Crystal Server

[Tryller]

Just because my name isn't in the canary files I have no relevance whatsoever, sorry little grasshopper

Post automatically merged: Mar 1, 2025

It looks like we’ve got a troubling pattern here. He implements half-baked solutions full of issues, then we have to scramble to fix everything on our end, only for him to copy the improvements. Check out these examples: commit cb87ea6 and commit 7c93fd4 show poorly executed changes that needed follow-up fixes, and then commit 5c1262b and commit 5dc827c pop up as if he “solved” something that was already broken because of him in the first place.<br><br>Even worse, he takes PRs in draft—ones that are obviously buggy, like this commit 1dee569—and throws them straight into his base. Where’s the testing for all this? Are you the one putting buggy code into your base because you don’t validate anything, or is it us, who are trying to work properly and then see our efforts copied without any discernment?<br><br>And there’s more: he keeps cherry-picking stuff from Canary, like in commit 809baa3 and commit a74f85f, takes the credit, and pushes it forward as if he knows what he’s doing. It’s pretty clear he has no clue what’s going on and just wants to reap what we sow. How is this sustainable for anyone?

Do you really want to discuss credits? you're going to be embarrassed, just look at the commis of Eduardo and Majestyt, where are the credits to me? Don't come to embarrass yourself you Any response coming from canary will be ignored, this is a thread for people to collaborate with Crystal Server and not a thread for birds to be embarrassed

 

[Beatss]

Do you really want to discuss credits? you're going to be embarrassed, just look at the commis of Eduardo and Majestyt, where are the credits to me? Don't come to embarrass yourself you Any response coming from canary will be ignored, this is a thread for people to collaborate with Crystal Server and not a thread for birds to be embarrassed

You are a joke. Cheers

 

[Felipe93]

We, from the OpenTibiaBR community and the Canary project, have noticed your recent actions, including cloning our repository and attempting to attract contributors while criticizing our image. As someone actively involved in the project (I’m Beats, by the way), I would like to address this directly and based on facts.
First, the Canary project is a collective effort, built on clear guidelines that exist to ensure quality, consistency, and respect for the work of everyone involved. These rules are not arbitrary; they reflect years of experience and adjustments to keep the project sustainable and useful for the community. Your decision to ignore these guidelines and create a fork without seeking dialogue or alignment goes against the spirit of collaboration that open source represents.

Second, cloning a repository is a right granted by the open source license, and we have no problem with that — it’s part of the freedom we defend. However, using this to try to compete unfairly, attracting contributors with vague promises while denigrating our team’s work, is at least unethical. If you have criticisms or suggestions, the correct path would be to open a discussion in our official repository, such as an issue or pull request, so we could evaluate and improve together. Opting for public attacks instead of collaboration only weakens the community as a whole.

Finally, I want to remind you that Canary is not “ours” in the sense of ownership, but rather belongs to the community that sustains it. We have dozens of active contributors, regular commits, and a solid history of deliveries — all documented and transparent on GitHub. If your goal is truly to contribute to the OpenTibia ecosystem, why not join forces instead of dividing? Otherwise, your approach only raises doubts about your intentions.

I invite you to reconsider this stance and seek an open dialogue with us. The door is open for those who want to build, not destroy. If you prefer to continue with your clone, I wish you luck, but I hope it is done with respect and without trying to erase the work of those who came before.

Best regards,
Beats
OpenTibiaBR - Canary Project

really agree with u have seen ppl doing this nonsense making another repositories taken from others branch and such, is simply stupid. and divides the community, leading everyone to nowwhere working in multiple proyect meanwhile it would be easier to work in only just one all together. but well... that's how this shit goes the same as always

 

[heitorzinhotc]

This base is the strangest copy I've ever seen, the guys just copy what they want from another repository but are ungrateful and don't give credit. LOL

 

[Tryller]

Where is TFS credit ?
You Just ignored

 

[Beatss]

Do you really want to discuss credits? you're going to be embarrassed, just look at the commis of Eduardo and Majestyt, where are the credits to me? Don't come to embarrass yourself you Any response coming from canary will be ignored, this is a thread for people to collaborate with Crystal Server and not a thread for birds to be embarrass I have never cried over credits and the person who does that to me is an idiot

Do you really want to discuss credits? you're going to be embarrassed, just look at the commis of Eduardo and Majestyt, where are the credits to me? Don't come to embarrass yourself you Any response coming from canary will be ignored, this is a thread for people to collaborate with Crystal Server and not a thread for birds to be embarrassed

I have never cried over credits and the person who does this to me is an idiot, if you did it, you did it and now you are claiming credit??If that were the case, we would have to cry to Microsoft in copilot because they trained us on public and private projects... You are just another person in the bread line.

