As we continue to look forward to rolling out Minecraft 1.13 on the Cemetech Minecraft Server, we are keen to bring you the best possible experience. Some of our most die-hard members no doubt know that we've encountered interest and occasional low TPS (ticks per second) during busy periods. We have done a great deal of investigation to try to narrow down the cause, so far with minimal success. "But did you try-" yes, we tried that. "Ah, but how about-" yes, that too. "I bet you didn't-" yes, we thought of that as well. Therefore, using our current Minecraft 1.12 server, we're going to apply the scientific method and determine which (if any) plugins are responsible, or if the builds and machines in the world are to blame.

But we need your help. It's summer, which means you no doubt are rolling in free time, so we need you to come visit our Minecraft server and continue working on your towns, your builds, your adventures, and your exploration.
  • Starting this Saturday, June 16th, we will start with Phase 1 by disabling some of our least-important plugins, primarily those that have some aesthetic effect only, or at most some minor effect on gameplay. We encourage as many of you to come and play as possible.
  • In Phase 2, starting next Thursday, June 21st, we will disable another set of plugins, and due to the set of plugins we're disabling, we'll be turning our current graylist into a whitelist (i.e., if you're not already a Member, you won't be able to log into the server).
  • Finally, Phase 3 will begin the following Tuesday, June 26th, during which we'll switch the server to pure Vanilla. It should go without saying that we expect the same maturity and mutual respect that is enforced by our plugins: we will have a zero-tolerance policy towards theft, grief, vandalism, or any rule-breaking, and you will be permanently banned from both the 1.12 and 1.13 (and later) servers.
Therefore, just in time for the hot summer sun and brutal humidity, you should get back into Minecraft, think of something impressive to build, or gather your fellow Cemetechians for some cooperative (or competitive) Minecrafting: you'll be helping us make the Minecraft 1.13 server as fast and enjoyable as possible. See you on the server!

I think this post should be renamed to "Cemetech Minecraft Server: Lag (and fun) Eradication".

Before you ban me from all things Cemetech, hear me out. If you disable all of the plugins, we just have a standard survival game (albeit with other players). No economy. No towns (I guess they are pretty unnecessary with a group of respectful individuals). No shops. Sure, there are people I enjoy playing with on the server. But the economy plugins keep me, at least, motivated to play. If there is no economy, then that makes it less fun for me.

Some of the commands are worth keeping as well. The home command lets players hop back and forth from their project to their homes (or otherwise, but this is how I use it). This makes building projects large and small so much easier. Without it, I will probably not travel as much. I would stay in one spot and just build a small cluster of buildings around my central house. I probably wouldn't go to the nether or to the end, because it is an unnecessary hassle to go from spawn to my place.

Let's talk about Eggcatcher. This plugin is probably the laggiest (because of people with too many chickens), but I think that it should stay. It lets us easily transport mobs. I put squids in the end a few times. How else was I supposed to do this? The other alternative would be to absolutely wreck spawn (which certain plugins disallow).

In my opinion, the towns can go. Even the map and stat collection things can go. They are cool, but are not essential to a good experience. But the Essentials plugin, Sign Shops (I think this is part of essentials?) and Eggcatcher should stay. These maintain the fun and encourage building, playing and more. Without them, in my eyes, I don't think I'll be online much (and neither will others).

Judging by your post it looks that you will be removing plugins and potentially adding them back later if they are not too laggy. I really, really hope this is the case. I am a relatively inactive player and I have a full 12 days spent on the server.

Crying
This won't be permanent; we'll add the plugins back after this period. We're determining where the lag is, not removing plugins until the lag stops and keeping the plugins that are left.

We're removing plugins in groups rather than one-by-one. Removing in groups lets us figure out our next step. If the lag stops in group 1, then we'll probably go through that group again but removing fewer plugins at a time. I think there's 10 plugins in Group 1, so we'd probably go back through and remove 2-3 at a time. Even if the lag stops in Group 1 we'll continue through until all the plugins are removed.

We're doing this so we can sort of pin-point where the lag may be coming from. We're removing all the plugins so we can also get a good benchmark of our hardware. We really don't think our hardware is the issue, but if we still get lag with no plugins, then that's important info to know. It's also good to know the numbers we should be seeing on our hardware with vanilla Minecraft.

Now, what if we kept 3 "absolutely necessary" plugins and we still got lag? What then? Is it one of those 3 or our hardware? Removing everything is the best way to eliminate variables.
Well, we do call ourselves a intellectual survival server. However, with all these plugins, we might as well have bubblewrap somewhere in that motto.

Getting mobs moved around requires some fun bit of trickery with minecarts and the like, but can be done without egg catcher.

