[Poll Inside!] It Takes a Village - A Discussion on Node sizes and Server Population.

Please take this poll on whether or not you'd prefer to have your character reside in a Freehold or an Apartment:

http://poal.me/umjfss

Thank you!

Now, onto the post.

Player Count.

A very trusted content creator told me some time ago that sustaining even something as small as a Village (a Zone 3 node) would take hundreds of active players contributing towards this endeavor. I did some math and even if we were to assume that all 103 (there is really 118) nodes were Zone 3 (this isn't possible) and each required 300 players to sustain. This is an active server population of roughly 31,000 players.

Let's dig into this a bit more. If we assume that each type of Zone is an order of magnitude larger than the previous one.

Zone 1 - 3
Zone 2 - 30
Zone 3 - 300
Zone 4 - 3,000
Zone 5 - 30,000
Zone 6 - 300,000.

This is obviously incorrect. It is far too many for any MMO, let alone any game, and so it's not realistic and probably not what Intrepid has in mind. Now let's aim for something more conservative. Again, using that Zone 3 value as a baseline.

Zone 1 - 3
Zone 2 - 30
Zone 3 - 300
Zone 4 - 1,200
Zone 5 - 2,400
Zone 6 - 4,800

This seems a bit more realistic, but is it really? No, and I'll explain and show why. According to Intrepid, each Zone 6 (Metropolis) is meant to have a Zone of Influence that comes in at around a massive 20% of the map. This means that each Metropolis holds dominion over roughly 20 nodes. And this is of course setting aside castles. Assume that the values I have here are correct. I know for certain that Zone 3 (Village) one is the approximate number given by Intrepid per the content creator. See below for what I mean.

This is a very rough mockup, but it shouldn't be altogether too far off if the server is extremely healthy. I need to state this again: if the server is extremely healthy this is what we can probably expect to see in each region. This comes out to a total of just under 18,000 so let's just round it up to 18,000. Now we know that there are five of just such regions in this example, bringing the total up to 90,000 active players. That's a lot of players. Especially if you consider that every node's zone level is crucial. Intrepid has said they all provide different things in the chain up and down, so even a node that is ranked 1 is important.

However, if we cast aside the idea that a Zone 3 (Village) will need several hundred active players to sustain itself then we can make the numbers anything we want. A particular set that maybe even comes close to matching what Intrepid's own supposed aspirations are for server sizes; which I'll get into below. And it's this (as brought up in this thread by Azherae):

Zone 1 - 0
Zone 2 - 4
Zone 3 - 16
Zone 4 - 64
Zone 5 - 256
Zone 6 - 1024

This brings the necessary threshold for sustainability all the way down to roughly 11,000 active players across the entire server working on nodes. That's pretty decent, but not so excessive that it makes you question the feasibility of the feature. Still, I haven't exactly explained why this isn't realistic. Even if it's as low as 11,000 or as high as 90,000 active players (not concurrent) on a server. These numbers are big (gargantuan even), but not out of the realm of possibility. Especially if Intrepid utilizes a very small number of servers. Small being half a dozen to a dozen. Intrepid themselves have stated that they plan on having servers start at around a registered player cap of 15,000 each which is then gradually increased to 50,000 over time. This isn't infeasible. No, what's infeasible is the size of the map.

Map Size.

As we know the map for Ashes of Creation (not including the Underground) is roughly 500km by 500km. This may seem large, but veteran MMO players will know this isn't a groundbreaking map size. To put this into context: 500km by 500km is about the size of Phoenix, Arizona. Once more, in theory, this should work but Phoenix is a major metropolitan area and much of it is occupied in a manner befitting of one, rather than an expansive setting of various relatively undisturbed biomes and large sections dedicated to PvE. Consequently, what this means is that the towns, villages, cities, etc. won't be to scale. The vast majority of players will have to subsist within apartments so as to not overtake their land allowance. This, however,
is absolutely not going to be satisfactory for most players when given the choice between an apartment and a home. In fact, I am going to place a poll to try and gather supporting data to show this.

