I am quite interested in this as well, I can understand it starts small but would love to see this game grow to see huge kingdom sized level warfare
I am quite interested in this as well, I can understand it starts small but would love to see this game grow to see huge kingdom sized level warfare
My dream is to make my settlement long outlive me Why not join me in being part of it? Send me a message
9B7323 friend code
Posted By *Spenraw * at 1:36 PM - Wed Oct 02 2019
I am quite interested in this as well, I can understand it starts small but would love to see this game grow to see huge kingdom sized level warfare
I hope someone with this knowledge can give us a rough idea. I understand that thousands of people in the same area will be a performance disaster. However they have not spoken (to my knowledge) on whether or not large scale battles and sieges will be instanced.
This is one of the big questions I have about this game. Especially as (and I may be wrong on this) the UE really doesn't have a track record of supporting a large number of players fighting together. Pretty much the reason why games focusing on mass pvp like Camelot Unchained are building their own engine.
Like Unity the UE code was not written for a large scale mmo. This can be remedied by rewriting these bottlenecks to open them up to eliminate that problem. For professional coders this is not a big deal and shouldn't be something to worry about.
Posted By Belteyn at 3:27 PM - Wed Oct 02 2019
This is one of the big questions I have about this game. Especially as (and I may be wrong on this) the UE really doesn't have a track record of supporting a large number of players fighting together. Pretty much the reason why games focusing on mass pvp like Camelot Unchained are building their own engine.
a concern of mine as well.
Some of these answers cant really be known till the stress tests at the very end. You can do math on paper to estimate but those can be very rough guesses.
Posted By Xonth at 12:52 AM - Fri Oct 04 2019
Some of these answers cant really be known till the stress tests at the very end. You can do math on paper to estimate but those can be very rough guesses.
Disagree. Something as vital as how the number of characters fighting together affects performance HAS to be designed for right from the start. It's too big too just leave to chance during stress test i.e. "well we kinda hoped 500 player battles would work well, but now seems any more than 20 players won't really be viable".
Posted By Belteyn at 10:04 AM - Fri Oct 04 2019
Posted By Xonth at 12:52 AM - Fri Oct 04 2019
Some of these answers cant really be known till the stress tests at the very end. You can do math on paper to estimate but those can be very rough guesses.
Disagree. Something as vital as how the number of characters fighting together affects performance HAS to be designed for right from the start. It's too big too just leave to chance during stress test i.e. "well we kinda hoped 500 player battles would work well, but now seems any more than 20 players won't really be viable".
have to agree with this. If it isnt one of your main coding concerns your doing something wrong. While stresstests will provide alot of information to better things it isnt a way to decide do we need to make an own piece of code for this yes or no.
In the recent MMOs Life is Feudal had the, in my opinion, best combat mechanics and wasn't even able to support more than 80 people without the servers lagging (even if they were coming from the DEVs). The Tourque Engine they used was non the less free to use and was said to sucesfully support at least more than 200 people without laggs, coming to an end a bit over 500 with server performance (Tourque forum devs and progammers said so). So LiF was bad coding overall because the DEVs were laymen in the beginning (which they stated at this point :D)
The trick in the UE engine is multithreading what multiple developers do not really use, since it's difficult, hard to implement and also hardware related. So if there's a sudden, important change in hardware, the code needs to be rewritten in the worst case. The engine itself is super powerfull but the base is for looks and easy routine stuff. They also implemented the blueprint system for easy object focused programming. The problem is, that with each blueprint gear into another one, the code becomes bigger and bigger and calculations become exponentially slower.
So in the end, it depends on how much they implement for combat, how many parameters interact and how many can be changed by the player in x-time and which effect it has on the combat. Also server speed, interaction speed..... Their idea about instanced battles for sieges says alot about SBSs expectations of server performance for a lot of people.
I will be absolutely surprised if more than 100 players can be "on the field" without most players lagging in some fashion.
Large scale multiplier is one of the greatest challenges all previous as well as current MMO game dev teams have faced, with most never overcoming it.
It really needs to be built in at the very beginning, then tested and retested regularly throughout the rest of development to ensure any added functionality doesn't negatively impact performance.
As I've not seen any evidence of such work (unlike other titles such as CU, CF, and SC) I'm not very hopeful for COE to support large scale PVP.
You must all be quick to listen, slow to speak and slow to get angry. JAMES 1:19 NLT
If I recall correctly the battles will be instanced and participants will be invited by the leaders who will have access to a tabletop UI from which they will enter orders.
Is there any dev journals written about this or any that mention numbers and the process thereof? (whether or not it will be staged)
Posted By *Spenraw * at 1:36 PM - Wed Oct 02 2019
I am quite interested in this as well, I can understand it starts small but would love to see this game grow to see huge kingdom sized level warfare
Because there is kingdoms in the game that contain thousands of people, I would hope that they would make it playable somehow when at least a couple hundred people are involved. Otherwise it seems like we are taking it to one extreme in terms of number of players involved on every other part of the game. Then when it does come time for people to fight the maximum number of players involved suddenly dips down into a small fraction of all other situations.
Honestly if it came to 50v50 I would be pretty disappointed, instanced or not.
From what I can tell theyve been dodging this question as it's been specifically asked alot with no real estimation or answer.
I don't know anymore.
Posted By Leiywen at 04:36 AM - Fri Oct 04 2019
Posted By Belteyn at 10:04 AM - Fri Oct 04 2019
Posted By Xonth at 12:52 AM - Fri Oct 04 2019
Some of these answers cant really be known till the stress tests at the very end. You can do math on paper to estimate but those can be very rough guesses.
Disagree. Something as vital as how the number of characters fighting together affects performance HAS to be designed for right from the start. It's too big too just leave to chance during stress test i.e. "well we kinda hoped 500 player battles would work well, but now seems any more than 20 players won't really be viable".
have to agree with this. If it isnt one of your main coding concerns your doing something wrong. While stresstests will provide alot of information to better things it isnt a way to decide do we need to make an own piece of code for this yes or no.
There is a bit of a difference between designing to handle large groups (in battle or otherwise) and knowing precisely where the tipping point is for performance.
I would be curious to hear how SBS has been aiming to handle large crowds, but it would not surprise me if they were reluctant to speak to a specific number before having full assets in place (pretty sure multiples of the same asset would be easier than the same number of different assets to handle in any rendering engine).
