Next I hustled my booty over to the Microsoft booth to speak with Bruce Shelley,
producer of Age of Empires 2:
Age of Kings. Bruce is an icon of the strategic gaming scene: he worked with Sid
Meier in the production of such minor titles as Railroad Tycoon and Civilization, before
starting his own company (Ensemble Studios) and developing the premiere title, Age of
Empires. Now Ensemble is working on Age of Empires II: The Age of Kings (scheduled
for 10/99 release) and I was fortunate enough to have a chance to chat with him about it.
Age of Kings is a continuation of the gameplay in Age of Empires, a
real time
strategic 4x game. While it's a clear descendant of the original title, AoK builds
significantly enough to be considered a major evolutionary step in the series. When
I asked him about this, Bruce said there are really three different areas in which Age of
Kings differentiates itself from it's predecessors AoE (and it's expansion,
RoR).
First and most obviously, the game Age of Kings takes place in a different era.
Age of Empires covered the Stone Age to the Iron Age, focusing on the development
of classical civilizations such as the Greeks or Babylon. The Rise of Rome expansion
for AoE covered, well, you can probably guess. In that sense, AoE was really the RTS
successor to Civilization. Age of Kings departs from that and moves the entire
timeline to 500-1500 AD. New races (mostly), new units, and new structures all
require a serious rethinking by any player familiar with AoE. This isn't a re-hashed
old formula - this is an entirely new strategic balance and in many senses, an entirely
new game.
Secondly, according to Bruce, they looked long and hard at the feedback from players
from AoE, and spent a major effort in addressing these concerns. The AI code was
totally rewritten from the ground up. The enemy computer players now have dynamic
strategies, responsive to what their opponents (be they human or computer) are trying to
accomplish. If the human player is developing a dominant lead in cavalry units, the
computer may examine its own economics and determine that competing with the human in that
field is pointless, and switch to the rapid production of pikemen.
It's not just the enemy AI that was rewritten - there are subtle but significant
changes to the helper AI that are a result of Ensemble's partnership with the massive
Microsoft useability labs. Bruce said they spent a lot of time watching players totally
unfamiliar with the product, who were struggling to figure out what was happening.
They then went back to the codebenches and used this knowledge to seriously reduce
the amount of micromanagement required, but - importantly - not to take any of the control
away from more advanced players. This is a huge issue which I fear will be missed by
most players - beginners will play it without realizing how very much work went into
"holding their hand without letting them know it's being held" (my quote, not
Bruce's). Expert players will employ advanced functions without even realizing that
they are bypassing supports for newbies. He listed a multitude of
management-reducing functions, some of the few I'll recount here:
Much-improved pathfinding AI - one of the early complaints about every RTS is that units
get lost when they run into trees, water, building, etc. AoE was no exception to
this as a first-generation RTS game and had its own share of issues. But AoK's
pathfinding is superlative. Selected units not only will move more intelligently,
but will actually form up into column and march to their deployment. Honestly, it
looks awesome.
Unit orders such as guard, patrol and follow - as well as actual formation movement -
will make the manipulation of large numbers of troops much easier. It's possible
that, with such commands, in AoK large-scale battles will actually resemble the historic
realities of lines and flanks, rather than degenerating into brutal mass melees.
The Town Bell - tired of your foraging villagers being helplessly slaughtered when the
enemy makes a rush on your city, simply because you couldn't click fast enough to order
them all to return? AoK implements a Town Bell, situated in the center of every
town's center. When first your watchtowers identify an enemy attack, you click on
the town bell and all your villagers run to the nearest structures and take shelter
therein from the marauding enemy. Almost equally as important: when the attackers
are driven off, another click on the Town Bell will send your villagers back
to exactly what they were doing before!
There are other gameplay changes too, such as allowing the garrisoning of structures,
the boarding of ships, and improved production queuing and waypoints all contribute to
simultaneously easing the burden while expanding the options for players.
Multiplay too was scrutinized, and Bruce highlighted a couple of
multiplay-specific
improvements. One was the addition of a different multiplay scenario called
Regicide. Rather than the simple RTS formula of "kill them all" Regicide
adds an interesting twist: the success (or failure) of a player depends on the
survivability of their single king/queen. Anything that keeps your ruler alive, or
will kill the enemy ruler(s), is an intrinsic good. It doesn't matter if you have a
city spanning half the map - if your king is killed, you lose. This should add a lot
of flavor to the online arenas. The other feature that really stood out for my was
the "highlight flare" feature for cooperating online players. For example
- in AoE if you were playing online with someone and you wanted to coordinate an attack ,
or draw their attention to something, describing it in text was practically a party game
in itself: "Send your unit to the third tree by the hill, not the brown tree, wait,
I'm not counting bushes. OK, he's . . .wait, go back." Now, when you want
to draw someone's attention to something on the map you simply scroll to the spot, and set
off the flare. Whoever you want can see the flare, making the map & chat systems
work in conjunction like a virtual whiteboard.
Last but not least, Ensemble made the graphics downright beautiful - the maps are
bigger, the buildings even more detailed and active. Bruce said one of the early
issues they confronted was the fact that, since the maps were much bigger in AoK than
AoE,
the artists went a little wild and were designing large, complex structures. As
Bruce put it, "Well, that's fine if you are an artist running your monitor at 1280 x
1024, but when you'd reduce the screen to 800x600 - as many of our players might have to -
the buildings filled the screens! So we took a week and forced everyone - the
artists included - to run it at 800x600. Then we went back and worked on the
buildings." he said with a smile. The result was, I think, a good one.
The buildings seem more appropriately scaled with the terrain and figures.
Importantly now, they will play just was well on 15" monitors at 8x6 as they
do on 24" monitor supersystems.
But lest you think that there was any sacrifice in the detail or esthetics, think
again. AoK will no doubt be heralded as one of the most beautiful RTS games to ever
hit the shelf. Units, structures, and even the terrain, have all been thoroughly
done to the minutest degree - for example, the animation of the Frank special axe unit,
the figure actually walks as if he's clearly carrying a really heavy axe to one side.
Needless to say, I've joined the slavering hordes waiting for this game to come out -
the release date can't come soon enough for me!
Bruce was also forthcoming with info about Ensemble, which now comprises
upwards of 50 employees on the East Coast, West Coast, and Texas. One thing I didn't
realize is that they now employ Sandy Petersen - yes, THAT Sandy Petersen, formerly famous
for everything in the gaming field from Runequest, to Call of Cthulhu, to id Software and
little titles like Doom, Quake, and now RoR and AoK. Sandy is one of the most
creative minds in the industry today and the fact that he's working togther with Bruce
suggests that Ensemble is definitely a company to watch.
In that vein, I asked Bruce what we could look for from Ensemble in the future.
He said that they would stick with real-time, since they (obviously, to my mind) do
that very well. But with 50 people, he felt that it was very possible for them to be
working on two projects at once and were considering their options for the spare capacity.
This "second project" he affirmed would probably NOT be real time.
With their experience, talent, and people - I can't wait to see what they have up
their collective sleeves!