Month: May 2015

“What Server Are You On?” Is A Question That Needs To Go Away

As my friends have gradually learned that I play FFXIV, the ones that also play have wanted to play with me. And that’s great! The whole point of a MMORPG as a genre is that you can play with loads of other people. Then we try to do it, and the same conversation plays out almost every time:

Them: “What server are you on?”

Me: “Cactaur.”

Them: “Oh. I’m on <someServer>.”

Me: “Drat. Well, maybe we’ll meet up some day.”

Now, I have a great Free Company and a lot of people I know on Cactaur, so I’m not unhappy with where I set up. But in this day and age, there is no reason for this conversation to have to exist anymore.

Individual Servers Are Outdated

Fundamentally, the problem is with the infrastructure. In the old days, games tried to have big worlds and few (or no) instances. In order to scale that up, you spun up another copy of the world and people could go there instead. That wasn’t a great model 20 years ago, but it was practical given the technology at the time. But it has numerous problems, the primary one being that you split your player base up on all those servers and they can’t really interact with each other.

It also scales badly. Consider a common MMO launch, where you have tons of players trying the game out and need lots of servers. A few months later, many of those players are gone, and you need fewer servers. You can’t just shut down the ones you don’t need anymore, as players are on them, even though some of them may not have enough population to make the game actually play well when it comes to group content. As a result, you now have cumbersome and difficult server merges to do.

If the population grows, servers get overloaded. You can spin up new ones, but people will want to play on the busy ones, so to stop that overload you have to close new character creation there. This happens in FFXIV a fair bit, and it can stop someone from recruiting a new person to the game. Do you want to join a game to play with me if you can’t get onto my server?

This was the best we could do 20 years ago.

Doing It Better

Today? There’s no particular reason for those limitations. Games are split up into smaller chunks, each of which can run on a piece of hardware in a data center. Instancing is plentiful. Some games have the ability to spin up multiple instances of a given area even on the same server, to control population. SWTOR did that last one – if Coruscant got too busy, Coruscant 2 would come into existence automatically to keep things flowing. Part of the reason that’s necessary is that the cost of communicating what players are doing to other nearby players increases exponentially as you add more players.

Consider:

  • If I move and there’s one other player around, the game has to tell one player. If the other player moves, it still has to tell one player.
  • If there’s five players and all five move, the game has to notify 20 people (four players per move, five moves).
  • If there’s 10 players and they all move, now it’s up to 90.
  • If there’s 50 players and they all move, now it’s up to 2450.
  • If there’s 100 players and they all move, now it’s up to 9900. That’s 10x the players and 1000x the work.

This gets unsustainable at very high numbers of people in a single area, but when players are spread out you can handle a lot of them. Thus if they get too concentrated in one area, adding another virtual area and splitting the players between the two will massively reduce load AND allow that load to be spread across two physical resources.

The net result is that if a game is designed with it in mind from the outset, a lot of scaling can be done by spawning copies of instanced areas on a server, allowing each “server” (which is actually a collection of physical and virtual servers) to handle far more players. If the population goes up, more hardware can be added to spin up new instances relatively easily. If the population goes down, resources can be taken away without having to shut the server down and merge it away.

Yes, I am aware that everything I’m saying is a simplification on the actual infrastructure.  🙂

End The Frustration

Games that weren’t designed this way because it wasn’t practical back when they were made are in a tougher spot. Blizzard has tried to get around it in WoW with things like cross server dungeons and chat. But for new games, they shouldn’t be using the same old tired model given all the problems that come with it.

I don’t know if Square is ever planning on doing anything about it in FFXIV given the cost and effort required, but it’s a pretty huge source of frustration when I have to choose which friends I want to be allowed to play with when choosing a server. It’s a serious weakness in an otherwise great game.

“Should Modders Get Paid” Is the Wrong Question

It’s New Blogger Initiative month, where people get together and encourage new people to start blogging. If you’re reading this and wanted to blog, go for it! Lots of people are out there to lend a hand right now. Belghast made a blog about it, and even mentioned “wayward bloggers”, or those of us who lost our way. I can probably fit into that group because I’m so erratic in when I feel like posting something.

Paid Mods Create A Shitstorm As If They Didn’t Already Exist

Well, everyone knows about this, right? Steam introduces paid mods, then pulls it only a few days later in the face of the Internet exploding as only it can. In some ways, I just feel like calling the whole thing #FirstWorldProblems and moving on, but it got me thinking.

