Github Repository for KO Development

[twostars]

So what happens after u finish the client,doesn't matter when,you finish it 100%,bug free,everything is working like it should,is it going to be open source so everyone can use it or you are planning on making a server ? (just ur name in any server would bring at least a thousand people ).

It's staying open-source. I wouldn't personally run a server using it, but working on it does help progress my private emulator project (haven't cared to focus much on new functionality yet -- this gives me the opportunity to research/reverse it for whenever I do end up implementing it there) which, IMO, is going to outperform it no matter how much rewriting is done to the public project. Don't get me wrong, apart from the fact it's incomplete right now, obviously the design is "fine" as mgame use it, private servers have used it, etc -- I just don't like the fact there's many points to bottleneck with it, which is something that I'm working on reducing but will probably never do better than my private project; thus far I've made some pretty major changes to the base (got a lot to go, mind), but even I have a line to draw there. More attention needs to be made on implementing server functionality at this stage (which I focused on a bit today, it's actually starting to feel like a usable server!).

I guess what I'm saying is maybe, but if I am involved with a server it'll be using my private emulator (not that there's anything wrong with using/developing the public one!).

[twostars]

And another month has passed.

Since, we've (some of these earlier ones may have been bordering on last month):
- implemented capes & clan symbols,
- implemented NP symbols,
- implemented support for the helmet visibility packet (where you tick the checkbox next to the helmet to hide/show it to other players. I'm not sure why it really exists, but we support it! ),
- implemented arena challenge/duel requests for clans/players,
- fixed the skillbar implementation,
- removed Aujard (it's all handled directly in Ebenezer, in dedicated database threads now),
- implemented the letter system,
- implemented buying merchants (more or less),
- removed MFC (so stuff's a LOT less bulky, which is especially noticeable with startup times -- it's insane how slow their database code is),
- reversed their item exchange implementation (i.e. for FT chests, abyss gems, etc),
- implemented transformations (type 6 skills),
- implemented better support for type 4 skills ([de]buffs),
- implemented better support for type 9 skills (visibility stuff & clan "Guards" [but these aren't supported yet]),
- written up the quest engine (no more EVT, it's all *proper* Lua -- unlike official scripts) & made support for their quest state storage (strQuest/sQuestCount) with a little bit more room,
- made several large changes to the skill system to aim for it to be reentrant & support any player/NPC caster/target (their existing system requires 3 separate classes for same behaviour, they only really exist because they didn't think to derive the player/NPC classes from a base unit class where all this was accessible for them to use anyway, which we've done). Huge work in progress, the system's continually evolving (so bugs may or may not appear each time this happens, but they're typically really minor and addressed pretty quickly),
- worked a lot on cleaning up the project (e.g. cutting off dependencies) to optimise compile times. So far we've had a lot of success here, I'm personally saving upwards of 30s per compile (was previously ~60s compiles, now it's typically ~25-30s) -- for a full rebuild (not something you tend to do often; with the trimming of dependencies, minimal rebuilds are also considerably improved). Note that I'm running it on an SSD, with fast RAM & a decent CPU -- so I'm sure the less fortunate people will benefit a lot more from these changes than I did,
- and of course, much much more (that's off the top of my head in what should be chronological order).

There's still a lot to be done of course (skill system support [though this is largely been underdeveloped while it's being merged, to avoid having to reimplement the same code multiple times. Ugh.], new clan system stuff, alliance system, king system, mining, upgrades, etc), so we could still very much use any help we can get.

With Lua support recently implemented (on Sunday!), we should also consider starting to write up some of these quest scripts, so that we know whether we need to change and or implement new functionality to better allow for them.
Our Lua implementation's what I'd personally consider a "proper" implementation, as opposed to mgame's official implementation which leaves a lot to be desired (on-demand file i/o for one, but referring more to the scripts themselves), as you're able to refer to player/NPC objects (and I'm tempted to expose the main server object also) directly & all code gets run through an entry point (Main()) in a basis for writing scripts that actually branch correctly & more importantly, make sense.

If writing Lua scripts is more your forté (or it maybe even simply interests you more than C++!), you should jump on our IRC channel @ irc.quakenet.org / #snoxd-server (QuakeNet Web IRC (qwebirc)) to get started (you'll probably want to read through this, http://www.snoxd.net/index.php/topic...h-the-project/ as it'll be easier to work with the project directly). Usually I'll be there, but even if I am there and not responding don't fret, I'm either asleep or just a little busy with something. Idling's the best way to go (but sidenote: while I read through Lua's official documentation quite extensively for implementation, I am still pretty new to the Lua language itself. For Lua specific questions, I'd refer to their site or some other source. For help regarding what our scripts expose & how they're meant to function, you can ask me :P).