The /home bit is meh. Honestly, I felt this way as well, but the more I've played in vanilla survival, the more I don't mind utilizing my elytra, or using a nether network, which gives you room to build cool things there, as well.
If you complete all the phases and determine it's not a plugin issue, I recommend using Carpetmod by gnembon as a fourth temporary phase. Its /tick health command can give info on whether most of the processing time for a tick is going towards block updates, entities, tile entities, or spawning. If it is mostly entities or tile entities, it can also determine which type (chickens, villagers, hoppers, beacons, command blocks, etc.) are causing the most lag. It also includes a setting to replace inefficient portions of the code (redstone dust, full hoppers, etc.) with more optimized versions, which could further show what is causing performance issues.

Also, since we are talking about Eggcatcher again, I agree that it is very useful. However, I consider the spawner thing a bug. I made a version of Eggcatcher which fixes it if you want to use it. I also agree that eggs were a poor choice for the material to catch mobs because the only way to get them in large quantities is to keep a ton of entities, and once you do so there is no "better" or "worse" design.
How is it a bug? It is a normal, intended feature of vanilla Minecraft.

If I wanted to create a build with emerald blocks, I could either set up a ton of entities, spend a ton of time, and create a villager trading thing and a few farms. Or, I could pop a vindicator egg into my spawner, which won't take up any more server resources than with any other mob, and grind for a few minutes as opposed to a few hours. This leaves me with more time to work on my build, which means that I can do something more interesting.

I do agree that eggs are a poor choice. I suggest spectral arrows, because they require crafting, can be fired, and really don't serve much of a purpose when hunting. They also despawn naturally, unlike the chickens spawned by throwing eggs.
It's not a vanilla bug, it's an Eggcatcher one. Checking the version history, the last major Eggcatcher update came out years before the vanilla feature of using spawn eggs to change spawners was added. The vanilla devs did not intend for spawners to be changeable in survival because spawn eggs are creative only in vanilla, and the Eggcatcher dev did not intend for spawners to be changeable because the vanilla feature did not exist when the plugin was created. Because neither set of devs intended for it to exist, it is a bug.

And my main issue with the spawners as far as the economy goes is that one player can get a small amount of any item with no effort. However, with emeralds for example, you could make a pumpkin farm with a stack of pistons that equals the speed of a vindicator spawner. This farm requires 2 entities, one trading villager and one hopper minecart, and the pumpkins and pistons themselves cause no more lag than an equal area of grass. This has the same lag impact as the spawner. The balance issue with the spawners is not speed, but the fact that because you can farm ANY item with no additional effort, it means that players are less reliant on the economy.

and now I am getting off topic again, maybe we should move the Eggcatcher balance discussion to one of the other topics I created for it
Just a heads up that the second group of plugins will be removed tonight. You will experience broken farms, dynmap will not work, and other worlds will not be accessible. Additionally, the whitelist will be enabled, if you are presently a member then you are on the whitelist.
commandblockguy wrote:
The balance issue with the spawners is not speed, but the fact that because you can farm ANY item with no additional effort, it means that players are less reliant on the economy.
Counterargument: rather than creating truly massive pumpkin farms and cadres of villagers, it encourages players to get creative with building machines like spawners that can easily be changed between light and dark, 1-high and 2-high, water or grass or neither, etc.

We believe that the current Phase 2 result may be a positive one (i.e., lag is gone), but we need more people on at once to verify.
The pumpkin farm is just one example. You can only trade a small subset of items in exchange for emeralds, so players wouldn't build pumpkin farms for everything.

In my opinion spawners actually limit the player's creativity, by allowing them to create a one-size-fits-all solution to mob drops. Yes, a basic spawner setup isn't compatible with all mob types right away, but it takes very little effort to change the setup to match the mob type. A spawner that can support any mob type at the push of a button does require effort to set up, but still encourages far less creativity than requiring completely unique designs for each type of mob.

For example, you can spawn and kill witches, guardians, and slimes, along with most other mobs, using the same machine which involves a spawner. This requires only one design, and there is very little opportunity for one design to have a speed benefit over another, as spawners all work at the same speed.

However, if the player does not use spawners to farm these mobs, separate farms must be made using totally different mechanics. The slime farm needs some way to attract slimes out of the farm and to their deaths. The witch farm forces the player to work with the constraint of a small spawning area. The guardian farm requires the player to remove an entire ocean monument and the ocean around it, and then some way to move the guardians, which move slowly in water, into the nether quickly to avoid filling the mob cap. Any one of these farms would provide as much opportunity for creative design as the spawner based farm.