For one thing, paid mods aren’t new. Anyone remember Zygor Guides? That was a paid mod for World of Warcraft that came out eons ago. It wasn’t particularly controversial, beyond people finding it goofy that anyone would pay for that. People did, though.

How about Cities: Skylines? Did you know that has paid mods? It does, in fact they cost the community $817 per building, and are made by a former Maxis employee who did art for SimCity. People don’t find that offensive either, in fact it was hailed as an inventive way to support modders. The key difference I suppose here is that because it’s a Paetron, the mod is actually free to individuals. A bunch of people banded together as a community and that is where it’s coming from, but it’s not actually “free”. Someone is paying for it.

How about Counterstrike? Team Fortress? Every MOBA that came after Defense of the Ancients? Those all started as mods. One of them created an entire genre, and that genre is worth huge money now. Mod makers that manage to make money aren’t new.

Then Valve came in, and things went pear shaped real fast.

“Should Modders Get Paid?” Is The Wrong Question

The whole problem with this argument is that people are asking the wrong questions, and talking about the wrong things. “Should modders get paid?” is a stupid question. For this recent example, we’re talking about an American based service in Steam, and American based game studio in Bethesda, and although the game has a worldwide audience, a lot of it is in America (so I’ll focus there).

America is a free capitalist economy (mostly). That means you have the right to go create something and try to sell it. Nobody is under any obligation to buy it. You have no right to get paid to make mods. Assuming you’re not breaking any laws or the game license agreements, you do have the right to try and charge for it, if you want. That’s the reality of the situation. Asking if someone should get paid is like asking if a musician should get paid. The answer is simply “if they can convince someone their work is worth paying for.”

Defense of the Ancients. Not worth paying for, according to lots of people, because it’s a mod.

I mean, if you’re saying “no, modders shouldn’t get paid”, think about that. You’re saying that Defense of the Ancients is something that nobody should have made money from, but DotA 2 is. Why? What’s the difference? DotA created the genre, and DotA 2 couldn’t exist without it. Tons of time, effort, skill, and love went into DotA, more so than a lot of other games that people consider it okay to charge for. Yet it’s not allowed to be monetized, while its sequel is? How does that make any sense at all?

That’s the problem with this whole question. It seems to assume that we’re talking about overly simplistic mods, rather than the high end mods that are essentially game upgrades or full games in themselves.

How Should We Pay Them?

The real question here is how to do this in a way that works for everyone, and this is where Valve failed in horrific fashion. In particular, Valve gave themselves a 30% cut, Bethesda a 45% cut, and the modder a 25% cut. If you pay $4 for a mod, the modder gets $1 and Bethesda (who did nothing), get almost double that.

That’s absurd. I know, people say that Bethesda created the game. So what? They were already paid for the game when you bought it to use the mod. They use all the mods as a marketing point to sell more copies of the game (as does Paradox with Cities: Skylines, and Blizzard with Warcraft 3 when stuff like DotA was massively prolonging the life of that game). They’re already making money because of mods, and there is no particular reason why they deserve a cut of someone elses work.

“But the modders are using their game!”

So what? If you create and show a Powerpoint presentation as a consultant, do you give Microsoft 45% of the consulting fee? No, of course not. That’d be stupid. Microsoft was already paid for the tool. They don’t get paid when you use Notepad, and they don’t get paid when Bethesda used Microsoft’s compiler (or whatever compiler they used) to build Skyrim. Black & Decker doesn’t get a cut of every job you do with a hammer, either. Those are tools. You pay for the tool, then you use it.

In the case of modding, Skyrim itself is a tool. It should be treated accordingly.

Anyway… Valve’s system was bad because modders got so little of the money (among many other reasons). The Paetron model used by some others is better in that regard, as the modder gets 95% of the amount you pay. The problem with that model is that it’s basically on donations, and lots of people get to use the mods without paying anything. That itself isn’t a perfect model, but it’s vastly superior in terms of dollar efficiency and has been used successfully.

There are ways to go about this without totally botching it up like Valve did, and I expect it’ll get figured out one day. People who create entirely new games as mods, or do other massive enhancements should have the ability to make some money by doing that, if people are willing to pay for it. Anything else is an entirely artificial divide where games made in some tools are more worthy of money than games made in other tools, because reasons… and that is entirely ridiculous.