Sponge in future


#1

Hello dear members of the sponge forum.In this post, I do not want to offend anyone, and in no case do not want to insult the creators of the sponge.
I am already familiar with the sponge for more than 1 year, starting with version 1.10.2. I use a sponge. After the first attempts to make my server stable, I left it. And every few months I returned to him and tried to fix the problems. I have never encountered so many problems that are very difficult to solve or even impossible. Since I have been working on servers for 6 years, I have something to compare with (starting with version 1.2.5). And I want to say that the sponge is far from ideal. Recall the same Cauldron with whom there were almost no problems. And the worst of all is that no one helps in solving these problems and new builds are very rare, even though there is still a lot of work with the sponge.

The sponge has a lot of potential, but you need to finish it. I apologize if I offended anyone. This is my opinion, I hope there will not be very many bad messages here.


#2

The main issue with people using SpongeForge is the enormous range of Forge mods it can be used with. Many mods do unexpected, and sometimes inappropriate things. This was always a problem with Cauldron too, and that was made more problematic by using Bukkit plugins that made different assumptions. That is the main reason Sponge was created; to manufacture a new API that allowed plugins to play nice with Forge mods.

However, there are always new mods appearing that do things in a different way, and trying to capture all of those expectations creates difficulty for Forge development, as well as Sponge. SpongeForge does work well with a wide range of mods, particularly with Minecraft 1.12.2 (which has the most developed version of SpongeAPI to date).

If you have gripes with mods malfunctioning with SpongeForge, it’s worth getting into the specifics. We’d like to know the details so we can get to work on what needs to be fixed - whether it be in Sponge, Forge, or a third party mod or plugin. Moreover, we do have a specific place for bugs and issues to be reported (github), and it is actively maintained by our issue wranglers and developers. If you really want to help, check if your issues have already been reported, and if not, then prepare a detailed report so people can look it over.

Remember, Sponge is an open-source, permissively licensed community project, and relies on the contributions, time and goodwill of volunteers. It’s shape, size, and features are all the result of collaboration from a wide group of developers. It will never be “finished”, unless development of Minecraft itself ceases.


#3

To add on to what @Inscrutable said, it’s very easy to get an inflated idea of how many crashing bugs there are in SpongeForge from looking at the number of crash reports on the forums. However, such an idea comes from a lack of understanding of the internals. Forge is an API with a defined way of doing things. But mods work with the base game itself, and can modify absolutely anything absolutely anywhere, and if they want to do things their own way there’s nothing Forge will do to stop them. SpongeForge however needs to hook into nearly every area, and provide accurate tracking for all events in those areas, and as such enforce Forge’s implied API constraints. If a mod is playing fast and loose with the rules, and doing things in a way it shouldn’t, SpongeForge will crash it, or the other way around. The other most common way for something to crash is when a mod is trying to modify base Minecraft code by rewriting the bytecode at runtime. SpongeForge does an enormous amount of this itself, and if a mod blindly writes ASM instead of using mixins, or uses a mixin to completely overwrite a specific method, it’ll be incompatible with SpongeForge’s modification to the same method. Some injection additionally won’t be compatible unless SpongeForge’s version gets loaded first, which requires renaming the jarfile because Forge loads mods in alphabetical order.

The key takeaway here is that there’s startlingly few places where SpongeForge is incompatible with mods. The real issue is when mods are incompatible with SpongeForge.


#4

I totally agree with you. But I use the same mods (even fewer mods) as on versions 1.8, 1.7, when I have not used a sponge yet. And then everything was fine. I do not argue, could add something new there, which causes problems. This is normal, but at least someone should do it out of the sponge, because the creators of mods do not take the sponge seriously and they always say that it is their fault.


#5

Huh, I had nothing but an unpleasant experience with Cauldron - its predecessor was fairly stable though (at the expense of a lack of mods though I guess).

There are a number of Sponge developers who have done extensive work getting mods compatible with Sponge (its worth noting that’s the effect of the work, not the actual work - mods that are incompatible with sponge and require changes, will be incompatible with other mods too), such as Aaron, through the form of communications and contributions to those projects. Its in our interest to have and maintain compatibility with a wide range of mods.

The trouble is, that a lot of mods are not designed with compatibility in mind - frequently doing things that harm its own compatibility with other mods! However, since any issues would only be noticed when 2 conflicting mods are installed - its sometimes difficult to pin blame. Unfortunately, some modders don’t see reason and have been hostile towards communication or contributions, to resolve their issues -_-

It sucks, but yes - some mods cause incompatibilities with sponge. However, as I say - we want to have high levels of compatibility, so do report incompatibilities and someone will look into them :slight_smile:


#6

tell me better where can i get help with my server? And it’s possible in general that with a minimal set of mods (where is one of the main nps) I can keep 100 people without lags? Because I already do not believe in it, since my server at 10 lags.


#7

If you’re after a 100-person Modded server, then you will have to make some compromises. Mods add a lot more load to the server, and it’s much harder to reach high player counts without serious impact on tps. Sponge does offer some solutions there, such as lowering entity-activation-range, tileentity-activation, and mob spawn limits (amongst other optimisations), but the most effective way is usually through lowering the server view distance (and/or pregenerating the world).

In general, you can get help just by asking here, or on the general Sponge channels in IRC and discord. However, general advice is unlikely to be a lot of help sorting out issues with a particular mod set. Be specific, and look at things like Sponge timings or Spark profiling to see where the impact is.


#8

I already somewhere three themes created about this here. One in the gitlab, one in Russian spunge and in the discord wrote.