War - What is it good for?!

[Destroyer_Bravo 1054]

so basically you've set up a gigantic gwent board with no hand size limit and a draw equal to potential rally size

[Fireheart 2777]

Thomas has been hard at work fixing DD! You can ask him for more information or give him suggestions on how it should be changed.

 

7 hours ago, The Elfen Lie said:

Also with the increase of allowed raiders, will the raid cooldown on settlements be slightly increased? For now it can make it more annoying when getting raided.

We will be testing this first change and then asking for more feedback and  the opinions of the playerbase after awhile. Already these set of raid rules are still highly watered down compared to past versions. Though, if this proves to be an issue we might change raid rules to be based off of two different sets. One for war times and one for peacetimes.

 

7 hours ago, Freya  said:

Still want to know why exactly we have to make more and more compromises in order to have warclaims. I'd rather compromise on professions or [insert bloatware plugin] if it meant we could have the glorious warclaims of old.

 

This is just depressing to see.

 

Stop playing footsies and tell us why we can't do warclaims like we have for the last four years, please, whilst I still have a shred of confidence in you guys left.

As Minecraft updates we are forced to adapt and change how we work warclaims. This is completely to blame on the way the program is gamed. Packets is a form of data that players send between one another in order to see moves and actions; ever since 1.9 we've had more packets come on from additions to Minecraft combat such as stamina bars, swinging, etc. Plugins is only a part of what makes the server lag and all we can do is ride the wave and hope we can make up solutions to our problems. Telanir has also outlined this to the community within his address of warclaims! -Was just adding on to what Dan has already stated.

 

4 hours ago, Pess said:

Was a post made by staff prior this decision then? If it was not known it was being changed no one can give input. This is kind of lame reply to ust point at the feedback forum.

Yes, there was a bit of a warning to all of this and a lot of feedback was given. This concept was taken from feedback suggested and discussed at several meeting.

Warclaims have always been an issue on the server as to "What system should we use?" "This one isn't working." "This one is too complicated." etc. Cosmic and I when we sit down and discuss these systems try to bring to mind one goal. Keeping it simple. All in all some people might feel this is a bit complicated, refering to iMattyz's comment above, however it is quite simple. Same war rules apply and the only difference is you'll rally in a room and sign up for the warclaim. This is no difference than signing up for an event or tournament on an RP forum post. Same idea with the hope of spreading out players to generate less lag. This has never been tested which is why I mentioned that we intend to do so with the next war, and if it proves to not work out we will look to try something else!

1 hour ago, Shalashask said:

Won't this just be abused by people who are intending not to turn up or are inactive signing the roster? As one of their friends fighting on another day can just take their place.

In terms of numbers this will not hurt the system's intended purposes. We hope to curve numbers down so instead of having giant 100v100 battles we have 30v30 instead. If someone can't come due to IRL reasons and shows up late someone might take their place, that does not hurt the warclaim whatsoever and allows that person who couldn't show up to possibly swap out spots for the person who took their place. That way they simply switch the intended battles they meant to go to, but still both fight.

 

33 minutes ago, SquirtGun said:

Did we think about waiting for 5.0 and seeing how the lag is first?

 

I thought the whole "world sharding" was to prevent large maps and lower lag to allow for a lower-lag environment.

 

Shouldn't we see how that goes, then idk do a server wide war-game or something to test the impact it has on lag before redoing the entire system?

This goes beyond the way we divide server resources. Our issue with Minecraft PvP is a networking issue due to the sheer amount of packets that have to go from the player to the server to the rest of the players. Each packet as a set amount of data it contains and when you have 10's of thousands of these being sent in a warclaim, it causes lag. Of course there are server side issues to it, but Minecraft restricts us slightly on what we can do due to how the game was made. It is a valid point and come 5.0 we can definitely test it for fun to see how it works, but I rather not hold up roleplay for a map still in development. This is meant to be a hotfix that allows warclaims to recontinue for those interested. 

[lemontide331 1615]

how come lotc is the only server that seems to ever be affected on a large scale by minecraft updates lol 

[Neri 3590]

5 hours ago, Pandan said:

 

This factor has been explained multiple times, now I'm not a dev but I'll at least try to give insight into why this occurs. And, well, its mainly our platform (minecraft), map and godknows what else. In actuality the nexus stuff barely is a fraction of what, for instance, swing timers added in terms of lag. Unfortunately even 'removing' swing timers wouldn't throw aside the lag as all that does it set them to a minuscule amount of time, still producing those damned packets.

