+- NEUS Projects (https://neus-projects.net/forums)
+-- Forum: Sigrogana Legend 2 (OOC) (https://neus-projects.net/forums/forumdisplay.php?fid=8)
+--- Forum: Suggestions (https://neus-projects.net/forums/forumdisplay.php?fid=7)
+--- Thread: On the topic of G6, Optimization and the Server Chugging to a Halt (/showthread.php?tid=11256)
Pages:
1
2
On the topic of G6, Optimization and the Server Chugging to a Halt - polly - 06-30-2024
Alright. This will likely be on the controversial side of things. But someone has to say it.
It is no secret that every time an event is ran, SL2 just becomes... unbearable to play for anyone not attending said event. It slows to a crawl, sometimes takes nearly a minute or so breaks between inputs all because of... what? Six encounters and a couple fancy effects?
We're in the year of 2024.
Even though we're using an outdated engine that is put together by hopes, dreams, tears and duct tape, I don't think there's any real reason why this should be the case. Yet, it happens, every single time.
And I believe there is one culprit at hand; Great Six.
It's practically a barren wasteland at this point. I don't think it's had more than single digit players in the last month. Almost every house so far has been abandoned and forgotten and the continent as a whole sees no real use nowadays. It's become little more than a resource hog that... well... results in exactly the problem I wish to discuss and fix in this thread. The server chugging over the most minor of things during events.
Now. I'll say this in advance before you get your pitchforks and torches; Great Six should not be removed. Period.
Even if it's hardly more than a legacy/nostalgic thing by this point, I think attempting to just phase out the continent would cause more problems than it fixes. I don't think I would want to see it go, among many others who had a lot of memorable and great experiences with it. It still serves as a fine enough "tutorial" for newer players, since it lets you level up and grasp the mechanics of the game a little bit easier(although this is not a point I fully agree with, many could argue in the favor of this. I thought it would be fair to mention in it's defense). Not just that, but there's several keys with both G6 and Korvara characters. Trying to phase one out, or separate them to different servers is just not a solution.
However, I do believe we should cut down on the amount of the mechanical events the server has to handle in G6. Starting with:
- Dungeons
- Global Events (Fishing Contests and Black Beasts)
But I do believe that the global events in G6 also have a heavy hand in affecting the server performance every hour shift or so. I'll start with the first change I believe would require less work to implement;
Just... outright remove fishing contests.
Nobody does these now. Nobody ever shows up, nobody ever attends it, and I'll be honest; Nobody really cares, either. I don't think this would give as significant of a change as the aforementioned cutting down on dungeon spawns, but again, we're running a game on Byond. A game engine that literally shits itself and breaks over the most minor of updates. Every tiniest bit of optimization counts. The fishing contest shop could simply be changed to a regular shop that sells it's wares for Murai. Having these hosted every couple of in game hours is just not really worth it.
And now, the heavy hitter;
Black Beast events need to go too. Well, not fully...
I do think them naturally spawning should, at least.
This is/was one of the most popular events in G6. It's one that people generally always flocked to - had dedicated 'militia' showing up to deal with them, and it's without a hint of doubt a staple in G6. Outright getting rid of it would more than likely just bring an uproar from older players, and it would make it harder to enjoy 'legacy content'. I'm sure people will eventually want to try reliving the old days and gather a couple of friends to do them with. So for this, I would suggest replacing Black Beasts' naturally spawning with it being a player-summoned event. One Black Beast raid at a time, at one specific town. It would trigger the Black Beasts like it always did - but manually in a specific spot rather than a randomly picked one every few hours. Costs some sort of resource to do so and has a cooldown to avoid people just spawning it over and over and over.
Both of these getting removed would also fix the age-old issue of RP in Korvara getting interrupted/clogged by global announcements that do little to inform people and more so irritates them. It may be minor thing, but I'm sure people would still like to see that go.
Again, it's not like people attend these. If I had to take a shot every time I saw a "non existent" defence effort announcement, I'd probably die of alcohol poisoning before the day ends.
Mines are minor enough that I don't think they need to be adjusted. They're literally just a node you walk up to to start an encounter. If we are leaning into extreme optimization, they could very easily go as well. They don't give anything you can't get anywhere else in G6, as far as I'm aware.
Now,
I'm not gonna pretend that these would be "be-all-end-all" fixes. Nobody apart from Dev has real numbers on what really strains the server, and these suggestions are little more than guess work hoping to fix the issue of the server practically begging to be killed every time an eventmin dares to spawn more than 2 encounters. I do believe that optimizing what G6 does/uses is where the key to fixing this issue lies. G6's player count has severely dwindled and I don't think the masses should suffer for the enjoyment of few. I've called G6 'legacy content', because that's what it essentially is by this point. It's seen no significant update ever since Korvara's release. It's old, deprecated and sees no real use.
I've made this suggestion thread in order to not just spark discussion regarding optimizing the server, but also to potentially give Dev the encouragement by the player base to take potentially more "drastic" measures to help lower server load(such as loss of functionality i.e. my suggestion to remove fishing contests).
If you've any suggestions regarding the matter, I think this thread would be the best place to put it.
tl;dr: server go brrrr when event. not good. g6 not used as much. g6 does lots. make g6 do less. make server not go brrr. thank.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Rendar - 06-30-2024
As a small aside. Dunno if it matters but.
Dungeons are spawned on the overworld, but their dungeon layout (See: inside the dungeon) and the mobs therein are not spawned UNTIL someone steps inside of it, and it spawns on a by-floor basis.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - polly - 06-30-2024
Yeah, it does.
