L2J Server

First of all I want you to know that I love your work,
I use your files/fixes all the time and I am greatful that you share.

Recently I oppened a server based on L2J,
I have about 500 online, but many people start quiting.

Some current reasons for players quiting are the following:
Siege Golem - Damage to Castle walls
Kamael Traps - Activation does not work/proc
Herb icon stucks with buff icon - Visual bug
Cannot exit Dimensional Rift

Also many skills need rework, I wait for Zoey76 to make skills work again for months now.
But he doesn't seem that active and wonder if his skill/effect rework will ever finish.

I understand your point on making something only retail like,
because if you provide a tempfix will make you ignore the problem and move to the next thing to fix,
making that problem never be fixed as it should.
But that could make a server cripple for months and make all players quit until you make it retail.

I know you try to do the retail things/solutions, but it seems that you do not have the people to it efficiently.
People with much less potential than L2J have done things that you have not been able to accomplish since Freya came out.

Why don't you provide/accept tempfixes for problems and a create a TODO list for what is already hotfixed,
in depth of time you could make it work 100% retail like. Why you refuse to make L2J better?

Use Stable

Although i agree on the part you are trying to say that they must focus on making the basics working properly and work on the new stuff or improvements on the side on BETA, that could make it usable for live servers and give the chance for faster testing on new features. Right now the way things are you cant use BETA for live servers and its only so much you can report by playing withyourself on a test server.

For instance here is what happened to me last spring. I started working on my new server using BETA. It was the time i did a lot of bug reports and helped get a lot of stuff fixed, however my server failed because of bugs and other related stuff. This was the reason i am now using stable and i have stopped reporting bugs for BETA.

Even though the idea of having a BETA and STABLE branch is good, on second thought is kind of counterproductive for the reasons i gave above.

Fact is, everybody wants to make the code better, not so much make the missing features, because the former one is simply easier to do. Arguably, all the skill stuff could have waited because there were only a few things that didn't work as they should, while now, after all the "fixes", too many things don't work like they should (or even did before the "fixing"), and the next update might also not be a magic cure-all, just like all the previous fixes promised to be.

I had suggested before to get the L2J team to officially ask for people to apply as testers, who then would simply test the required stuff either on their own test servers or, better, on a common private test server for L2J use only. For whatever reason, somebody didn't like it, my guess is because the L2J devs like to be so secretive about their work.

Beta branch is suitable for live server only if the owner/developer knows l2j and java well enough to temp fix all problems until they get properly fixed by us.

Stable branch is considered to be stable for a live server use it until we get beta stable enough to replace previous stable rev.

For the problem with teleports i provided u with a location of the problem just remove the lines i showed and it will no longer does problems for u.

We focus on fixing existing code issues problems. If you wish i can provide a patch in which i will show you over 50+ wrong usages of one method. Where arguments passed to it are ints from an object instead of the object or ints wrapped into objects for unknown reason. Fixing one of this more well used methods can take ages and can eeasily lead to fixing 100+ errors. Sad thing is that anything you wish to fix properly later leads to big changes. We are trying to do this now so later when we start implementing features that will be easier.

pandragon wrote: Also many skills need rework, I wait for Zoey76 to make skills work again for months now.
But he doesn't seem that active and wonder if his skill/effect rework will ever finish.

LOL, I have done a rework of thousand of lines of new code, just to get home and read this, I work every day for L2j countless hours, the amount of crappy code I have removed couldn't be really factorized.

Instead of writing such threads, people should start learning coding in order to have a server, myself with all the knowledge and experience do not own a server, I couldn't imagine how others could do.

For your information the rework is almost finished and it's a complete rewrite of the effect and skill system, which was required due L2j was "working" but really far from retail.

About that herb stucking icon bug, with less than 10 lines of code could be fixed, I hope a server owner can manage such little thing.

I don't think any of the current core or soon-to-be core developers like adding temp-fixes to the SVN, they are shared in forums (sometimes by developers) and you can use them if you want.

xban1x wrote:We focus on fixing existing code issues problems. If you wish i can provide a patch in which i will show you over 50+ wrong usages of one method.

That is such bullshit, there is nothing wrong with passing ints instead of the object they're taken from, is is simply longer notation. You don't fix a damn thing by changing that, you simply "prettify" the code, instead of fixing REAL problems or implementing missing features.

Zoey76 wrote:About that herb stucking icon bug, with less than 10 lines of code could be fixed, I hope a server owner can manage such little thing.

If there was a problem and a fix was known, it should have been committed, even if it was temporary. No need to be a smartass about it by saying "if you can't do it, what are you doing here?" That's just bad attitude.

Zoey76 wrote:I don't think any of the current core or soon-to-be core developers like adding temp-fixes to the SVN

Think again. Or, better, actually ask around.

Why would i like to write or read xxx(obj.getX(), obj.getY(), obj.getZ()); instead of just xxx(obj); . I care about maintaining code and making sure it can be readed simply. You just care about raw performance.

You will never agree with me because we have diffrent views. But don't harass me cause of it and i don't agree with temp fixes inside svn. Except if it's in stable.

I'm not asking you to like it, I'm saying there are more important things to fix.

jurchiks wrote:I'm not asking you to like it, I'm saying there are more important things to fix.

That is your opinion, each dev has his own. We contribute to project freely and we contribute what we wish. If we don't share our opinion with yours or anyone elses feel free to express it but on the end me and any other dev, we still do what we want to do not what we are told to.

xban1x wrote:i don't agree with temp fixes inside svn. Except if it's in stable.

and this is totally wrong, dev's commit to svn without test and broke stuff, but commit fix for that broken stuff dont want, who will use beta branch then if code all time broken? no one, because no one want have crap server from where player quit in few day

lion wrote:

xban1x wrote:i don't agree with temp fixes inside svn. Except if it's in stable.

and this is totally wrong, dev's commit to svn without test and broke stuff, but commit fix for that broken stuff dont want, who will use beta branch then if code all time broken? no one, because no one want have crap server from where player quit in few day

Since when beta is used for live server? It's called beta with a reason.

You know too damn well there are tons of people who use beta for a live server because trunk is damn outdated.

jurchiks wrote:You know too damn well there are tons of people who use beta for a live server because trunk is damn outdated.

They should realise the risk or sync everything except effect rework manually.

Doesn't justify breaking code in the SVN.