In this month's exclusive interview, the designer of the new D&D adventure Lord of the Iron Fortress discusses playtesting challenges, math for advancing PC levels, and building fortresses on other planes.

Wizards of the Coast: Lord of the Iron Fortress is quite a dangerous adventure, taking PCs to the Infernal Battlefield of Acheron. What inspired your design?

Andy Collins: I've always enjoyed the challenge of high-level D&D -- not only playing it, but designing adventures for it. I'm also a big fan of the outer planes, particularly as envisioned by the Planescape campaign setting. Wizards of the Coast hadn't really addressed planar travel in any of our adventures as yet, so I wanted to include it as a significant part of Lord of the Iron Fortress. This was reinforced by the fact that the Manual of the Planes was in development at the time, giving me lots of great ideas.

Wizards: What's the process like for playtesting an adventure?

Andy: Playtesting can be a grueling process, full of stops and starts, dead ends, and other frustrations. You're rarely testing a complete and edited document, so the DM has to interpret as best he or she can. It's not uncommon to re-run an encounter after revising it based on feedback. In a best-case scenario, you'd want to use a different group (to avoid preconceptions), but that's not always possible.

I've actually never run this adventure all the way through. I've playtested some sections of it, but we generally prefer to have someone other than the author playtest adventures, to ensure that it can be run "as written" rather than relying on the author's opinions of how he or she thinks the adventure should work.

Wizards: And how can you tell if an encounter is too powerful for players versus the playtesters simply making mistakes?

Andy: That's a place where you have to rely on the experience of the DM and the players. We try to "debrief" after playtests, to discuss why things worked or didn't work, and that sometimes results in decisions of "if we had (or hadn't) done X, we would (or wouldn't) have succeeded." Since a typical playtest group around here has something like 80+ years of total gaming experience and 25+ years of professional game design experience, that's perhaps the most potent resource one can have. When in doubt, you test it again.

Wizards: We don't want to reveal too much to players who might come up against the adventure's villain, but Imperagon is quite the nasty. How do you go about creating a powerful adversary without creating a stereotype or a knock-off of some other well-known villain?

I look to the rules for inspiration. Rather than saying "I need an evil spellcaster," I say "What would happen if I took a rakshasa and gave him additional sorcerer levels?" I look for good combinations of monsters and classes (including prestige classes), and try to make sure that villains aren't extraordinarily vulnerable to any common tactic.

Also, I think it's important to give your villain high-quality hench-beings. That lets the villain shore up his or her weaknesses, while simultaneously giving the PCs more to worry about, whether in a climactic battle or simply along the way.

Wizards: Have most of your playtest groups succeeded in defeating Imperagon and halting his master plan for domination?

Andy: Since most of our playtests focus on individual encounters (or groups of encounters) rather than the entire adventure, I can't really say. My guess is that this adventure will prove . . . challenging . . . to most characters.

Wizards: An understatement, no doubt. Do you find, after a few times DMing an adventure that you've created, that certain sections of the overall tale have consistently greater excitement or depth than others?

Andy: I think that's definitely true. There are a few adventures that I've run multiple times over the years (whether self-created or otherwise), and whether it's due to the designer's skill, DM's attention or interest, the players' involvement, or some combination, there are always some scenes that work better than others.

What's even more intriguing to me is how different groups can take the same scene and make it memorable in entirely different ways. I've run two groups through Bruce Cordell's TheSunless Citadel, and each group handled the climactic encounter in completely different (but equally exciting and memorable) ways. Go figure.

Wizards: What are the challenges in developing an adventure that takes place in another plane? Does it give you as a designer more freedom or more restrictions?

Andy: From a rules-mechanics standpoint, designing for the planes has never been easier than with 3rd edition D&D. The Manual of the Planes provides simple, basic mechanics for adventuring on the planes, and it's easy for a designer to highlight those.

I think one of the keys for good planar adventure design is to focus on what makes the plane (or planes) interesting and unique. Don't just put your dungeon on another plane without considering how that makes your dungeon different. That might be unique monsters, strange planar qualities, or whatever else comes from locating an adventure on the planes. I find that to be a great stimulus to my creativity.

