On 9/28/07, Vincent Archer <archer@frmug.org> wrote:
>
> Fundamentally, each monster in game is a monster. Only orange.
Rather an important point. "This monster is just like the last one,
only it loads mesh 1467 in your client!" ;)
> You achieve variety by having the monster:
>
> - using different attacks (he's a melee, he's a caster)
> - being vulnerable to different attacks (he's got high AC, he's
> vulnerable to blunt weapons)
> - having different stats (he's got lots of HP and low atk, he's got
> little hp but hits like a truck)
These three situations are normally what trigger the perception of a
"different monster". In D&D, for example, you have goblins (melee),
goblin shamans (caster), goblin chiefs (different stats), and nilbogs
(vulnerable to different attacks). The biggest problem you face in the
modern MMO world is that no matter what you do, your world is going to
be compared to WoW and your system is going to be compared to D&D and
there's not a damn thing you can do to stop it.
What creates the problem - as opposed to simply a state of affairs -
is when you are doing something radically different from the gold
standard. If you build yet another D&D clone system in yet another WoW
clone world, you may be adjudicated "not as good as" the standard, but
there's still a comfort level because everything is familiar. When you
introduce something extremely different, as you're doing here, it has
to look different and feel different in as many ways as possible so
people know it's supposed to be different - and even then, you need to
introduce CLEAR SUPERIORITY in several patently obvious respects.
The system as presented so far is fundamentally an XP system with a
few added hoops through which players are required to jump, and
they're not jumping through them for personal benefit... they're
jumping though them for the benefit of others. Altruism is not, as I
understand it, a large motivator for the average MMO player. Hell, I
was treated like some sort of leper when I hung around the new player
entry areas and passed out gold on Realms of Despair (I had far more
than I could spend, and never forgot how #$@!ing hard it was to buy
decent equipment on day one in the game). The concept was just so
foreign to the player base, they simply couldn't fit their heads
around it... and if anything, it's gotten worse.
My aim here is not to make you abandon your plan so much as to
encourage some kind of modification that introduces a clear
superiority. I have no clue what that would be, but I have an idea
where you may be heading with it...
> - being in different contexts (he's alone in a camp, he's patrolling)
> - being in different tactical setups (he's alone, he's in a group, he
> has a healer backup)
This would be great!
Unfortunately, I understand this type of context sensitivity is also
rather intractable for the average game engine. I'm not sure how you'd
do it. I'm not sure how you keep the contract with the player.
For example, here's goblin shaman and there's goblin chief. If I
attack either outright, they will both be alerted; goblin chief will
lay about himself with a greatsword, and goblin shaman will diligently
heal goblin chief throughout the fight while pelting me with fireballs
when goblin chief is relatively undamaged.
What if I attack this problem laterally? If I drop a large rock on
both of them, killing them before they're aware of me, is that the
same as a frontal attack? What if I attack the shaman with a ferocious
blow that kills him outright - do I get credit for the chief's
supporting role, even though they couldn't attack me at the same time?
If I kill the shaman, then while fighting the chief ANOTHER shaman
wanders in and starts doing what he does, do I get proper credit?
It just seems like a very, very hard problem. It seems much easier and
much more likely to succeed if you do XP, because it's pretty hard to
screw up XP. "You kill the orange monster and get 4 XP." Meanwhile,
players are raising hell at your system because the shaman walked in
and healed the nearly-dead chief - forcing the fight to take twice as
long - and then they didn't get the "chief with shaman support"
achievement because of some strange artifact in the context engine.
Easier doesn't always equate to better, of course. If you really think
you can do it right, I'd rather see you try and fail than take the
easy route - everyone can learn from a solid postmortem of a failure,
but nobody learns anything from the success of the Same Old Thing(TM).
And besides, what if you succeed?
> The attendant problem being, of course, that the players are pretty good
> at figuring the type of mob that gives the highest xp/minute ratio, and
> end up killing that single mob over and over. That was the EQ model for
> years, and probably the single most boring thing...
This is where I see your task structure really breaking down. The
players WILL figure out the "best" way to get from point A to point B
for any values of A and B, and that method will propagate rapidly,
such that over time you see a half-bell curve that starts at 0 people
using the best way and then ramps up to asymptotically approach 100%
of players doing it. At that point, the vast majority of your content
is wasted because virtually nobody uses it.
> You're mixing quests and achievements. A designer takes an hour to build
> a quest because he has various tagging to do, find a quest giver, write
> dialogs, and so on.
I fail to see how it's any easier to say "you get an achievement for
crossing this line". You may "only" need to draw a line and attach a
tag specifying its level, but determining where that line goes and
what level gets attached to it is going to be somewhat more involved.
I think it's really the level that creates the issue. If the level
weren't involved, I can see some clear advantages to this approach,
but it's the whole idea of "this mob is level 1 and this mob is level
2" so if you kill the level 2 mob first you can never get any credit
for ever killing the level 1 mob. Essentially, no matter how many
level 1 tasks you create, nobody will ever get credit for more than
one of them.
So I think we need to ask that question. Why does there need to be a
level on the task? How does the system fail - if at all - when you
toss that idea out the window?
> On the other hand, achievements are not completed in 1mn either.
That's true, but PLAYER time is something you need to CONSUME, while
DESIGNER time is something you need to CONSERVE. It's okay if a player
has to spend five hours finishing your quest, because that's five
hours of the player having fun, and all your thousands of players will
each get five hours out of it. That's value. But if your designer
spends fifty hours on one achievement, that's several thousand dollars
you just spent on one achievement (assuming you have GOOD designers,
who are EXPENSIVE), and that really adds up fast.
> You do need to get in a dungeon where you will have
> 4 mobs, of sufficient level, at the same time.
If players ever NEED to do this one thing and it's the only possible
thing you can do, you have failed. The entire attraction of games like
this is choice. You can be whatever you want. You can do whatever you
want. When you start attaching conditions to things, people get
annoyed, and when you make something mandatory they will resent you
for it.
> As most MMO have problems providing varied
> content, the cop-out of having repeatable content is fairly common. Once
> you start expecting your players to have to repeat content, then making
> sure it is rewarding to repeat content is important.
What you seem to be missing is that players LIKE to repeat content.
They don't like to repeat it forever, but if I kill something and it
makes a weird squeakly noise that I think is funny, I want to go kill
another one. If that doesn't move me along the advancement path,
people like me - who enjoy experimenting with one thing for an
extended period of time - are going to end up stuck far behind their
companions who constantly seek new things to do and try to pass
through them as quickly as possible.
> You can reward repeating content. In fact, since a game built around the
> achievement model would probably face the same "player needs to find
> something to occupy 1000 hours of game play" problem, you have to find
> rewards.
I think it may be most productive to use this as a parallel system.
Your fundamental advancement doesn't have to be on the task path;
tasks could be a separate path. It would be interesting to see how the
paths could be divided using something like that, so the player is
encouraged to pursue both at once.
> if you're interested in character capacity, then the xp-derived level
> is not a particularly good indicator.
In my experience, NOTHING is a particularly good indicator of this,
and I strongly suspect any effort to automate such an indicator is
doomed to a dismal failure. The only way I've ever been able to
determine whether player X is capable of taking on mob Y is by playing
with player X in situations that demonstrate his capability. Even the
opinions of other players (reputation systems) tend to be utterly
worthless.