Probably makes the first point less of a point, since it's as optimized as can be. And not like people even walk in to trigger the dungeons fully spawning.
Still, I do believe my point of optimizing the server lies in optimizing G6 even further.
I don't think the server wanting to die over the most minor of events being hosted is something we should just accept.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Lolzytripd - 06-30-2024
A big part of lag during events is everyone downloading resources that have been just uploaded to the server by the EM
Saving, loading, adjusting encounters between phases of the event
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Poruku - 06-30-2024
All these suggestions are working off of the idea that these g6 things do in fact make us lag. While it's true that when the hour shifts, g6 does make it take a little longer, that has nothing to do with why events cause lag
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - WaifuApple - 06-30-2024
I think ultimately we shouldn't touch things based on speculation about what does and doesn't cause lag. We could assume that G6 is dragging the game down with all these things, but we just don't know that, and changing things there on assumption alone would be a relatively odd thing to do.
I'm sure the lag issues are something Dev can look into, and for all we know it might be due to unoptimized stuff in EM code, which is why events in particular are heavy on the lag. But even that's speculation, and while I do think it should be looked into for what the problem is, I don't think we should be going about it touching stuff we don't definitively know is the problem.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Trexmaster - 06-30-2024
While we can only speculate about server performance without having our hands on the profiler Dev uses it's worth considering how to improve performance after so long of every single event notably impacting the entire server.
For contests/black beast events (as the op stated) I believe those could be changed to instead be manually started by players on G6, and have a cooldown equivalent to however long it would be normally (it's roughly 3hrs/24 in game hours iirc). That way if no one is on G6 the events will never start.
The same could be done for dungeon generation on G6. If no Sigrogana players are online it could simply skip anything pertaining to G6 dungeons if it isn't the first check post reboot (so that there isn't absolutely nothing after a reboot).
Though the biggest culprit is likely just how event tools are handled by the server. All we can really do is leave it to the EMs to provide feedback from their observations in using the tools to help optimize their performance.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Autumn - 06-30-2024
(06-30-2024, 05:51 PM)Trexmaster Wrote: While we can only speculate about server performance without having our hands on the profiler Dev uses it's worth considering how to improve performance after so long of every single event notably impacting the entire server.
For contests/black beast events (as the op stated) I believe those could be changed to instead be manually started by players on G6, and have a cooldown equivalent to however long it would be normally (it's roughly 3hrs/24 in game hours iirc). That way if no one is on G6 the events will never start.
The same could be done for dungeon generation on G6. If no Sigrogana players are online it could simply skip anything pertaining to G6 dungeons if it isn't the first check post reboot (so that there isn't absolutely nothing after a reboot).
Though the biggest culprit is likely just how event tools are handled by the server. All we can really do is leave it to the EMs to provide feedback from their observations in using the tools to help optimize their performance.
While I am unwilling to hedge my bets on anything purely speculative or anecdotal, Trex mirrors my thoughts on how optimization on that end would be best tackled. As you don't want to take away features from that part of the game either, it'd be kind of like kicking someone while they're down.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - polly - 07-01-2024
(06-30-2024, 08:55 PM)Autumn Wrote:Yeah. I do agree that maybe my suggestions may have been far too drastic and unfounded. It's one of my least proud threads so far, but I do believe it still highlights a concern with the game/server.(06-30-2024, 05:51 PM)Trexmaster Wrote: While we can only speculate about server performance without having our hands on the profiler Dev uses it's worth considering how to improve performance after so long of every single event notably impacting the entire server.
For contests/black beast events (as the op stated) I believe those could be changed to instead be manually started by players on G6, and have a cooldown equivalent to however long it would be normally (it's roughly 3hrs/24 in game hours iirc). That way if no one is on G6 the events will never start.
The same could be done for dungeon generation on G6. If no Sigrogana players are online it could simply skip anything pertaining to G6 dungeons if it isn't the first check post reboot (so that there isn't absolutely nothing after a reboot).
Though the biggest culprit is likely just how event tools are handled by the server. All we can really do is leave it to the EMs to provide feedback from their observations in using the tools to help optimize their performance.
While I am unwilling to hedge my bets on anything purely speculative or anecdotal, Trex mirrors my thoughts on how optimization on that end would be best tackled. As you don't want to take away features from that part of the game either, it'd be kind of like kicking someone while they're down.
However, I disagree with it being like kicking someone while they're down.
If people were playing G6, sure. It would actively fuck over players still trying to make it work. But... Nobody is. G6 at most caps out at like six players. Even that I feel like is a generous overestimate. And even with that, they more than likely hardly play the game for longer than an hour or two at a time, in contrast to Korvara which has a consistent player base almost every single hour of the day.
I'm not suggesting to entirely phase out G6. I'm suggesting cutting back on the resources G6 uses so the server (and thus the majority of the player base) can benefit from it. And just like how Trex put it, we could simply make the hourly events triggered by players.
Removing fishing contests may have been a bit too drastic - I don't know why I hadn't thought of making it function the same way how I've mentioned making Black Beast raids work.
RE: On the topic of G6, Optimization and the Server Chugging to a Halt - Poruku - 07-01-2024
Would honestly be better for the people that do play g6, too. Log on with some bros, do a beast raid, rp in the player owned bar, idk. Having it on command sounds healthy for the game and for g6
It would also have the huge benefit of informing people when peeps are actually running black beasts