What this all means is that the living areas within the game won't be nearly so large as to support the number of people required of them and that the actual size of servers will need to be much lower in order to allow for the game to operate as intended (large areas undisturbed, PvE zones, metropolises that can't possibly be to scale with real-world cities, etc.). What I suspect is that Intrepid is going to have to rework these values massively when Alpha 2 occurs and they get some genuine data that supports this situation. I also believe that servers, as a consequence, will have fairly low player caps when compared to the potential 90,000 that we might otherwise see. Now, when I say low, I don't mean just a few thousand, but rather perhaps 5 to 15,000 with much lower active player requirements. This would solve a lot of infrastructural network issues for Intrepid and also balance the scale of the map with the number of players they might expect to see.

Find more posts tagged with

Comments

daveywavey

Do I have to pick one or the other? Can I not have both? What about Static Housing?

Azherae

I don't wanna go back to 'speculating ways that things Intrepid plans don't make sense yet', since technically, it's not feedback until they give us something working to look at, in the case of something like this, but I do believe that generally, the discussion has merit and I've thought about it a lot.

The numbers literally don't add up no matter how I crunch them and it makes me uncomfortable about the concept basis of the game.

It's not because I expect the population of each area to not work, it's moreso that I expect the distribution to not work. even with that system where a node passes on some exp to its parent node. But they have lots of time to tweak these numbers. Also, if you didn't see it on the wiki, the expectation is:

Around 8-10k concurrent users per server is projected.[2][3][4][5]
Initially there will be a limited number of registered accounts (approximately 15,000) per server to help mitigate login queues.[6]
This limit will increase over time to around 50,000 registered accounts per server.[6][7][8]

Anyways, I bring you this data to help you edit your 'premise' if you wish it, since it makes assumptions that will mostly just 'be challenged' and then 'make your data collection invalid'.

Having played Alpha-1 I can also tell you that your primary concern of 'there not being enough Freehold space for everyone' is probably correct, but only based on the things we already know. Also, cities have a number of houses and mansions internally, to the point where a village would be able to hold at least 8 people.

Your orders of magnitude are therefore off. If you're taking suggestions, try the standard double bitwise:

Stage 1 - 0 (no actual housing, just people wandering by)
Stage 2 - 4 (still no housing)
Stage 3 - 16 (This is covered by in-node housing, with space for about an equal number of freeholds)
Stage 4 - 64 (some of these people live in the villages and freeholds of those villages)
Stage 5 - 256 (see above)
Stage 6 - 1024 (never seen it, can't say it'd work, but there's meaningful extra space around Village nodes now, even before you start to take up spots that might be for Freehold placement, but this was in Alpha 1).

The number of suggested players appears to be implied by 'the fact that once a Metropolis is built, people are probably still trying to raise the surrounding towns, which in turn sustains the Metropolis.

tautau

Correct me if I am wrong, but you population estimates seem to assume that all of the nodes will rise to the highest level that the game structure allows, right?

Isn't it possible that a well populated server will establish only two or perhaps three metros? That would mean that there are areas of the map where there might be level -0- nodes (or a bunch of level 1 and 2 nodes) around the edges of the 'civilized' regions. Thus the total server population would be lower while still having a healthy game.

In fact, in order to have the maximum number of metros on a map, that would require that the metros be almost perfectly located on the map to fit them all in. That would seem to be an unusual alignment. Fast growing nodes might tend towards the center, being too crowded together for perfect placement. So a server might have three metros sort of centered in the map with level 3 and 4 nodes between them, and the edges of the map might have areas of level 0, 1 and 2 nodes, maybe a level 3 off in the corner. This would result in civilization near the center of the map and wild areas all around the edges, which could be a pretty cool setup.