This probably seems like far too much work for one player. You can't expect anyone to put in work for each type of farm, especially if they don't need chests upon chests of items from each. However, this is good because it promotes the use of the economy by making players reliant on each other for different types of items. Each player can build a few farms, sell the items from it to the whole server, and buy items that they don't produce from other players.

Once again, we should probably continue this in one of the other threads, as this is getting very far off topic.

I am interested to hear the results for the lag testing. I assume that you will narrow it down to one plugin now that you have determined the group?
I do like the way that you think commandblockguy. The way that you treat the digital minecraft world like the real one. I do however see that as both an addition and a fault to your logic. I think that spawners are great as well, mostly because of one simple reason:

People do not like to farm their entire playtime.

Or even most of their play time. I mean, correct me if I am wrong, but I think most players would much rather be doing anything but tend to large farms. By having spawners, this elimanates the issue of tending to large farms and makes players able to get profit without the tedious work. But, this also allows large monopolies to form and creates an unstable economy. One where mutual benifit does not always exsist.

Solution: I believe that, in order to maintain a healthy server, a trade off of the two is required. Perhaps players be allowed to own one or two spawners and if they wish to expand further, they must devise ways to use the enviorment and grass.

Do whatever you wish, but that is my viewpoint


EDIT: Should have made this clear... I am an outsider looking in Very Happy. I have never actually played on the cemetech server (although I should). I was just giving my two cents but now that I see that things are not as I saw them, the above post stands irrelevant.
LAX18 wrote:
Solution: I believe that, in order to maintain a healthy server, a trade off of the two is required. Perhaps players be allowed to own one or two spawners and if they wish to expand further, they must devise ways to use the enviorment and grass.

Do whatever you wiah, but that is my viewpoint


What would happen to large spawners like the BM one, which is public (maintained almost entirely by myself) and contains five spawners?

Centralization is good because it means that you can have one open facility as opposed to multiple facilities. This means less chunks loaded (if it is shared, the chunks loaded by a player existing will overlap), and on an active day it can mean less spawners.

I'm not saying the server should become socialist. But for some of the laggier farms, this is a good option, I think.
Well, the farms that I was referring to are ones that run automatically, without player intervention. This type of farm requires players to spend a lot of time designing them initially, then run without needing the player to do much manually afterward.

Spawner blocks have the restriction that players must be within 16 blocks for them to work, which is another factor that makes them more suitable to getting a small amounts of any item. Many automatic farms do not have this restriction, which means players are not required to spend all of their time doing one thing.

And currently, I don't feel that there is a monopoly on the server at this point in time. I feel like the opposite is true, where there is little service one can provide that is anything more than a convenience. Any player can farm items for themselves rather than buying them from someone.

EDIT: Left this open for a long period of time and iPhoenix added a reply between the time I last previewed the draft and when I actually sent it, so here's my response to his post:

How does the amount chunks loaded relate to this? If you have two farms, only one player can load each at a time, regardless of whether they are spawner farms or vanilla farms. If you mean that one centralized location has all the chunks overlap, then you have a bunch of farms loaded when you are only using one of them, causing a ton of lag. If you mean that all the spawner blocks are in the same location, that's really only one farm running at half speed for each person.
commandblockguy wrote:
If you mean that one centralized location has all the chunks overlap, then you have a bunch of farms loaded when you are only using one of them, causing a ton of lag.


That's what I took it to mean and I share that opinion as well. I believe it's much more beneficial for the server to have things spread out. If each person makes a farm that can sustain themselves for one or two things (and a shop), Point 3 in this post, then lag would be minimal and we'd have an incredible economy because players will be farming items as needed.
That's not what I meant. I'm suggesting that we have decently sized community farms. Of course, these would not be close together for the reasons stated by commandblockguy. I'm saying that each farm itself would be centralized, as opposed to, say, 4 different server members having their own farm. The goal here is to significantly decrease the need for every single player to build their own farm, because there will be always be the time where you overestimate your need or your need decreases, etc. Instead of building this in your base, which will be loaded pretty much whenever you are online, why not have fast and efficient farms that are not loaded when not in use.

The enderman XP grinder is a great example of this (though it is mostly by virtue of how it works, but that's besides the point). It is away from any activity, and lets players gather exactly what they want (xp/enderpearls) very quickly.

Switching topics slightly, I don't think such an economy will happen, because people want to sell more than one or two things. Competition happens.

Command stated this really well.
Quote:
And currently, I don't feel that there is a monopoly on the server at this point in time. I feel like the opposite is true, where there is little service one can provide that is anything more than a convenience. Any player can farm items for themselves rather than buying them from someone.
(my bolding)

Also, let's say that one week I am completely out of inventory in my, say, pumpkin farm. To meet this increased demand I might take the time to expand my pumpkin farm.