Wizards: As a designer, do you ever find yourself favoring a minor secondary character, like the hunter Mundroot, or wanting to develop subplots around seemingly random encounters?

Andy: I try to keep my actual plot-writing to a minimum in an adventure. I prefer to see myself as a "frame-builder" rather than a storyteller -- the DM and players are going to create the story; I'm just setting the scene.

That said, I like to think that even secondary characters or random encounters can provide a springboard for further adventures, so I feel obligated to include at least a little bit of information that can help the DM in that direction.

Wizards: Well, then, of those secondary characters and random monsters -- the new monsters, spells, and magic items that you created for this adventure -- do you have a favorite?

Andy: Before I even started writing, Todd Lockwood sketched a beautiful metallic catlike creature that he thought might work well in the adventure. That creature became the "steel predator," a nasty beast that will harry the PCs throughout the adventure. You can see it in the lower left corner of the cover illustration.

Wizards: While designing, do you keep open the possibility for a sequel to the adventure?

Andy: Without using the "s-word," I'll admit that I think any adventure that doesn't leave a couple doors open for exploration by the PCs is really dropping the ball. These could be as simple as an NPC who's likely to escape or as significant as major events put in motion by the PCs' actions (or inaction).

Wizards: So, how has the new edition of D&D changed the rules for designing?

Andy: The new rules have a robustness that's never before existed in D&D. Classes, skills, feats, spells, magic items, and monsters interact in clearly defined (and thus easy to exploit, for a designer) ways.

I love the fact that I can make a salamander a half-fiend 4th-level fighter/3rd-level rogue wielding a spiked chain to inflict nasty sneak attacks on unwary PCs. I love that I can include challenges that focus on the PCs' skills (such as climbing, bluffing, jumping, swimming, and so on) and have a clear guide as to how to handle them.

I tend to design adventures -- particularly those that will see print -- in a very modular fashion. I first figure out about how long I want the adventure to be, whether in number of encounters or total amount of experience points I want to award. I then break it down into a list of target encounter levels, and try to fill those from a list of monsters than I've already picked out. The system of XP, Challenge Ratings, and Encounter Levels in the Dungeon Master's Guide makes that an incredibly simple task.

For instance, in Lord of the Iron Fortress, I knew that I wanted to advance 15th-level PCs about two levels. That meant that I needed to include about thirteen EL 15 encounters and thirteen EL 16 encounters. Since having encounters of all the same Encounter Level is pretty boring, I started by tweaking the numbers a bit (but keeping the total expected XP rewards the same). This resulted in a "menu" something like this: four EL 13 encounters, four EL 14 encounters, seven EL 15 encounters, ten EL 16 encounters, two EL 17 encounters, and one EL 18 encounter. Then I started filling in monsters that I knew I wanted to include, which led to further tweaking as necessary. An Excel spreadsheet helped me keep track of expected XP rewards, to let me know that I hadn't strayed too far from my target numbers.

Wizards: Wow -- who'd have thought it was so complex? Now that all that work is behind you, what plans do you have for future designs?

Andy: Most of my adventure design energy these days is focused on my personal campaign, titled "Bloodlines." Among recent adventures, the PCs have thwarted a slave ring, taken on a dragon cult that had been infiltrated by yuan-ti, ventured to the Plane of Shadow, and stopped a barghest warlord from overthrowing the capital city of the kingdom.

Professionally, I'm something of a switch hitter, going back and forth between designing and editing. After Lord of the Iron Fortress, the next example of my design comes with the Epic-Level Handbook (coauthored with Bruce Cordell, Thomas Reid, James Wyatt, and John Rateliff). I also recently finished a big article for Polyhedron Magazine, featuring the "return" of a popular D&D setting from years past. And I'm a frequent contributor to Dragon Magazine, including a bimonthly column due to debut early in 2002.

As for my next design project for Wizards R&D . . . it's scheduled for release so far into the future that I can't really discuss it. But it's big, both in scale and subject.

©1995-2008 Wizards of the Coast, Inc., a subsidiary of Hasbro, Inc. All Rights Reserved.