 

[dgprado]

Open Tibia is a project that has been in development for nearly three decades. The Open Tibia project has always been characterized by the community's participation. I find it regrettable that, in recent years, the Open Tibia project has turned into a machine that exploits the work done by the community, ultimately limiting the project's development to finance the interests of a small minority.

And this is precisely what Canary has been promoting in recent years: the rise of super servers and the stagnation of the Open Tibia community. There are reports from former Canary core members claiming that the private project is sold for R$ 10,000.00 and that part of the content was allegedly obtained from "customers."

Not to mention that users are treated like criminals, when in reality, they only want to help the community. Many of the project's collaborators are blind to the situation and do not realize that they are dedicating hours and hours of excellent work to a project that is not moving forward, while a restricted core takes advantage of these people's efforts.

 

[Beatss]

Open Tibia is a project that has been in development for nearly three decades. The Open Tibia project has always been characterized by the community's participation. I find it regrettable that, in recent years, the Open Tibia project has turned into a machine that exploits the work done by the community, ultimately limiting the project's development to finance the interests of a small minority.

And this is precisely what Canary has been promoting in recent years: the rise of super servers and the stagnation of the Open Tibia community. There are reports from former Canary core members claiming that the private project is sold for R$ 10,000.00 and that part of the content was allegedly obtained from "clients."

Not to mention that users are treated like criminals, when in reality, they only want to help the community. Many of the project's collaborators are blind to the situation and do not realize that they are dedicating hours and hours of excellent work to a project that is not moving forward, while a restricted core takes advantage of these people's ef Did you take your medications before commenting this???

Open Tibia is a project that has been in development for nearly three decades. The Open Tibia project has always been characterized by the community's participation. I find it regrettable that, in recent years, the Open Tibia project has turned into a machine that exploits the work done by the community, ultimately limiting the project's development to finance the interests of a small minority.

And this is precisely what Canary has been promoting in recent years: the rise of super servers and the stagnation of the Open Tibia community. There are reports from former Canary core members claiming that the private project is sold for R$ 10,000.00 and that part of the content was allegedly obtained from "clients."

Not to mention that users are treated like criminals, when in reality, they only want to help the community. Many of the project's collaborators are blind to the situation and do not realize that they are dedicating hours and hours of excellent work to a project that is not moving forward, while a restricted core takes advantage of these people's efforts.

Did you take your medications before commenting this???

 

[heitorzinhotc]

Open Tibia is a project that has been in development for nearly three decades. The Open Tibia project has always been characterized by the community's participation. I find it regrettable that, in recent years, the Open Tibia project has turned into a machine that exploits the work done by the community, ultimately limiting the project's development to finance the interests of a small minority.

And this is precisely what Canary has been promoting in recent years: the rise of super servers and the stagnation of the Open Tibia community. There are reports from former Canary core members claiming that the private project is sold for R$ 10,000.00 and that part of the content was allegedly obtained from "customers."

Not to mention that users are treated like criminals, when in reality, they only want to help the community. Many of the project's collaborators are blind to the situation and do not realize that they are dedicating hours and hours of excellent work to a project that is not moving forward, while a restricted core takes advantage of these people's efforts.

LOL

that was the biggest piece of crap I've ever read, by the way, what can we say about TFS then?!

Thanks to Canary, because if it depended on everything else, we would be failures, and that is the reality. TFS and Otland are killed long before Canary and yes, the Brazilians keep everything else active, without the Brazilians everything would be dead thank them for that. There are even Brazilian otservs that make new players get to know Tibia.

And you seem to be a Brazilian who is envious of Brazilians. That's strange!

 

[carlospessoa]

Is this repo a joke or what?

 

[Terotrificy]

Wow such a big drama. I only questioned the fact of 50 fixes per month was not a good sign, but it's not like TFS is the greatest open source engine out here, TFS is garbage, stop it lol.

 

[Itutorial]

Wow such a big drama. I only questioned the fact of 50 fixes per month was not a good sign, but it's not like TFS is the greatest open source engine out here, TFS is garbage, stop it lol.

Which one is better?

