The Ultimate Torment Interview Of Ultimate Torment, Pt 2

I very much want Torment: Tides of Numenera to be excellent, because the world needs more Torment. Not in the literal sense, of course; the world is a miserable place. But Planescape Torment was a wonderfully different sort of RPG set in a wonderfully different sort of world, and another descent into the gnarliest bowels of fuckweird would be quite grand. Numenera's still a ways off at this point, but inXile seems to be on the right track. Yesterday we talked combat and why quality is more important than size, and today we continue on by chatting about why Planescape Torment *wasn't* perfect, what that means for Numenera, the recent delay, and why we won't just be able to attack any old random NPC. All that and more below. 

RPS: Have you come up against any design dilemmas where you decided to do something that - from the outside looking in - would appear to be in direct opposition to Planescape's design? What other factors do you take into account in these situations? Beyond the obvious surface-level, "Well, PST did it this way or that way" stuff?

Heine: In most cases, we're just taking what PST did and pushing it further. The Tides are a prime example of that. PST pretty much broke the rules of D&D's alignment system to achieve its goals. Now that we're not tied to any system, we thought, "What were those goals, and what's an alignment system that could accomplish them even better?" The result is a more organic and nuanced system than what PST had.

When we do break from what PST did, we have to ask ourselves why PST did it that way, what they were trying to accomplish, and is our proposed solution better than that? For example, early on I had assumed that combat would be allowed anywhere - because that's how PST did it, and because I, being a relatively old-school gamer, had never played a game where you couldn't do that. Others had assumed the opposite.

The ensuing discussion forced us to ask important questions. Did PST allow combat anywhere because it was the right thing to do or because that's how the Infinity Engine worked by default? Was it a critical part of PST? This is a tricky question, because for any given aspect, there will always be some people who believe that it was. Did it work and was it a good decision for PST?

We determined somewhat reluctantly that it wasn't critical to PST and that it added a lot of work for the designers and scripters at the time. Then we had to decide: is our proposed solution better? What does it gain us? What do we lose by it? In removing the possibility of combat anywhere. We lost some perceived freedom, but we gained more focus on our core vision - no trash mobs, quality, handcrafted encounters that support the narrative, etc, and a heck of a lot of time that would otherwise be spent designing, implementing, and debugging reactivity to handle the case where any combination of NPCs might have died. Because that time would be spent improving quality and reactivity elsewhere in the game, where it would be more likely to be seen by more players, we decided to drop the "kill anyone" approach.

But then as dialogues and designs started coming in, we realized it was almost too restricting. I mean, sure, we don't have to cater to the player who just wants to slaughter everyone to see what happens, but if an NPC is in your way and really pissing you off, shouldn't you have the option to smack them down? The problem with this is that our Crisis concept demands a limited number of handcrafted situations, but we couldn't go through the whole game handcrafting every possible scenario where the player might want to get into a brawl.

The solution was what we called mini-Crises, or Tussles. They're basically shorter, non-handcrafted combats that are always entered into by player choice or occasionally by player failure, but usually the player will be aware that he's trying something that could start a fight. In this way, we can give the player freedom to attack people that are reasonable to attack, while still maintaining control over which NPCs can die and when. It also gives players who want to focus on combat more opportunities to do what they're good at.

Though, as with everything, we still need to prove out how well these will work, or how much extra effort will be necessary for them to work well, before we can commit to it. If Tussles as we currently imagine them prove too ambitious, we have some fallback ideas that would allow for this type of freedom in other, simpler ways.

RPS: PST wasn't a perfect game, despite how revered it is. What are you hoping to improve, fix, or redesign from the ground up?

Heine: Combat is the obvious one. Many felt that PST's combat was tedious and uninteresting, and we have set out to improve that from the start. PST also many times had to wrestle with or completely break its own RPG system. While we didn't choose Numenera because of its flexibility, it has turned out to be a huge boon on that front, giving us the freedom to do many things we could not have done otherwise. The only pushback we've really received from Monte Cook Games has been in terms of how we portray the Ninth World, and because we are eager to present the Ninth World correctly, we have been extremely happy with this feedback.