The central metros could have shifting alliances, occasionally with one being overthrown and a nearby level four node replacing it as a new metro. I will even suggest that this might evolve into the most common server situation and that a server with five metros would be quite rare.

Ulfbrinter

Azherae wrote: »

Your orders of magnitude are therefore off. If you're taking suggestions, try the standard double bitwise.

Thanks for the great feedback, I actually brought the point up about server caps with the person from whom I received this information because, like you, it didn't add up. I could genuinely see a Metropolis needing 1,000 players, that's just three maxed-out guilds and some stragglers who want to live in a city. What completely threw off my expectations was being told that: "No, a village will require hundreds of players supporting it." and this knocked me off my feet. I don't think it's all doom and gloom though. They just need to tweak their expectations a bit about precisely how many people can actually be on to do the things they want and how many servers they expect to have spun up.

If you don't mind, I will edit your example into the OP because I personally feel it is far more realistic despite going well against the solid information I have from someone who genuinely seems to know every last shred of detail (even things no longer able to seen due to deleted channels and DM's) about this game.

McShave

I highly recommend going back to the September livestream and looking at the node simulation they had going. In that clip, you can see how many nodes of each level there are at a time, and how the number of nodes at each level change over time. It also gives examples of how many nodes of different levels are under a metro node of influence too.

Ulfbrinter

McShave wrote: »

I highly recommend going back to the September livestream and looking at the node simulation they had going. In that clip, you can see how many nodes of each level there are at a time, and how the number of nodes at each level change over time. It also gives examples of how many nodes of different levels are under a metro node of influence too.

I watched, the clip is also in the Wiki. I didn't see a single one get above Zone 3 before collapsing. Maybe you did, but I didn't.

Azherae

Ulfbrinter wrote: »

Azherae wrote: »

Your orders of magnitude are therefore off. If you're taking suggestions, try the standard double bitwise.

Thanks for the great feedback, I actually brought the point up about server caps with the person from whom I received this information because, like you, it didn't add up. I could genuinely see a Metropolis needing 1,000 players, that's just three maxed-out guilds and some stragglers who want to live in a city. What completely threw off my expectations was being told that: "No, a village will require hundreds of players supporting it." and this knocked me off my feet. I don't think it's all doom and gloom though. They just need to tweak their expectations a bit about precisely how many people can actually be on to do the things they want and how many servers they expect to have spun up.

If you don't mind, I will edit your example into the OP because I personally feel it is far more realistic despite going well against the solid information I have from someone who genuinely seems to know every last shred of detail (even things no longer able to seen due to deleted channels and DM's) about this game.

I don't mind, and I actually don't think that the person is 'wrong' either. I think what might have led to that data, is the fact that a parent node takes some exp from its Vassals, which means that a Village surrounded by 6 Encampments technically has 40 'residents', indirectly, not '16'. And of course this isn't counting all the travelers that pass through.

I feel like 'a Village that gets no visitors', even if 'people generally live in the outskirts' would be approximately right.

Similarly, since the Encampments aren't allowed to level up to Village while the Village isn't yet a Town, technically all those people (assuming just basic play) want this to happen. They want the exp to percolate up to the Village so that it becomes a Town, and then they want to 'race' with the other 5 to be the first to become the new Village.

So, projecting from that...

"If an Encampment only had 4 supporters, it might not level'.

"If a Village only had 40 supporters (counting those in Encampments), it might not level"

And so on.

So at that point, in order to raise a Village to Town, using this method, you probably need 10 or so players 'per Encampment' and whoever lives in and around the Village itself. A hundred players wouldn't be unreasonable at that point, to assume, but since it's 100 players over 7 nodes, the usual napkin math tells us that 1500 players would be enough to cover the whole map in this config if you spread them all out perfectly.

If Ashes changes the ways it levels Nodes, obviously everything I say here is useless.

That said, I think I can answer your question now, and hit the poll.

