A way to potentially fix servers is to dedicate server capacity based on players on said server (except USW2 and any other trading servers) but make sure that each server gets a base (reduced from current) amount of server space. And have a proportional amount of server capacity added to each server for each player who is 25+ stars and moves a certain amount of space/has some stats maxed/gear (to prevent someone from flooding a server with bots ect). Since each server (I belive) gets the same amount of server capacity EU servers during NA time are ghost towns and it’s a waste of space. Since Halls sometimes has 150+ (record was 200ish) people on a single server it causes massive amount of lag. This could be a way to prevent lag deaths and make the game better to play for people.
Not sure if it’s just me, but I’m having some difficulty understanding what you’re trying to say. So I’m gonna summarize it as my interpretation before talking about it. You can correct me if I’m wrong.
Moving players with 25+ Stars, good gear, and maxed stats are considered “non-bots”, while players with less than 25 Stars, worse gear, and no maxed stats are “bots”.
For example, if ten 25+ are in Server X, the capacity depends on that. So if the ratio between “non-bots” and server capacity was (let’s just say) 1:2, Server X would have a capacity of 20. With the extra 10 slots, “bot” players can join.
I see a lot of things wrong with this.
First of all, DECA wouldn’t go through all of this trouble, at least until Unity is well-established. It’s so much simpler to have the same server capacity for all servers world-wide, no matter how many “bots”.
This brings me to the second problem. According to this page in RealmEye, 52.86% (yes, I counted them all out with a calculator) have 24 or fewer Stars. Note that these only include “active” players that have been last seen fewer than 14 days ago. Also note that these statistics do not account for private profiles (I don’t think so). Having the server capacity be solely dependant on less than half of the RotMG community seems wrong. If we are considering the entire community (even the “inactive” players), 97.16% of the players have fewer than 25 Stars (still not counting the private profiles), which means that the server capacities are dependant on less than 3% of the community.
That is quite ridiculous if you think about it. Why?
Let’s say that there are no 25+ Star players in Server X, while there are 100 in Server Y. It doesn’t matter what your proportional amount of capacity is, because Server X would shut down. Meanwhile, Server Y has a massive capacity, meaning if someone is around the Server X area, they would be forced to play on other servers (eg. Server Y). This would cause a flood in players within Server Y, crashing all servers.
Overall, I do believe server management is necessary. However, this is definitely not the way to go.
Take all of this with a grain of salt, though, because I may have gotten the entire thing wrong.
RWT : Fixed
I’m not sure if capacity is the right word and I’m not sure if this is even possible.
It makes it harder for bots to flood servers to crash them intentinally.
You have to understand how many people created accounts and then stopped playing, or a massive amount of bots that were banned. Potentially change it to 10 stars. Alot of the “good” (aka the players who partake in halls alot) have private profiles.
Then they wouldn’t create lag?
So the proportion of under 25 star accounts is not as much.
Could you please specify the exact proportion? Given the fact that this is just extra unnecessary coding work for the developers, I’m not even sure if this is within the realm of possibility.
Or potentially decrease the amount of server data space for servers out of their timezone.
This runs into more issues.
First off, the most obvious one – a server’s data space cannot be decreased while it is being run. Another week-long game shutdown is unnecessary as well as mess up with players’ daily rewards.
Second, this may not be possible. I’m not sure what database DECA uses to run RotMG. I have heard that they use services from Amazon, which uses SQL (pretty simple to decrease in space), but I can’t be sure. I also can’t be sure if non-SQL databases can easily decrease specific servers through timezone.
Lastly, like said above, this is just extra work for the developers.
Your reasoning behind this is that if 150+ players join a single server for LH/Shatts/etc., the server lags out. But this is true for any case, even yours.
What does the server proportion do to stop this? I honestly don’t understand how this helps.
For example in fame trains/large ass halls people die due simply to a massive amount of lag/black tiling. Like the parasite event with 150+ in it and like 20 8/8 deaths. Potentially deca could slightly decrease the amount of server capacity after a server restart at like 1 am local time in the region and have it as kind of a QRF (basically to be used on crowded servers) and then when it becomes like 8-9 am transfer it back.
So your idea is to decrease the server capacity of other servers…? We’re getting somewhere, but the capacity of other servers almost never interfere with your gameplay, as they are hosted through different network routes. (I think, at least. I’m not really the best with this stuff.)
I assume they dedicate a certain amount of data/actual servers to each server, so they could divert server capacity or whatever to boost servers that are crowded from servers that are outside of their timezone.
You realise it really isn’t due to lag, but trolls kicking barrels into top left
It’s a mix of both. When you have 150-200 people in a single dungeon while another 200 players are on the server it causes lag and litterly crashes the servers since like 1/3 to 1/2 of people on are on a single server running a already lag filled dungeon. Nests are even worse.