Thanks.

[Blaze]

Sounds great. Few years from now I finished my study(Game Development) and I might be in a position to help out. For now I wish I could.... Anyways keep up the great work!

[twostars]

Looks like I haven't updated this thread in a while.

Since the last post a lot of work was done on getting skills happy and merging a lot of the redundant player/NPC code so there's a whole lot less to focus on (and is considerably easier to read through).
The foundation of the King system was written; timers and stuff run, candidacy is implemented, notice boards, and also on the fun side of things, the King events. I recall getting a little bored of all the handlers at some stage, so it's definitely not all that finished (the rest of the election process needs doing, I recall starting work on rewriting it to clean it up.).
Off the top of my head, the other major thing I can think of is that the anvil (upgrading normal items, not accessories/rebirth items -- yet) has also been implemented.

We used to have some semblance of a progress thread on the forums originally, but that was long removed. Recently, I've posted a new issue on the tracker with the current (known) server statuses, so it's probably a better indicator of progress than simply stuff off the top of my head (though, that also is... but it's been a work in progress! :-P):
https://github.com/twostars/snoxd-koserver/issues/78

Portability has been another main focus, since the last post. We made use of new C++11 functionality to speed up (and instigate, really) the porting process. VS2010 does not support C++11 enough (it was designed for an early C++0x draft) to use it, so we were going to drop support completely... however, since some users still use Windows XP (which doesn't support VS2012) the code's been left backwards compatible. Where possible, VS2012/C++11 should be used though.

As a result of the porting process, we now have support for Linux (FreeBSD/MacOSX are lesser concerns, but I'll throw up kqueue [socket system] later -- there shouldn't need to be any other changes). Of course, you still need a Windows database server to run Microsoft SQL server (we're actually looking to see how well MySQL fits with the current ODBC database code, though. If it goes well, native MySQL support - rather than using ODBC - may happen... but it's not a priority. :-P).
The build process is (fairly) GNU-compliant, so it shouldn't be too hard to hook up. I've written up a quick guide (work in progress) here to help answer any questions with that: https://github.com/twostars/snoxd-ko...arted-on-Linux

I don't however advise using Linux unless it's something you're familiar with. Porting the server to other operating systems was never a priority, it just became so because: a. it became feasible without too much effort (C++11), and b. a couple of potential contributors find it extremely inconvenient to resort to using Windows purely for the project's development. Native Linux support definitely helps there.

So, yeah. It's definitely still being worked on. If you're looking to contribute, it's much appreciated -- you can find us in IRC (see the main post). Note that it's IRC, so... sometimes we may not reply for a while. It doesn't mean you're ignored, it's just that IRC hasn't been checked yet (or we're asleep- ... yeah, right.).

Fun times.

[ScreameR]

Seems to be a great project! Looking forward to see the results. To bad i am not able to help out. Just wondering, what can i expect out of this, will this be like the newest, most updated Knight Online private server database for every developer to use for his/her own server? Because if i read this correctly, you are actually a team of developers working on this same project.

Why not make 1 good server out of it? Which will be exactly like USKO or even better, i bet with the name's of these developers on it as creators, the great quality i expect of this server, this can turn out as one of the greatest servers ever made, if not... the best out there.

Since USKO and EUKO are both not as good as real KO used to be, with heavy PK's lot of people trying to level up. Which hey.. took a lot lot lot of time. No babashopping, people actually had to get some dirty hands before they could join the fun.

This could actually bring back a really big part of that. All i read while reading these forums is people complaining about how dead a server can be, or even worse. I see this project as an last opportunity for KO to get back to life again. Atleast, the fun great memories we can't get back to right now with the current ''failing'' US and EU Knight Online. In other words. Advertise, work hard ask for people's opinion try to get out of it what me and i hope alot of other people's are hoping for. a whole new server which brings back to life, what once was Knight Online.

[twostars]

Seems to be a great project! Looking forward to see the results. To bad i am not able to help out. Just wondering, what can i expect out of this, will this be like the newest, most updated Knight Online private server database for every developer to use for his/her own server? Because if i read this correctly, you are actually a team of developers working on this same project.

I realise a lot of people confuse the term 'database' with 'server files', so if this what you meant, yes. It's an attempt to have publicly available, open-source server files which natively supports new clients and their new functionality (e.g. events) properly (so no more stupid hacked up events that aren't quite the same as the real events on official servers due to limitations on the client/server ) and by default (which isn't really a restriction on the client so much as it is the primary direction we're taking. Perhaps in the future we'll go back and support older clients.).