I can't imagine being Apartment based, I don't like cities in MMOs much, and even if they're not fully safe, I prefer not to be too close to safe areas as it's less fun for me. Given my goals, Freehold sounds much more fun, and with the system for letting friends at least use some of the facilities, it'd suit my group better.

Ulfbrinter

tautau wrote: »

The central metros could have shifting alliances, occasionally with one being overthrown and a nearby level four node replacing it as a new metro. I will even suggest that this might evolve into the most common server situation and that a server with five metros would be quite rare.

This is also something I have thought about a lot, and back in July I made an extremely thorough thread on this very topic that dives into ways small groups could actually control massive areas of the map. And that the countermeasures Intrepid have spoken of to try and dissuade J4G/Zerg groups (or just clever, coordinated players) from just swarming a map to be pretty unconvincing. I also feel that Intrepid has this idea (which probably has some value) that due to the map being large enough and fast travel being fairly limited, people will want to centralize on locations where the most people are purely out of convenience. It's not a farfetched notion. Also, the 5 metro thing comes from Steven, this is his idea of how a server should look. Whether or not it actually will, in practice, is another matter. Especially with servers that have had their population counts drop off massively but cannot be merged into a new server since that'd mess up so many things.

tautau

Keep in mind that, while a player might spend a lot of time around 'their' node, they are also spending time in nodes where quests take them, nodes where friends are, exploring and other activities. So there isn't a pure player/node connection. It may average out, though. But nodes with popular/semi-required quests will have an advantage, as we saw in Alpha 1.

I think Steven said something like 'a maximum of five metros' which isn't the same as saying that five metros is how a server should or would look.

Noaani

Just reading the OP, I have a few points to make that I think are missing, and would drastically alter the outcome.

The first is in the node graph provided. It would not be possible to have three level 5 nodes like that, as these nodes extend a ZoI (even within the level 6 nodes ZoI) that prevents other nodes getting as high as it.

As such, it is unlikely that a metropolis will have more than two level 5 nodes attached to it, and they are very unlikely to be right next door.

Second is in the number of players. Nudes under level 3 have no upkeep, no housing, and thus require no players to maintain them.

Third, that figure you were given was the number of players that are likely needed to level the node up, not the number needed for upkeep.

Just by reading your OP, I think either your content creator friend knows less than they think, or there were some lapses in communication.

Edit to add, in regards to your poll, there are three types of housing, not two, and a character can have all three types of housing if the player wishes. There is no need to decide between one or the other.

Ulfbrinter

Noaani wrote: »

Second is in the number of players. Nudes under level 3 have no upkeep, no housing, and thus require no players to maintain them.

Do you have a source for this? Because the Wiki (from what I saw) makes no distinction between zone atrophy for Zone 1's and 2.

Noaani

Ulfbrinter wrote: »

Noaani wrote: »

Second is in the number of players. Nudes under level 3 have no upkeep, no housing, and thus require no players to maintain them.

Do you have a source for this? Because the Wiki (from what I saw) makes no distinction between zone atrophy for Zone 1's and 2.

No player buildings, no amenities, no citizens, no government, nothing to maintain.

Nodes under level 3 are transient. With the exception of the first day or so of the game, people are not expected to specifically hang around.

ButtercupClover

We are also under the assumption that experience is the only thing they are talking about when talking about maintaining a node. Perhaps the hundreds of players part is in reference to the materials and gold needed to keep the buildings functioning while the XP progresses.

There's also the facts that the radius of influence expands so that more people can contribute to the XP growth of a node with more areas to farm. There's also the vassal system that would increase the number of players contributing. Players wanting a neighboring node to level will contribute to the parent node.

It doesn't seem too far out of expectations to think that it's possible. Perhaps it will be too hard at first and need tweaking but I think the player count factors in more than just the XP needed to maintain node progression.

Ulfbrinter

ButtercupClover wrote: »

There's also the facts that the radius of influence expands so that more people can contribute to the XP growth of a node with more areas to farm. There's also the vassal system that would increase the number of players contributing. Players wanting a neighboring node to level will contribute to the parent node.