But someone only needed a few stacks of jack o' lanterns to light up a project, so my increased supply is met with nothing.

In an ideal world where I have infinite time on my hands and never get distracted, I would probably take a few seeds out of my farm. But there are things to build! People to play with! Mobs to kill! People to kill! Shops to shop in! Shops to kill! (Maybe I got carried away)

If the idea of centralized farming gets tossed out, what if the number of hoppers users could craft and the number of shops people could have were limited? This would obviously reduce lag (nobody would play). It means that users couldn't farm everything, unless they were really clever and organized, and even if they did, they can't ruin the economy by selling it all.
I agree, we shouldn't keep farms in players' bases. However, this doesn't mean that they all need to become community use just because they are not located near the player who built them. Without changeable spawners, it is difficult to build a farm for a particular item. It would be impossible for one player to make an automatic farm for every item, or even a majority of items. There is also less incentive to build a farm that already exists on the server than one that hasn't been built before, as competition reduces the profit you can make. The difficulty in building farms, rather than some arbitrary restriction on shops or hoppers, should prevent players from farming everything. Spawners (as they exist now) provide very little of that difficulty.
There is a potential serious issue with switching to pure vanilla: For aesthetic reasons, some areas depend on Towny to prevent mob spawns instead of using light. While the Bazaar is the first area that comes to my mind (incidentally, my internal base is actually lit-up for no mob spawns), I know there are other areas. I think, for example, Turq's base is pretty dark, too.

Are players on their own to clean up all creeper damage? Will the server be fully or partially reverted?
commandblockguy wrote:
Spawners (as they exist now) provide very little of that difficulty.


I do agree that spawners decrease the difficulty in gathering resources to build farms. But grinding for redstone in the mining world is not difficult or engaging. It's tedious time-consuming, and does not foster creativity. I'd even venture to say that it limits creativity because I'll spend all of my time mining (either for lapis to buy it or directly) instead of designing, planning, and building something creative/interesting.

Quote:
The difficulty in building farms, rather than some arbitrary restriction on shops or hoppers, should prevent players from farming everything.


This is most definitely better than the idea that I came up with.

However, I do not think that forcing players to grind is the right idea. On a SP world, a middle to endgame player will probably have a witch hut farm providing you with a steady stream of redstone/gunpowder/glowstone.

Because there is only one witch hut on the server and it is not open for public use, spawners replace this. To use my example from one of my previous posts, instead of building a large villager trading hall that contains lots of laggy (tile-)entities, I can use a vindicator spawner, which will get me a decent amount of emeralds in less of my time.

A spawner is not entirely OP. The spawner requires you to build it! In the case of the BM spawner, which originally had three spawners barely in range of each other that were really close to bedrock and the ocean, this was a big challenge. I still spend the majority of my time on the server maintaining the spawner.

Quote:
However, this doesn't mean that they all need to become community use just because they are not located near the player who built them.


Of course. People should still reserve the right to build their own personal farms. They put the effort into building the farm, they should be able to decide the rules for it.

I see where you are coming from, and I agree with most of the things you are saying. blah blah blah survivorship bias blah blah blah

If spawners appear to be to OP, I would not be opposed to a plugin or otherwise that reduces spawn rates.
Quote:
But grinding for redstone in the mining world is not difficult or engaging. It's tedious time-consuming, and does not foster creativity.

So why not farm redstone then? Razz

More seriously, I was not talking about the resource difficulty while building farms, but the creative one. All spawner farms function the same way, and there is no real purpose to using obscure mechanics to improve the farm. On the other hand, farms using vanilla mechanics all have unique restrictions from one another, encouraging creative design.

One of those restrictions is materials, but that basically just means that you can't make the best farms right away. You can start off with a slow or manual redstone farm, and using the resources from that you can work your way up to a more complicated one.

I see your point with the witch hut and other limited things like guardian temples. Hopefully with the next map we will have more area to build. If not, maybe permanent spawners for witches and guardians could spawn underground the way zombies, skeletons, and spiders do now.

It's not that the spawners are OP and need to have their speed reduced. Most vanilla farms are much faster than spawners. In my opinion the inbalance comes from the wide range of items you can get from one spawner farm. If you cut the speed in half, players would spend as much time building the farm as before, and get the same range of items, but would need to spend twice as much time grinding for items.
I like your idea, but to retain the properties of spawn eggs as tradable items maybe they could spawn as dungeon loot.

Unfortunately, most people creating a farm are not going to design it themselves. They will use a tried and trusted design found online.

This isn't really creative.

I suppose that if I did not have the resources I would probably try to make my own with what I have, but usually the result is slow and not optimal for lag.
  
Page 1 of 2
» All times are UTC - 5 Hours
 
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum

 

Advertisement