Why not make 1 good server out of it? Which will be exactly like USKO or even better, i bet with the name's of these developers on it as creators, the great quality i expect of this server, this can turn out as one of the greatest servers ever made, if not... the best out there.

Since USKO and EUKO are both not as good as real KO used to be, with heavy PK's lot of people trying to level up. Which hey.. took a lot lot lot of time. No babashopping, people actually had to get some dirty hands before they could join the fun.

The one thing to note here is that this project aims to develop the server files for anybody to use. It is not, strictly speaking, a project intent on providing a private server for players to play on (i.e. doing up spawns, drops, rates, etc) -- that comes later, by whoever uses the server. Having said that, when it comes to development, I try hard not to make any promises I may or may not be able to keep -- but I suppose it is possible, in the future, to be involved with a server using these files (notably, I was against it early on). I cannot personally speak for any others who collaborate, as I am sure they have their own intentions.

I guess what I'm trying to say is, when the project's reached a more usable/runnable state, this possibility can be better explored.

This could actually bring back a really big part of that. All i read while reading these forums is people complaining about how dead a server can be, or even worse. I see this project as an last opportunity for KO to get back to life again. Atleast, the fun great memories we can't get back to right now with the current ''failing'' US and EU Knight Online. In other words. Advertise, work hard ask for people's opinion try to get out of it what me and i hope alot of other people's are hoping for. a whole new server which brings back to life, what once was Knight Online.

I really hope so, the server being open source opens up a lot of doors for developers wishing to change or even implement never-seen-before functionality. Working with binary files is seriously limiting -- the bar required for getting in and changing things to your fancy is simply too high for most people. Changing/implementing major functionality is also typically out of the question for feasibility reasons. When you're working with source code though, you can really do as much as the client lets you.

I think private servers in their current form have been done to death. The files themselves simply aren't stable/reliable enough on their own, but the limitations induced by them have spawned private servers that have all been, IMO, more or less the same (i.e. there's not been all that many distinguishing features about them). I think, on that front, something needs to change. KO is a fine game, and most people would insist that it changing (officially) is the only thing that's ruined it, but neverheless -- I think it does need room to breathe. Buggy behaviour cleaned up, and a bit of room to breathe for some fresh creativity.

I understand a lot of people will disagree with me, as they feel adding new stuff ruins the game -- and well, in regards to official servers, that's probably more or less true (though there have been a handful of interesting features amongst the bad). I don't think this shouldn't be reason to dismiss all further creativity from private server developers' part, though. I think there's so much that can still be done with the game, the ideas just haven't been able to be expressed.

[ilovemyko1]

I realise a lot of people confuse the term 'database' with 'server files', so if this what you meant, yes. It's an attempt to have publicly available, open-source server files which natively supports new clients and their new functionality (e.g. events) properly (so no more stupid hacked up events that aren't quite the same as the real events on official servers due to limitations on the client/server ) and by default (which isn't really a restriction on the client so much as it is the primary direction we're taking. Perhaps in the future we'll go back and support older clients.).


The one thing to note here is that this project aims to develop the server files for anybody to use. It is not, strictly speaking, a project intent on providing a private server for players to play on (i.e. doing up spawns, drops, rates, etc) -- that comes later, by whoever uses the server. Having said that, when it comes to development, I try hard not to make any promises I may or may not be able to keep -- but I suppose it is possible, in the future, to be involved with a server using these files (notably, I was against it early on). I cannot personally speak for any others who collaborate, as I am sure they have their own intentions.

I guess what I'm trying to say is, when the project's reached a more usable/runnable state, this possibility can be better explored.


I really hope so, the server being open source opens up a lot of doors for developers wishing to change or even implement never-seen-before functionality. Working with binary files is seriously limiting -- the bar required for getting in and changing things to your fancy is simply too high for most people. Changing/implementing major functionality is also typically out of the question for feasibility reasons. When you're working with source code though, you can really do as much as the client lets you.

I think private servers in their current form have been done to death. The files themselves simply aren't stable/reliable enough on their own, but the limitations induced by them have spawned private servers that have all been, IMO, more or less the same (i.e. there's not been all that many distinguishing features about them). I think, on that front, something needs to change. KO is a fine game, and most people would insist that it changing (officially) is the only thing that's ruined it, but neverheless -- I think it does need room to breathe. Buggy behaviour cleaned up, and a bit of room to breathe for some fresh creativity.

I understand a lot of people will disagree with me, as they feel adding new stuff ruins the game -- and well, in regards to official servers, that's probably more or less true (though there have been a handful of interesting features amongst the bad). I don't think this shouldn't be reason to dismiss all further creativity from private server developers' part, though. I think there's so much that can still be done with the game, the ideas just haven't been able to be expressed.