 

[Error 502]

I prefer an active community that reports issues, resolves them, and where pull requests are tested and approved by the community in a timely manner. I also value innovation, with PRs that bring improvements like Prometheus metrics, full Tibia 12+ features, cast system, multiworld, among other features.
that some people sell for over $200 in TFS or that are stuck in PRs that were never finished.

For example, in TFS, there are currently 185 issues and 59 PRs, most of them obsolete or untested by the community — some have been waiting for more than 4 years (Mar 4, 2021) to be either rejected or approved . o issues with more than 10 years (Jan 12, 2014)
The latest commits in the main repo are minor and insignificant changes.

Every community has conflicts that slow down development, and TFS is no exception.
For instance, merging the NPC Lua system by Evil, only to revert it shortly after due to Linux-related issues.


Developers leave the community( omarcopires ) because of situations like this.

or


etc...

I even ask myself: if I submit a PR to tfs, how many years would it take to get approved? Would anyone test it? How many years would I have to keep it updated?

About Crystal Server vs. Canary:
Who cares , Crystal only has 1 developer, while Canary has many.

and the debate is why didn't they give each other credit?
Crystal is based on Canary, and canary uses tfs as a base. That’s it — why emphasize something that’s obvious?

 

[StreamSide]

developer is a big word

 

[dgprado]

LOL

that was the biggest piece of crap I've ever read, by the way, what can we say about TFS then?!

Thanks to Canary, because if it depended on everything else, we would be failures, and that is the reality. TFS and Otland are killed long before Canary and yes, the Brazilians keep everything else active, without the Brazilians everything would be dead thank them for that. There are even Brazilian otservs that make new players get to know Tibia.

And you seem to be a Brazilian who is envious of Brazilians. That's strange!

heitorzinhotc, you, in particular, deserve a round of applause from the community for the time and energy dedicated to creating all that shambles. Only you could get Majesty to work. kkk

 

[Terotrificy]

Which one is better?

I have no idea, no public distro of tibia out there is good enough when it comes to architecture design, code quality or ram usage.

 

[Codinablack]

I have no idea, no public distro of tibia out there is good enough when it comes to architecture design, code quality or ram usage.

The ram usage is only an issue when you are running maps like global tibia with a billion tiles and still using the "otbm" format that was created as a revolutionary memory saver (compared to xml) like two decades ago. The tile class, the otbm, and all other data structures centered around those two, desperately need reworked to become lighter, and that will aid in the endeavor to lessen the RAM usage... but there are a lot of other places for these potential optimizations as well, its just not a huge focus for teams built for replicating real tibia and it's features.

I have personally already removed the virtualization of the tile class, as well as removed all its leaf classes, in attempt at lowering the memory footprint of each tile. There are of course several other things that need addressed that could reduce the memory footprint of the server itself, but most projects just aim to emulate tibia, and so it will never be their priority for such things (as far as I can tell).

 

[Itutorial]

I prefer an active community that reports issues, resolves them, and where pull requests are tested and approved by the community in a timely manner. I also value innovation, with PRs that bring improvements like Prometheus metrics, full Tibia 12+ features, cast system, multiworld, among other features.
that some people sell for over $200 in TFS or that are stuck in PRs that were never finished.

For example, in TFS, there are currently 185 issues and 59 PRs, most of them obsolete or untested by the community — some have been waiting for more than 4 years (Mar 4, 2021) to be either rejected or approved . o issues with more than 10 years (Jan 12, 2014)
The latest commits in the main repo are minor and insignificant changes.

Every community has conflicts that slow down development, and TFS is no exception.
For instance, merging the NPC Lua system by Evil, only to revert it shortly after due to Linux-related issues.

View attachment 90656
Developers leave the community( omarcopires ) because of situations like this.
View attachment 90657
or
View attachment 90658

etc...

I even ask myself: if I submit a PR to tfs, how many years would it take to get approved? Would anyone test it? How many years would I have to keep it updated?

About Crystal Server vs. Canary:
Who cares , Crystal only has 1 developer, while Canary has many.

and the debate is why didn't they give each other credit?
Crystal is based on Canary, and canary uses tfs as a base. That’s it — why emphasize something that’s obvious?

It seems are you comparing paid distros vs free distros. With a game this old you aren't looking at a huge community willing to develop things for free especially when you finally do and all you get in return is "finally you did something" as if it is owed.

Canary's code base is grotesque. I am not even trying to talk bad about it there is so much code everywhere including copy paste and plenty of uneeded code. Canary is better for having more features but TFS is better for overall quality.

