control only one stat. Passive skill is a good option, but your current quote is really focked up (mul accuracy by 18, mul reflectDam by 0 and so on).

Merge of 3 existing AIs. What more to say, except you could do it even better. Ty for share, that avoid some rework at least, lol.

Dunno if you know or don't know it, but there's such "garbage clean" at each server restart. It deletes all useless stuff. It's made on IdFactory, and run at server start (and you can eventually link it to an admincommand if you feel like an urge to use it...). protected IdFactory() { setAllCharacterOffline(); cleanUpDB(); cleanUpTimeStamps(); } Point is, you normally can't have clones.

Zuuuummmm, I was on PK = player killer. A FK doesn't improve performance, as no index is created. That's good for DB integrity, adding constraint, but if you're not totally dumb, you don't need it (such as XML validator L2J currently uses on postIL chronicles). I'm not such a pro on SQL, so correct me if I'm wrong.

Explain better what you want. Personally I didn't understand a shit about. Give eventually a simple example.

Roflol yellow title No way :P. I got the link now (and you search it through darkness :P), but still that's not a good solution. The only good solution is to add an id column with a unique number for each register. But what's the point :P. Rev 202 is up Rev 203 overview Beast Farm is under correction (near fully debugged). Added with some quests, it will be rev 203. Topzone link As crappy fans told before me, the CTS (Crappy Test Server) is up. The other new is a Topzone link exists. Makes the promotion of the pack and test directly ingame to see moves :). Vote for it on Topzone ^^

Use JDK 1.6, as dpbBryan told. Or refactor Random to use Rnd instead.

What's your remark is for ? Spawns at death is random, and there's random chaotic respawn points aswell. I don't find the link between "PK" and "NPC" in your sentences :P. You expressed 2 different ideas, or one of your sentence got a wrong word :D. Regards.

Well I'm not a pro and never touched to that part, but I guess the bmp *isn't* a bmp anymore, but a flow of data (well an image is a flow of data...). On aCis as on L2J latest version (nothing different from the 2 packs), there is nothing related to graphic library aswell, so answer is all found... You must take it as a data type. That's only a guess answer (I hunt posts in fact), and you never read my answer if it's wrong (if it was right, well you read it ofc). Btw, I guess some *administration tools* (account managing, and such) surely found the solution in the past. Regards, Tk.

1. Drop the line concerning football crap in your MasterHandler (you had it in the past, but not anymore) OR add the missing admincommandhandler. 2. In AdminAdmin, do a ctrl+shift+O to correct imports OR drop the complete reload system for FakePCs if you haven't this system anymore. PS : saying you got errors is not enough. I gave you both possibilities, the good one I can't know which is it, as you say NOTHING about your system, what you wanted to do, etc.

or all, depends of PoV. Project is project, there isn't "stable" version. We move on, each commit corrects bugs. If you miss 10 revisions, you got between 20 and 40 bugs and/or less content for sure compared to latest version (average quests commited per commit is 3, you calculate). Use latest revision each time. The only bug I introduced those days are CH doors. Previous bug was inventory refactor, 3 months ago and knownlist refactor, 5-6 months ago.

You're right, that's only during server loading. And you can find on inet 100+ tips about how to make mySQL faster :P. That's a project itself...

IRC has been deserted since a long time (months ago). Basically it has never been populated, to be honest, lol. 3 max isn't a * real* success, but an honorable fail :P. Spawnlist doesn't have primary key from basic. So far, I think it's a generic L2J mark (correct me if I'm wrong), from IL (and surely before) to current times. You have to add one column only for that if you really need.

Well in previous post I gave link to another thread (which itself links to another thread)... Just use http://trac.assembla.com/acis_project/changeset/115 and analyze how I implement it. It's easy, as the timeline is clean you will see only related stuff speaking of zone implementation. PS : I never opened a Java book, and learnt only by analyzing L2J existing code. 1,5 year ago, I didn't know how to compile a pack, not even the difference between a compiled pack and sources. I never read a tutorial aswell, except Eclipse checkout/compile. You learn everyday and that's the funny part. Same about people who help me doing quests. My main helper is correcting the first quests he did some months ago. He is able to make easy and medium quests from A to Z, without need to test them. That's a long process, but when you're able to do what you want, well... That's the interesting part :). Don't give up, do it all alone and you will see in a couple of months you will laugh reading that post ("how noob I was"). If you got troubles to adapt, just post your errors, and we (I) will answer.

If that continue, I would have to create T-shirts with aCis logo for fans ^^. Well we spoke about it with sterios on MSN, I agree with minideed about the fact all L2J packs are bugged when you talk of IL (and basically, any chronicle got bugs so far lol - L2J Freya got big fail on Formulas.java since years and list of bugs is so long than you could die just listing them all - I don't throw stones to L2J at all, it's a really big project). The only good criteria for me (if I had to choose, but if I made my own pack, that means there wasn't good alternative for me) is the developer's capabilities to implement original stuff (by original, I mean add a missing AI - speakingNPCs, hotSpringsAi - which doesn't exist on any public source, or correct an AI -Baium in my exemple - following retail videos/infos and only on that) and to correct stuff. And eventually, to avoid to add more bugs than original pack. If you're chained to the initial pack you based your pack on, basically it's pointless to begin a pack :P. And that's the error of many packs. They're happy with what they got, add some customs and that's enough. Personally I don't search to avoid bugs. Each big rework I made (items, skills and since some months, quests) was needed. I'm at 116 quests/AIs atm. People often complain about broken packs, but if they were working instead of crying on their fat asses, they could have a decent IL pack really fast ^^ (I don't say that for ya sterios, your reports are precious). Regards. Ok, a fast scoop : I'm in fact at 119 quests. Implementing Siegable clan halls, 3 "working" (according to L2J) atm. And ty BiggBoss for the good work :P.