One thing we keep pushing on in TTON is that there should be no best solution, no best ending, and no preferable playthrough. PST did okay in this, but if you went in with a high Int/Wis/Cha character, you would experience a deeper and richer story than other character builds. And although PST's endings were pretty nuanced, there was one that is generally considered "best". In TTON, we hope that each player will feel that they got the best playthrough, because it was the playthrough that was best for them. This is a high bar, and I don't know if we can reach it, but we're sure as heck going to try.

RPS: On that front, you've got more than a decade of hindsight to work with. How is that different from making a sequel to a game that's only been out for, say, a year or two? Where do you even start? Do you second-guess yourself more? Less?

Saunders: I don’t know that this is a universal truth, but in making TTON, we second-guess ourselves less. PST was a defining game for many, with its uniqueness being one of its striking qualities. We can’t really provide the same experience while also embracing the uniqueness players loved about it. We’re focused on creating an experience like the one PST provided.

We also have to remember that with so much time having passed, we will be compared with players’ sometimes-distant recollections of PST – not even the reality of the game they played! This is a greater challenge, I think, but on the other hand, it means we can more freely do what we deem best for this game instead of getting bogged down with trying to directly compete with a classic. We defined what we think makes a Torment game very early in the process - per the four pillars described during the Kickstarter - and it is our vision for the game that guides our decisions.

RPS: Can you give me any basic examples of companions you've come up with? How are yours different from the rather wide gallery of archetypes many RPGs pull from these days?

McComb: In general, we’re trying to think of our companions as family members, and fitting them into the party in that relation to the PC. We’ll give the player a choice of two companions at the start of the game. Due to a side effect of certain of the player’s abilities, which can be honed throughout the game, the player can pick only one of these two to adventure with - or can pick neither.

Since they appear almost at the start of the game, I think it’s probably safe to reveal that one of them is a fallen Aeon Priest whose arms are covered with intricate, living tattoos. He can pull these tattoos from his body and siphon their energies to create a blazing wall of force or a spray of acid. The other companion is a brash nano, a mad-scientist type whose recklessness has turned her into a series of quantum shadows echoed through multiple realities. Either of these can act as a mentor, a replacement for the father-figure of the Changing God.

More bluntly, while our companions might contain echoes of archetypal companions, we intend that they’ll prove themselves different in the execution.

RPS: There's a temptation, I think, to equate "different" with "better" in games like this. But sometimes it's not. Trying to avoid every trope possible can even be detrimental in certain cases. Have you run up against any situations like that? Are you ever tempted to fall into a mindset of, "Oh, let's subvert this by doing that, and that by doing that, and etc etc etc" instead of, "OK, how do we make this good regardless of what we are and aren't subverting"?

Saunders: We aren’t seeking to subvert tropes in an explicit way – that is, we aren’t combing through known tropes and coming up with the ones we want to play with. And we recognize, too, that trope subversion is now itself a trope. How it does come up is through our embracing ideas that, if we were making a game for the mass market, we might have considered too bizarre or unexpected. And we’re also more often rejecting ideas that we feel are overdone.

RPS: How much does having George Ziets on board full-time impact the project? How specifically? Why be lead area designer instead of narrative designer, a role he's occupied on previous games?

McComb: Kevin told Adam and me that George is a great designer. I have now seen that firsthand. His first turnover was incredibly detailed, remarkably living, and provides a touchstone for any future area designs for Torment. His existing relationship with Kevin has allowed him to transition seamlessly onto the team, and he completely gets the vision. He’s a strong advocate for Adam and me in-house, and he’s able to provide direction and feedback at both a high level and a micro-level. His experience, his thoughtfulness, and his attention to detail all combine to make him an incredibly valuable asset, and I’m honored to be working with him.

Ziets: To address the last part of this question - some of the reason, at least, is due to my personal preference.