Do you not think by having them so anybody can use that it would ruin the private server community by having the whole population spread out?

[twostars]

Do you not think by having them so anybody can use that it would ruin the private server community by having the whole population spread out?

This argument has come up before, and really, there's so much more to it than that.

Starting with the development process... if not open-source, there'd be a lot less people watching over changes (i.e. and spotting any bugs)/bug reports/improving the files in the begin with. A lot less input.

Take my private emulator (yes, that thing's still there!) as an example of a private project.

The amount of work I've put into that is insane, yet it's not done (less help), and not all functionality works (officially) as it should as there's been a lot less input on how such functionality is intended to work. People want to change this? Fat chance, the code's not available.
There's also been a bit of oversight on my part when checking in changes; I find when working on something others will look at, I spend the extra time double-checking to ensure I didn't leave anything out and that there's no typos. This isn't always going to happen, but when it does, these problems can be noticed fairly quickly in an open-source project (case in point, this commit right here: https://github.com/twostars/snoxd-ko...11cd5b978a2a64 -- thanks Max!).

(edit: snipped incoherent ramblings of a crazy person)

Above all that, I may be naive, but I honestly think if the server's unique enough, stable enough, and managed properly, it doesn't matter whether the same server files (perhaps without the same unique tweaks specific to this server, but that's the idea -- making things unique) are used on 50 other servers. That is, if it's worth playing, it will be played.

I think this project will help accomplish that, at least by providing the established server files to work with. It give servers an improved baseline, taking the focus off so many lesser issues and helps leave room for the other jobs that need doing. Servers with staff that try to take every role don't typically survive all that well. Taking the pressure off at least one component of running the server (the horrible files and the problems associated with them) should, at least lighten the load...

I don't think the server files stand on their own. They help, sure -- but servers are ultimately still run by people. Not code. People need to be involved; dealing with the community's needs, maintaining the server, pushing the server by fuelling it with fresh ideas and ensuring they get implemented. I think that the right people working together with the right files can make magic happen.

"Spreading the population" really isn't an issue I think anyone should be worried about. If your server's not attracting attention, it's probably being done wrong -- even if it's just through lack of advertising. The server should stand on its own, the files can only speak for so much.

[Chunk]

A rush on this if you don't mind, the servers that are coming out lately just smh.

[Kallop]

A rush on this if you don't mind, the servers that are coming out lately just smh.

Rome wasn't built in a day.

[twostars]

A rush on this if you don't mind, the servers that are coming out lately just smh.

Feel free to lend a hand...

There's a bunch of issues just sitting there waiting for more information to be provided on certain things. You don't need to know how to program to help out; if you play KO (especially - but not just - recently), it can be a pretty big help.

[Nagato]

why don't you focus on building a 1534 version server?

USKO sucks, with all these new useless effects, armors and skills.

imo, 1534 version is the best thing happened to KO. Working skills, long capes, active hp/mana, party bars. it's a balanced version between old MYKO and new USKO.

[twostars]

why don't you focus on building a 1534 version server?

USKO sucks, with all these new useless effects, armors and skills.

imo, 1534 version is the best thing happened to KO. Working skills, long capes, active hp/mana, party bars. it's a balanced version between old MYKO and new USKO.

Actually, most of the difference is actually database/client-related -- stuff's just added on. All the features that existed before then should remain compatible (minor tweaks here and there to some of the main packet structures, but they're negligible).

Skills, rates, items + their stats, level caps, etc are all database-related and subject to the owner's discretion.

Regarding the server, using 1.534 or 1.886 makes no difference at all, it all accomplishes the same thing. You could use the 1.886 server to run a server with 1.534 functionality.. or you could make the few slight tweaks required to "downgrade it" so the old client is supported.
In time, I wouldn't mind going through and implementing some logic to handle that -- but right now, there's absolutely no point; all the existing features work the same, so we might as well be working closer to official to get the newer features reversed too.

Fun fact: an enterprising individual did this already with a few month old copy of our project. It's released as "SOHU v3 1.534" or some such, i.e. official files (they're not official though, obviously).
It'd probably take a LOT less than an hour to "downgrade" it.

[Nagato]

i got you and that's awesome,

everything is fine for me, but the only thing i'm afraid of is minor cooldown, lol. that's why i worry so much..

[Chunk]

Feel free to lend a hand...

There's a bunch of issues just sitting there waiting for more information to be provided on certain things. You don't need to know how to program to help out; if you play KO (especially - but not just - recently), it can be a pretty big help.

How could I help, if you don't mine me asking. I've read through most of them, one I was thinking is your premium system. I wanted to ask if the text would be on your chat bar or you would have it in the top right corner with a drop down system like USKO.