Ignoring paid distros (because you can't compare apples and oranges) we see two options.
1) The code is updated all the time which comes with unclean, unorganized, and large amounts of broken features.
2) The code is updated slowly giving plenty of time for what is already included to be fixed.

I will always go to the code base with more quality no matter what stage of development because one day everything will have to be redone. One day there is going to be enough bloat it is completely unusable.

It's cool you started your own distro but if you really think the limited community that even works on these projects are just waiting for someone to press the merge button you are mistaken.

My current PR to update TFS pathmatching has been 3 years in the making. During those 3 years I have turned it into probably one of the fastest versions you will find even if you paid. Obviously I didn't work on it for 3 years but I worked on it whenever I had free time. If the code was pushed into the repository anytime before the last month or so there would of been too many problems to even want it merged. So all that time was needed for me to take the time to learn all the TFS source code involved with monster and player pathmatching behavior and create a system that is really great.

Someone better may have done it in a week but we aren't working with 100k professional developers. We are working with a small community that are willing to try to improve code for free. ON OUR TIME.

 

[Luan Luciano]

Interestingly, you say that 99% of some fork code comes from Canary... Why didn't you mention that 99% of Canary code comes from TFS and its forks xD

the best answer kkkkkkk

 

[Terotrificy]

The ram usage is only an issue when you are running maps like global tibia with a billion tiles and still using the "otbm" format that was created as a revolutionary memory saver (compared to xml) like two decades ago. The tile class, the otbm, and all other data structures centered around those two, desperately need reworked to become lighter, and that will aid in the endeavor to lessen the RAM usage... but there are a lot of other places for these potential optimizations as well, its just not a huge focus for teams built for replicating real tibia and it's features.

I have personally already removed the virtualization of the tile class, as well as removed all its leaf classes, in attempt at lowering the memory footprint of each tile. There are of course several other things that need addressed that could reduce the memory footprint of the server itself, but most projects just aim to emulate tibia, and so it will never be their priority for such things (as far as I can tell).

I agree, I mean using TFS or any other public distro as it is today is fine for playing with friends or a small community, but it is very far from being a serious productive server if you don't personally fix it's flaws, no matter if it's canary, nostalrius or any other public distro as long it's forked from TFS.
There are several posts that address the problems and TFS developers are very aware of it, I ignore why these problems haven't been fixed to this date.

But to address the most criticals:

Bottleneck: Single core design, main thread doing live database calls can freeze entire world. Most noticeable when players log in or logout.
Pathfinding: Very expensive resource wise.
Items duplications: As long as you can crash the engine, it's very likely to be able to dup items on TFS. Since players are saved upon logout or death, they are not aligned with house tiles or other players, so the data persistency is not consistent. It just takes a player to:
pickup an item from a house tile, logout (item is now saved into player inventory), crash server. Since house tiles are saved only on server save, you will have now item on house tile and also dupped on player inventory. You can do basically the same with two players, no house needed.

If you take a look at the tarball you will instantly notice depot inventory is loaded on step in, not on player login, that's a core mechanic of tibia that shouldn't be so off in TFS (because it also takes part on decay mechanic, which is also not working properly on public distros) and would help to improve greatly the bottleneck, not to mention the use of pthreads that cipsoft did back in 2007, and using a proper querymanager to communicate between the server and database.
Cipsoft also addressed items dups decades ago, and we have the binary files there.
TFS would improve greatly just taking a look at it.
Even TFS's raid mechanic is non-sense, checking every x seconds to throw dices and see if it should trigger or not a raid. If devs would take a look at the real thing, the tarball just threw the dices once on server startup and scheduled all the raids for the day, no need to be wasting resources very x second to do calculations the entire time. Not to mention raids have two categories, BigRaids and SmallRaids, which are critical to properly schedule them.
Right now, as how TFS is, you won't be able to trigger rookgaardrats raid or any other intended daily raid using the daily value parameter, because it can just miss all the dices throws in a day.
There are many other things that might be improved, like making custom loot for creatures on raid (instead of creating a custom orc warlord to just add amazon loot lol), dissapear monsters after x time (just like cipsoft did with many raids), etc.

I may be wrong in some points since It's been a while I don't check at TFS, my apologies if some of the points I stated are wrong. But I believe the correct direction would be to reverse engineering the tarball, get the best possible things out of it, implement on TFS, address the bugs and then keep moving on to catch whatever tibia turned into nowadays, if that's even still the goal lol.