 

Then we've also got the map combined with minecrafts god awful resource.. splitting? spread? We've just got over 30,000 entities constantly loaded on this map alone. Its.. pretty damn harming, compared to lets say for instance PVP servers in which they've got maybe 100 assets at most on a single world. We're trying to process 300x the amount of load other servers are, and thats also a large contribution as to why we lag so god awfully. 

 

Right now we're just trying to find solutions that'll at least give you something to do in the mean time. There isn't any quick solution to fix the lag, but in reality we're dependent on so many external sources (Mojang, Spigot, etc) that due to the nature of our server, which minecraft isn't nowdays explicitly designed for, will very much cause excess lag. 

 

Edit:

 

 

Thanks you for being honest and answering my questions.

 

And yes I agree, removing swing timers wouldn't remove the extra packets causing the n^n problem whenever you add a new player into the equation but it would at least help the players not feel the lag quite so badly.

 

If entities are causing a problem, why has little been done to cull them? I still see hundreds of owned horses roaming the map, I see thousands of boats floating in the wilderness, I see farms with 1000+ animals within them vibrating into some form of animal singularity. Is the amount of entities we've got on the map not being compounded by the fact that nexus is always actively tracking and effecting them (eg: animals still starve in unloaded chunks in my experience). Surely a good cull would help the server a bit? What about making a fresh and smaller mineworld (it's still half untapped).

 

I understand that minecraft isn't really designed for what we aim to do (see mojang adding minigames as a vanilla thing to windows10/console editions) but i'm sure we can try to trim the fat so we don't need to turn warclaims into a WoW battleground, that's just miserable.

[zaezae 1098]

Honestly, why not invest in a separate machine, lift a portion of the map where the fight will happen, remove entities and make people log on there to fight on 1.7 or whatever version of minecraft did packets correctly, and just grant people iron armor and swords vanilla?

 

 

[Aislin 4354]

This is exactly what people wanted. More chances for GM bias and/or errors to dilute the result of a war between factions! The solution isn't to make warclaims less fun so that they work, it's to find other ways to make them work (such as cutting down on map size or plugins used).

[Pandan 1452]

21 minutes ago, zaezae said:

Honestly, why not invest in a separate machine, lift a portion of the map where the fight will happen, remove entities and make people log on there to fight on 1.7 or whatever version of minecraft did packets correctly, and just grant people iron armor and swords vanilla?

 

 

 

Because the cost isn't something we can endure right now, unfortunately. Our finances are not 'code red' status however they're also not the strongest they've been since the massive forced re-shuffle with the ol' perks. I'd love to be capable to dropping £100-£200 a month to get something like that operating, yet anything lower (such as the build server, ask anyone working on 5.0 how bad it can be) would simply not be capable of handling the 200+ potential players. 

 

Regarding 1.7, its been that long it'd just be very broken I imagine. I can't say for certain however its very difficult to get legacy versions working across entirely new windows updates, computers, graphics cards and even server software. Again, this is me presuming. I'm here to talk not to.. code. 

[Runabarn 1028]

While you're at it, mind stating in a clear and concise manner what exactly constitutes a raid? There is no official statement anywhere and there has been confusion over it.

[zaezae 1098]

Just now, Pandan said:

 

Because the cost isn't something we can endure right now, unfortunately. Our finances are not 'code red' status however they're also not the strongest they've been since the massive forced re-shuffle with the ol' perks. I'd love to be capable to dropping £100-£200 a month to get something like that operating, yet anything lower (such as the build server, ask anyone working on 5.0 how bad it can be) would simply not be capable of handling the 200+ potential players. 

 

Regarding 1.7, its been that long it'd just be very broken I imagine. I can't say for certain however its very difficult to get legacy versions working across entirely new windows updates, computers, graphics cards and even server software. Again, this is me presuming. I'm here to talk not to.. code. 

I understand, but the server itself is red-lining on PVP. I know that is stating the obvious. The only other thing besides a WC system that might not help (but might help) is to, somehow, switch to a version of the map with zero entities. No signs, animals, mobs, etc.

