Server sized shards. So they don’t have to merge realms. That is all it is.
There was an idea floated on this forum (forget who brought it up. Maybe a blizz employee. Lol) to make groups of servers that would merge together once the pops dropped. Name them Blackrock 1, Blackrock 2, and so on. Keep names unique amongst the 5 so there would be no re-naming when merged. Pops drop merge the servers.
All blizz did was take this basic idea and mix it with server sized shards. Then called it layering. Kind of brilliant honestly. Giving it a new name made sense. Sharding has a bad name round here.
Either way. I think it get shut off when it’s been as long as it takes for pops to drop. We may see an extended phase 1. Oh gawd. I can see the time gate posts now. /shudders
Which begs the impossible question…What if the pops don’t drop? What if this really is the 2nd coming of Vanilla and it exceeds all expectations? Again.
Ok, cool. I don’t want to come across as a jerk or a troll, i just feel like they haven’t actually explained layering in a very good way and when you watch it and play it on the beta and stress test it’s ‘feels like’ sharding which makes me question why bother doing it.
That “all” is a fair bit of work still though. And clearly still more to be done.
Then (as I repeat constantly) all the same mechanisms Vanilla had, come into play. Queues, server transfers etc.
But if Blizzard is watching their metrics properly, they’ll be adding new servers by the time the first server starts showing far too many L20s getting through. Cap the old ones off and establish queues on it to encourage new players to go elsewhere.
The way it was described isn’t what we’re seeing yet, so more likely its a work in progress than they decided to tell us the wrong thing.
ITT: People unironically praising shardi— sorry I meant to say Layering system which will divide people ( simply logging to an alt places you in another layer BTW ) and ruin the economy through hopping. These are people who supposedly also want the classic experience.
And we’ve also proposed fixes to the layering selection and hopping problems, because attempting to stop sharding/layering is a pointless waste of time. Layering is happening whether we want it or not, so better to discuss the solutions for ameliorating the issues, than whining about it.
As long as the exploits get kinked out, like not being able to layer hop in combat, having a CD on layer hoping, etc… I saw Monkeynews on stream layer hopping via logging out and into alts and back into his 60 to get extra Thorium nodes.
Their servers have vastly improved over the last 15 years, now capable of holding many more thousands of players on one server compared to back then (it’s talked about in the dev interviews).
In Elwynn Forest alone you can easily have 1k players. The reason why they wanna use layering, or similarly use sharding in retail, is so that gameplay goals aka questing can be accomplished and queues are lower during that launch. Which made our modern launches so smooth as you said.
Also unlike back then, they can very quickly open up another server in case of such high demand that it would be necessary to put up more servers to handle the load.
Additionally they can get an idea of how many people they can expect by looking at the name reservations they’ll have up shortly before the release.
Tech really isn’t the issue with the launch. It’s from their PoV the gameplay inconveniences and potential server health issues if players drop off drastically.
I think layering will stop when it can stop, if it’s tuned right it should go away by default when realms stabilize close to their final populations. That said I don’t see them suddenly adding a huge queue time onto realms or delaying phase 2 if populations are unexpectedly high when phase 2 hits.
I hope like hell they add more servers during launch because their metrics showed that a server would be overpopulated, not add it when the server is over populated and they’re trying to remove layering.
If they can’t predict this, they’re going to have to do far more cleanup work at the end, almost like if they used more servers and no layering. (clean up work in reverse for dead or dying servers).
So why not put a stamp on your login- something akin to, “By virtue of your login, you belong to X layer” ?? This way hopping on any number of alts you want won’t make a damn bit of difference, as they would all be on the same layer? Is this too hard or something, technically/mechanically (I’m not computer- or server architecture-savvy)?
As it is now, it seems like logging in determines what your place is with regard to layer- almost like a “soft” queue that you don’t actually see. The first 3000 per server go to Layer 1; the next go to Layer 2; etc. This apparently becomes problematic with guild members and friends who logged in earlier/later and wound up on different layers, but want to group with you.
By virtue of invite, one of you gets shunted to the appropriate layer for co-play. This is exploitable, obviously: “I found a Rich Thorium, let me invite you so you shunt over and can mine it.” Etc. So how do you fix the problem of being shunted around by groups split up to farm nodes of resources? Someone above mentioned a time-out period that would make the nodes ungatherable (word?) for a time, akin to Honorless Target debuff on zoning.
These seem to be the biggest concerns. The logout/login thing seems pretty simple to solve, actually, so the questions become 1.) Is Blizzard implementing this apparent “simple” fix of branding you with a layer designation across your whole account so you can’t individually hop layers; and 2.) How does Blizzard plan to address F&F invite exploits for farming and other things like WPvP avoidance?
I took the time to read it, I hope they do limit it if they so intend on using it to minimize abuse / exploits. There are already large discord communities forming with hundreds of people intending on mass hopping / strip mining areas with layer hopping techniques.
In my first listening to the Dev Interviews, I thought that’s what they were doing. i.e You could group and change layers, but you would always return to the default layer on leaving, and relogging wouldn’t change layers.
That begins to fall down when you need to collapse layers though, because they need to do a lot of work to know how to merge 4 layers into 2 such that both new layers are populated appropriately.
Dynamically assigning layers on login is far simpler to manage collapsing.
Bet, but what about wanting to play with my IRL buddy of years and years and years, who gets off work later and wound up on a different layer? One of us has to shunt, yeah?