Welcome to Raph Koster's personal website: MMOs, gaming, writing, art, music, books.
Welcome to Raph Koster's personal website: MMOs, gaming, writing, art, music, books.

The whole Web
Raph's Website


Essays
These are full-blown essays, papers, and articles.

Presentations
Slideshows and presentation materials from conferences.

Interviews and Panels
Reprints of non-game-specific interviews, and transcripts of panels and roundtables.

Snippets
Excerpts from blog, newsgroup, and forum posts.

Laws
The "Laws of Online World Design" in various forms.

Timeline
A timeline of developments in online worlds.

A Theory of Fun for Game Design
My book on why games matter and what fun is.

Insubstantial Pageants
A book I started and never finished outlining the basics of online world design.

Links
Links to resources on online world design.



This is quite an old series of posts, from 1999 on Usenet. Players were complaining about how tough the mobs were in EQ (which had just launched at the time), and at some point the thread got cross-posted to both the UO and EQ newsgroups, and Brad McQuaid jumped into the thread. Much to everyone's dismay, I think, we mostly agreed on things.

Database deflation

Intelligence will always outweigh the game mechanics. Give it time…

Give it time--I am telling you that in time, players will find tactics to defeat these monsters. Yes, the support staff will be right behind them scrambling to correct the method used, but nonetheless it will happen. I would be very surprised if there had not already been approximately a 5 level decrease in the capability of mobs in EQ relative to players of the same level (in fact, I don't need to be surprised, I've already heard of methods specific classes use to kill 5 levels above their capability).

My magic crystal ball says that in a year, Lady Vox will be soloed or at the least taken by a group of 3 players. ;)

This is a very common thing in muds. Let me describe a "Lady Vox" I put in a mud once: it was a vampire that resided in a closed room with a locked door. The instant you walked in, it slammed and locked the door behind you, blinded you, summoned helpers to take you out, and attacked with a full arsenal of magic spells.

When it first went in, it was utterly deadly. It took a carefully coordinated group of 10 people to kill it.

Today, it is routinely soloed. Despite its capabilities having been upgraded in the meantime, and player capabilities being downgraded to match. Players figured out tactics to deal with it, and soon were superior in organization and execution.

There are massive pitfalls [to just adding tougher mobs]. To list the two most important ones:

  1. This means (mathematically) that the game as a whole loses a degree of accessibility to newbies. Features and things to do will tend to be added at the high level in order to satisfy the high level players demanding greater challenge. It is a very common thing to see muds where 90% of the zones are intended for groups of maxxed out characters only. Newbies are reduced to one or two areas per level range, and the entire process of levelling up is seen as just "the prelude to the real game." This renders the game less friendly. UO suffers from this ("build your character to GM, then go play") and so does EQ ("the game doesn't really start until level 10"). It is a bad thing.

  2. It runs the risk of unbalancing the entire economy and creating a Monty Haul system. The addition of higher level areas implies the addition of cooler and more powerful items to serve as rewards in those areas. Over time, these items filter down into an economy that has insufficient drains. The average standard of play rises as a result, and the previously difficult areas are rendered less challenging. So, the designer can always add tougher mobs, which means even more powerful items, which...

I predict that the longer the game runs, the less time it will take for people to reach level 50 from scratch; I also predict that the longer the game runs, the less time it will take to exhaust a given expansion.

Well, the assumption there is that Brad & co don't redo Lady Vox, of course. :) I doubt that will be the case (cf his reply) so it'll be hard to judge the bet. So let me clarify the situation under which I think this would occur:

If those two things are that way, then yes, I stand by it. Of course, Brad might upgrade Lady Vox, or he might not expand the game on the high end (that's breaking from the announced pattern, but hey, he might change his mind).

If a year from now there are level 150 players with magic items 5 times more powerful than what we have now, then why would it even be surprising to you if Lady Vox is taken down by a group of three?

This isn't a "defending UO" thing--this is just what happens on muds that follow that development pattern.

Yes. By saying "common" I didn't mean to suggest it was an absolute. After all, there's always systems like EQ's where there are actual level restrictions on what you can attack!

