Ever since my earliest days at SSI with games like Wizards Crown, I've sought to make party composition matter. With SDS I'm working to achieve that in a couple of different ways. First, the ability systems allows players to create a wide variety of characters, providing the option of creating specialists, such as front line warriors, or creating generalists who can deal with more situation. Allowing players to choose and experiment with character types, party composition and tactical strategies is a key goal. In addition, we're working on bringing tactical options to the environment, and enemy AI.

With Wizard's Crown, I was happy to make a system where putting together a shield wall with a row of long weapons behind could be an effective strategy, so party characters mattered as part of a unit, not just as a needed slot in a fixed, ideal composition. With the Gold Box games, we had a limited amount of terrain, but I was happy to see players taking advantage on what was there, anchoring lines, or positioning enemies into the perfect line for a lighting bolt bounce for double damage. Many of the strategy games I've done, such as Panzer General make terrain analysis critical.

In the video, you may have noticed the Shade Elf leap to a higher elevation level. The game is 3D, so we take height advantage into consideration. You can reach some heights by conventional movement (stairs, ramps), but others require special movements such as Mystic Leap or Teleport. Leap is an expensive ability available from a couple of Specialties, while Teleport is limited use, so each has its own flavor.

Cover matters, and we are still evaluating how best to convey valid cover to the player. Using cover on the player's part leads to a more static feel to the battle, meaning that shield or armor walls are important to shield squishier, ranged characters. Limiting the directions of approach matter, so support spells, locations of impassable and difficult terrain and aggro control come to the fore.

When an enemy chooses to stick to cover, or utilize a static defense, then character mobility and altering of terrain can be deciding factors. In the video, we see the mage turning the rock wall from cover, into a cone of damaging debris. Having the right sort of attack type to utilize this strategy turns the tables on enemies. Surefooted characters can move through difficult terrain, such as loose debris fields without penalty. So enemies counting on a flank being covered by an expanse of open, slow travel will be in for a rude surprise. Similarly, the enemy counting on a high rock to guard an approach becomes unhappy when the leapers occupy it as a sniper's nest.

So destructive terrain can have multiple uses. I covered an example of removing cover, and using destruction to induce damage, but some terrain can become difficult terrain, limiting movement in the region. I'm working with the design team on options where removing a piece of terrain can have wider effects; doing some initial tests on triggering floods. The goal is to make the terrain as dynamic and tactically interesting as practical. Sometimes its as useful as an extra party member, at other times, it multiplies the effectiveness of the enemy.

Working on multiple AI approaches to fighting the party. From the simplest, path to nearest and hit it, to more static overlapping defense, or favoring more diffuse enveloping techniques. Like the player, I want the computer to switch things up and ideally recognize valid strategies against specific party composition.

In the broader context of the adventuring zones, I want party order and relative position to matter. Having a picket well out front is an approach with advantages and disadvantages. That character lacks support, but if she is both perceptive and sneaky, she can allow the party to evaluate enemies before the enemies realize. Cautious parties may be able to navigate around, while aggressive players can position themselves for optimal use of a surprise round. Pickets can also be caught out, without cover, and go down before others can react.

Other aspects of party order can involve whether to position a line of heavies in front of lighter characters, or keep one to the back of the party. How far to string the party out is another consideration. In short, I want tactical considerations to suffuse as much of the system as we can.
​