[Pandan 1452]

5 minutes ago, zaezae said:

I understand, but the server itself is red-lining on PVP. I know that is stating the obvious. The only other thing besides a WC system that might not help (but might help) is to, somehow, switch to a version of the map with zero entities. No signs, animals, mobs, etc.

 

I feel that's a very last resort type of thing, at the end of the day RP will always take prevalence over PVP. There comes a point when amending these issues when we have to consider if the cost to RP, our main purpose and goal, is worth the benefits to PVP.

 

This isn't me stating I despise PVP or anything, in reality I dont particularly mind it. Even if I am bloody awful at it. Nevertheless our target market is the roleplay community, sometimes I think we come close to forgetting that, playerbase and staff alike. 

[Naj 545]

The decision is one that is not ideal. We all desire large war claims and epic battles. However, the fact is, that the staff team is faced with a in the problems they've outlined in their descriptions of packets. I think the only thing we can do as a player base is to continue to move along with the systems implemented and, after they've had some time to test out, create feedback to either change, or improve that system.

 

In regards to 1.8, the war claims were hardly better. The lag would clear once a decent number of people had died. However, the beginnings of the battles were unbearable. The last war saw suggestions similar to this from both sides as a way to fix the issue of lag. 

 

So **** it, lets try it.

[argonian 14254]

9 minutes ago, Joe / joenaj / Yemekar said:

In regards to 1.8, the war claims were hardly better. The lag would clear once a decent number of people had died. However, the beginnings of the battles were unbearable. The last war saw suggestions similar to this from both sides as a way to fix the issue of lag. 

 

So **** it, lets try it.

They had just fixed 1.8 warclaims right before we updated. We had a massive test fight that went very smooth with shittons of players.

[Telanir 6975]

1 hour ago, Freya  said:

 

Thanks you for being honest and answering my questions.

 

And yes I agree, removing swing timers wouldn't remove the extra packets causing the n^n problem whenever you add a new player into the equation but it would at least help the players not feel the lag quite so badly.

 

If entities are causing a problem, why has little been done to cull them? I still see hundreds of owned horses roaming the map, I see thousands of boats floating in the wilderness, I see farms with 1000+ animals within them vibrating into some form of animal singularity. Is the amount of entities we've got on the map not being compounded by the fact that nexus is always actively tracking and effecting them (eg: animals still starve in unloaded chunks in my experience). Surely a good cull would help the server a bit? What about making a fresh and smaller mineworld (it's still half untapped).

 

I understand that minecraft isn't really designed for what we aim to do (see mojang adding minigames as a vanilla thing to windows10/console editions) but i'm sure we can try to trim the fat so we don't need to turn warclaims into a WoW battleground, that's just miserable.

 

I personally run around as often as I can taking care of stray boats, overpopulated farms, and etc.—unfortunately they are repopulated within days.

 

Unlike run of the mill Bukkit plugin dev's, every decision we make here we have to consider its effects on server performance. Nexus does not compound the issue, for example, the crafting subplugin only does minimal calculations and these occur only when you access the actual workbench, and causes absolutely zero lag in the event that no workbenches are being accessed at that moment. Delegating calculations to moments of access or on asynchronous threads means that the main thread (that if slow, is responsible for lag) is interrupted as little as possible. What you should take from this is that our custom plugins are not part of the issue: removing them to solve lag is like chipping away at a brick with only your nails.

 

I'm on vacation this month so I won't be too much help at the moment, but the rest of the Dev team I know for a fact is finding solutions to the performance issues. I've already found and resolved about 17.7% of the lag we were experiencing two weeks ago (according to our profiler), but we've got a ways to go.

 

Let me know if that answers your question, thanks Freya.

[Birdnerdy 1539]

Two options:

 

We could invest in a "legacy" 1.7 server where all players would receive the same kit for pvp. As far as I know, versions prior to 1.8 did not lag very much during warclaims

 

We could also consider reaching out to a large pvp server and asking them if we could use a large server of theirs for warclaims. 

[Fireheart 2777]

44 minutes ago, Whimsylicious said:

While you're at it, mind stating in a clear and concise manner what exactly constitutes a raid? There is no official statement anywhere and there has been confusion over it.

There is a very clear outlined statement of the definition of a raiding under war rules. I have underlined the section you are questioning in red!