Good tactics on the part of a player, given a system that actually permits the attack, can overcome extremely large power differences. Some of these tactics might be called "exploits," certainly, and most of them are usually due to poor tactics on the part of the victim (be it insufficient AI on a mobile or lack of smarts or expertise on the part of a human player)... even in level-based systems, it's pretty common to see this happen. Indeed, in designing level restrictions like EQ's, you actually assume it occurs and formalize the expected power differential at which it can occur (eg, a guy 10 levels lower can take out a level 20, if he's smart--that's the point at which you put your hardcoded restriction in).

Well, of course [Brad will upgrade Lady Vox before that happens]! You're a diligent maintainer. It comes back to the point about the support staff working to correct this (and "correct" is perhaps a bad word--it's natural for it to occur, so it's not like a mistake was made). The problem there is that as deflation occurs because of the database growing, you then have a larger amount of data that needs to be revised. Eventually, you stop trying to revise the entire database and leave in these older creatures because you get more bang for the buck in building new tougher ones than in upgrading the old ones that are no longer competitive.

I call this "standard of living." The actual standard of living for the playerbase as a whole has risen, and was use to be expensive is now common, what used to be rare is now frequent... it's a natural process given static, non-evolving datasets.

So in your opinion, should that Lady Vox or vampire be upgraded or not? Letting it remain in its overtaken-by-events state actually makes for interesting conversation too ("wow, I just soloed Lady Vox...! Anyone remember when she was the deadliest mob in the mud?")...

Traditionally, designers haven't been able to keep pace [updating content to fit the new capabilities of players]. :) We will, of course, have to see how it does with the new breed of paid maintainers and admins.

…[Muds] generally have a much higher (orders of magnitude) admin to player ratio than the commercial endeavors do. It'd be interesting to run the math on admin hours devoted to maintenance. I would guess that most muds are still higher than the commercial games on a per capita (hours spent per player in playerbase) basis.

Yeah, the issue is the workload (and the consistency of the work done, too, since estimating the amount of deflation is very difficult when you're talking in units of orcs and magic gloves. ;)

The hidden issue then becomes budgetary, naturally--something we generally do not bother the players with. :)

Hurm, not sure you need to go to a totally closed entropic system. There's also the "floating level" notion, which I've never seen done but is an interesting thought experiment.

In the old Encyclopedia Brown books there was a mystery involving a map on a boat that supposedly got wet when the tide came in through an open porthole. Ludicrous, of course, because as the water level rises, so does the porthole--and the map stays dry.

If your definition of "level one" or of "magic item" were not based on hard numbers but instead were the result of an equation analyzing the actual current day capacities of players--then rabbits would actually spawn stronger to match the current capacities of newbies killing them. Balance would remain exactly the same--a level 1 mob matching a level 1 player. The water level raises not only the water but also the porthole...

Never seen it done in a mud though. Standalone games often do it. Especially platform games, where the enemies will be smarter or dumber based on how well you are doing, or the game will supply extra powerups if you are doing poorly.

There are many pitfalls to [upgrading only the content at the high end, or whatever range is too easy for the players]. To cite one major one, if you do this then you have a discontinuity in your availability of mobs for a given level. A level-based game has to pay very careful attention to the quantity (and frequency of death) of mobs of given levels, and their geographical distribution.

Well, [newbies running around in banded armor] does mean that the balance of the game as a whole has shifted. You have compressed levels 1-10 into the equivalent of one level's worth of difficulty (or more accurately, levels 1-9 fell off the bottom end of the scale). This reduces the advancement ladder and therefore increases churn.

I think that ultimately, the answer, as it is for most everything involving online games, will in some manner involve:

How to do that for this particular issue is a thornier question.

I do think that over time, we will move towards less and less hand-created content in these games as the scope required for the game grows and the audience increases. Leveraging player-contributed content and algorithmic and/or "smart tools" content generation is going to be very important and companies drive towards better profit margins and larger audiences. A matter of bang for the buck: where does the hand-tuning help the most, and where can you get away with fractal map generation, or algorithmically defined creatures?

Only a few weeks after this exchange, Lady Vox was taken down by a single group of six players, and the "exploit" was promptly fixed.

Child's Play


A Theory of Fun
for Game Design

Cover of A Theory of Fun

Press

Excerpts

Buy from Amazon


After the Flood

Cover for After the Flood CD

Available on CD
$14.99


More stuff to buy

Gratuitous Penguin 2006 Wall Calendar

Gratuitous Penguin 2006 Wall Calendar
$18.99


Receive CafePress Updates!

LegendMUD

click here to visit the Legend website

"The world the way they thought it was..."