For a good part of my career I was more of a generalist designer, and from one day to the next, I’d work on a variety of tasks – designing quests, laying out levels, writing dialogue, etc. It was only later that I began to shift into a more specialized narrative design role, which in practice meant that I spent nearly all my days writing dialogue. But I’ve always found that I’m more creative and happier when I can switch off between tasks, rather than focusing on just one thing. So the lead area design role seemed like a good fit – it’ll give me a chance to lay out levels, design quests, and think about combat encounters, in addition to the writing tasks of a narrative designer.

RPS: What's your process for designing an area? What do you prioritize? Can you give me a concrete example of how this plays out in an area you've designed so far?

Ziets: I’ll address this from the perspective of designing a zone, one of the large sections of our game that contain many individual locations, or “scenes”. For example, in Planescape Torment, the Hive could be considered a zone.

My first step in designing a zone is always to read the design constraints provided by our narrative lead, Colin. This is a usually short document that describes the high-level vision for the zone, as well as the story events that need to happen there.

With that in mind, I usually plot out the main narrative for the zone first. How can I string together the critical story events into a fun and interesting experience? What will the player’s main objectives be? How can I provide multiple paths for the player to achieve them? Simultaneously, I’m brainstorming cool locations for the player to explore and how I can incorporate those into the zone narrative.

Once I have a rough plan for the zone narrative and locations, I start fleshing everything out. I spit out as many cool locations and characters and encounters as possible, trying to find ways to fit them all together, and I don’t worry much about scope or prioritization. If the narrative is cool and exciting enough, it’ll spawn lots of ideas for side content. I try to craft side content that arises naturally from the overall context of the zone – the story, themes, and conflicts – so that the whole zone feels like a coherent whole. This is a messy stage in zone development, when some of the ideas I initially thought would be critical get pushed to the side, while other (more exciting) ideas arise and become more central. But it’s all on “paper” at this stage, so it’s a great time to iterate.

When I finally have a coherent plan for the zone, with lots of characters, quests, and events, that’s when I start thinking about prioritization. We have three levels of prioritization on TTON. A-priority refers to critical content that will be implemented no matter what. B-priority is not as critical to the core experience, but we assume we’ll make it, and we include it in the production schedule. If we find we need to cut content for a specific zone in order to achieve our quality bar, then some B-priority content may be left out. C-priority content would be nice to have, but we assume we *won’t* make it, and we don’t include it in the schedule. In other words, C-priority content is pre-emptively cut, unless we have time later to give it a last-minute reprieve.

So I think about all the content I’ve laid out on paper, and I ask what’s critical to the main narrative of the zone including multiple paths to complete it. That content becomes A. The rest gets prioritized as B or C. It’s a somewhat brutal process, since I just poured life into all sorts of characters and locations and events that I now have to de-prioritize… but when you’re developing games, you need to have thick creative skin.

Here’s how it all played out for the Bloom. As always, I had a high-level conception of the zone from Colin, as well as story events that had to occur. Since the Bloom is a somewhat unusual setting, I spent a lot of time just thinking about how the Bloom worked. The whole zone is a living thing that creates portals to other worlds… and it’s also highly dangerous because it’s a predator that eats people, thoughts, and ideas. So how would inhabitants interact with this thing? How did they survive inside it? What would players need to do to navigate the city-creature, learn about how it worked, and potentially manipulate it to get what they want? What different ways could players do that? I also came up with various factions and characters who lived in the Bloom, thought about how they’d interact with one another, and how they’d become involved in the player’s narrative.

Once I’d figured out the basics of how the Bloom worked and who lived there, I crafted the player’s narrative around them, making sure that the Bloom itself was a central “character” in the zone. Then I started fleshing out the locations and “neighborhoods” within the Bloom, populating them with characters and quests. Since I already had a solid foundation for what this place was and how it worked, ideas came fast and plentiful. Ultimately I had far more than I needed.

Finally I prioritized everything, spending a lot of time determining which elements were critical to the main narrative, which would lead to the coolest events and encounters, and which hooked into our themes most effectively. To give an idea of the final breakdown – I started with twelve scenes in the Bloom. Eight of them contained some amount of main story content, so I marked them as A. Two others were not crucial for the story, but they further explored the narrative for the zone, so I marked them as B. