Fanfo
patreon
MEGA POST | Official Server | Patreon Playtesting
🕑 Added 2024-12-24 02:52:32 +0000 UTC(At the moment, I don’t have access to my computer, but I had already promised this dev-blog a few days ago. I’ll update it with images as soon as I can.)
Happy holidays to everyone! I hope you’re enjoying your vacation with all your loved ones. Sending you my best wishes! This post will be long—I have a lot to explain about what we’ve been working on in secret over the past several months. Here you’ll find information about the official server release, the current state of the next shifter, and some final thoughts I’ve had over the past few months.
Since the release of Shingeki no Craft 2.2.4, the project has surpassed 100k downloads—what an incredible milestone! It’s amazing to see the growth rate increase every day. From the bottom of my heart, thank you all so much! Without further ado, let’s get started!
1. Official Server
Welcome to the Shingeki no Craft server! A server designed to deliver the best mechanics inspired by the show Attack on Titan in Minecraft! Showcase your combat skills with the ODM gear, fight as or against Titans, or recreate iconic roleplay scenes—all in one place, with no downloads or mods required. Just connect and play!
1.1 Server Access Information
IP Address: sncraft.minecraft.best
Version: 1.21.3 (Currently, the resource pack only works with this version.)
Creating this project has been a challenge for the entire team. I even had to learn how to make plugins in Java. I collaborated with Hope to develop a hub and the first game mode that fully integrates SnC with its ODM gear. The idea was to create a space where fans of both projects could enjoy an excellent multiplayer experience in every way (accessibility, performance, balance). I hope it’s something we can sustain in the long term despite the maintenance it requires.
1.2 Technical Information
I’d like to document the functionality and architecture used. Unlike other servers created in the past, the most important thing is that this is NOT a single host. A proxy was used to connect multiple servers, which communicate through reading and writing to a common database. It’s somewhat similar to how Hypixel operates and is the first step in creating a "serious" massive Minecraft server.
Initially, we opted to use "BungeeCord" as the proxy. However, thanks to the Smithed Summit event, I connected with more experienced individuals who recommended migrating our setup to "Velocity," the latest proxy technology available.
The need for a proxy arises from Minecraft servers/datapacks being single-threaded, a bottleneck issue that makes vertical server scaling (in this case, adding CPU power) infeasible due to SnC's costs (and eventually computational limits). The solution is to prevent bottlenecks by distributing the load across multiple processors. In other words, instead of having a single "powerful" server with 16GB RAM, the load is divided into four controlled environments with 4GB each (numerical values for illustration).
This approach inspired the idea of creating multiple game modes with high quality, engaging content, and mechanics tailored to provide a comprehensive SnC experience.
1.3 Bastion Sentinels
This is the first of the mentioned game modes. Its focus is on combat against Titans using Hope’s ODM gear. The lessons learned from developing this game loop will inform the creation of future game modes. This includes database communication through queries, runnables managing game states, scoring systems, and more.
The premise is simple: Defend the castle from waves of Titans that increase in intensity over time. Meanwhile, resources and terrain will gradually diminish until dawn.
There’s still work to be done, including:
Replacing redstone blocks with empty models.
Adding weak points and balancing existing ones.
Increasing Titan waves progressively over time.
Finishing the interior construction.
Adding a "falling tower" animation if the castle is destroyed.
However, the current state already features a completed game loop, making it possible to open the mode to testers and Patreons for bug hunting. I’d love to hear your feedback on this mode—it should be entertaining enough to attract new players and replayable for existing ones.
2. The Next Shifter
It's going well, surprisingly well. I'd even say too well if I compare it to the server. Basically, development has been on pause for a few weeks because my original plan was to do a reveal and a playable release on Patreon alongside the server release. This way, the players who support me financially could use it on the same day as the reveal, while also providing an incentive to fight a flop. "If you want to play the next shifter, you need to play on the server," so to speak.
With much regret, I have to say that until I consolidate the remaining loose ends on the server, I cannot reveal the next shifter. However, that doesn't mean I can't talk about some of the work that's been done behind the scenes.
2.1 Current State
So far very good, the few people who have seen the animations have been very impressed. The quality we are looking for is excellent, and the shifter has been playable for some time with it's basics abilities (walk, run, jump, rest, die, born). My "Done" flag for this version will be when I feel happy about a improvement in the feeling when controlling the shifters, I've been working in by revisiting the movement and the addition of well requested cooldowns for the abilities. I have already some ideas in mind, some of them already implemented, and I would like to explain in further detail.
2.2 Movement Improvements
I call them "Motion" and "Shifters Control", the first one refers to every movement Shingeki no Craft can apply to any entity (A bullet, a Titan, A Soldier, etc...), is one if not the most important system on the datapack. On the other hand, "Shifters Control" refers to how the player interacts in order to control their shifter, this aspect have always been very limited due the fact that datapacks don't have access to global key bindings.
Motion system is now instantaneous. It calculates and assigns movement at the moment the command is executed using a macro. Previously, the system would wait one tick before performing the calculation. Aside from improving precision, this change also allowed me to remove a check that was in the main tick function. If there's one function where you want to remove as many checks as possible to prevent overflow, it's this one, as it processes all entities. I’d expect this change to also result in better performance.
Shifters Control was my Christmas gift from Mojang. In 2.2.3, they added a predicate that allows for precise detection of vanilla keybindings (running, moving forward, left, right, etc.). This made it possible to greatly simplify the implementation for detecting movement and even expanded its functionality in some ways. For instance, you can now detect if the player is running, jumping, or walking while riding a vehicle (e.g., a skeleton horse). This made it possible to highly reduce the performance impact on movement triggering, remove any delay from the controller and enable players to toggle whether to run or walk in shifter form by pressing Ctrl.
3. Bitter Sweet
I wanted to create a server compatible with both Java and Bedrock editions. That’s the main reason for the massive delay on updates this month—this post was originally supposed to go out at the end of November. We knew it would be a challenge, but I pushed for it to surprise the many fans who have been asking for a way to play on this version. Unfortunately, it’s simply not possible right now, and I’m not sure if it ever will be.
I set up a proxy using Geyser, a plugin that, as I understand it, simulates a Java client on the server and sends packets back to Bedrock. I assigned the necessary ports, tested performance, and planned out the viability of this idea. It had potential—the server could have a steady flow of active players—but we fell short at the very last step.
Right now, I don’t know how to convert the Java texture pack (.zip) into Bedrock format (.mcpack). I’ve researched various sources and tried different approaches. I think the resource pack we have is so complex that an automated solution, one that doesn’t require manually go through each file, isn’t feasible at this time.
We’ve decided to move forward with Java Edition only. We’ve already spent enough resources on this initiative. There were some significant advancements, and I feel that I learned a lot in the process. I guess we’ll see what happens in the future, but right now I can’t promise anything related to this.
4. Final Thoughts
Overall, I’m happy and quite eager to see the results when everything we’ve been working on comes to light. Despite the ups and downs I mentioned, I feel like I’ve challenged myself to learn things I never thought I’d need, even if I can’t take full advantage of them right now. Besides that, I’m more busy than I've ever been—it seems like every day I have more and more on my plate.
I’m planning to upload a video in January, after the server officially launches, then I’ll talk more in-depth about my situation (SnC's situation). But who knows, maybe things will have changed by then.
If you’ve made it this far, thank you for reading. Sorry for the mess—I had to make time out of nowhere to write this, haha. I’ll be back to normal on the 26th. See you on the server if you want to join me for some building!
Happy holidays, everyone! Take care!
Comments
Death Note
i love you