As I am not going to reveal who it was that told me this, or divulge the exact messages that were exchanged you will have to trust that this conversation was about the number of people in a single node needed to support the Tier 3 village in that node. I at first believed that maybe they were referring to the total number of players between the five or six nodes surrounding that Tier 3 node, but no. I do not believe they are wrong and that this is what Intrepid intends. However, I do not believe it will work in practice. As you point out and so did another in this very thread, the likely situation is that it could be tuned for these values initially but once finding that simply do not work they'll need to be tuned downwards in a big way.

Noaani

Ulfbrinter wrote: »

I do not believe they are wrong and that this is what Intrepid intends.

See, this is the part that makes me think either your friend or yourself misunderstood.

Intrepid are not making detailed enough plans for things like this yet. At the absolute best, they would say that it is their current thinking, or maybe where they plan on beginning their tests from - but that it isn't actually an intention for the game once it goes live.

If your friend is presenting what they said as their intention for the live game, you should have words with them.

Azherae

Ulfbrinter wrote: »

ButtercupClover wrote: »

There's also the facts that the radius of influence expands so that more people can contribute to the XP growth of a node with more areas to farm. There's also the vassal system that would increase the number of players contributing. Players wanting a neighboring node to level will contribute to the parent node.

As I am not going to reveal who it was that told me this, or divulge the exact messages that were exchanged you will have to trust that this conversation was about the number of people in a single node needed to support the Tier 3 village in that node. I at first believed that maybe they were referring to the total number of players between the five or six nodes surrounding that Tier 3 node, but no. I do not believe they are wrong and that this is what Intrepid intends. However, I do not believe it will work in practice. As you point out and so did another in this very thread, the likely situation is that it could be tuned for these values initially but once finding that simply do not work they'll need to be tuned downwards in a big way.

In general, Intrepid is not necessarily in a position where they are telling us 'this is what we definitely will do'.

In software development of all kinds, you can definitely go 'here is our design spec, this is what we expect to happen and our plans', but that isn't a commitment to those things in the same way, which is something very hard for laypeople to understand sometimes, and why game developers hate to give roadmaps. It's actually not usually because they can't hit their roadmaps, it's because they aren't sure that if they hit them the experience itself will be good.

They can choose between 'implementing the thing exactly as they have planned it', with whatever comes out of that, and 'adapting when they realize new things', but then make their previous statements invalid. It's one of the few reasons I am still following this game (yes I know I backed it, not a huge bet for me or my group financially). There are things that make this 'not a game for me' or 'look like they won't work', but my experience tells me that they have a pretty good chance of changing some of them, even if it contradicts their previous statements.

EverQuest Next was cancelled, I would BET, in part because of this. They designed 'to spec' and then analyzed the experience and went 'this... isn't actually fun, we should have changed more things' and then decided to cut their losses instead of falling into a sunk cost.

I stand firmly on the 'I don't believe Intrepid is this clear on most of the things they have said in the past, yet'. I believe they are doing their best to give us the overall 'feeling' of what they said, and I hope that the way they choose to do it is the one I agree with (or at least that it is the other one that I disagree with but works really well, so that others can enjoy it).

But anything to the effect of 'Node sizes', or even 'positions of nodes on the map', or maybe even 'the precise number of active nodes', I expect can still change, and I'm sure Steven regrets ever making some of these specific statements.

But that's what happens when you show someone your schema on paper and start talking about it to 'customers' or 'clients' and then find yourself needing to change things. There will always be that person who makes a whole YouTube Video titled "Ashes of Creation: 107 Nodes instead of 108! Is Intrepid Studios compromising their Values? Did Steven Sell Out To The Prime Number Illuminati?"

daveywavey

Azherae wrote: »

Did Steven Sell Out To The Prime Number Illuminati?"

Bah, I hate those guys. They ruin everything!

Child Item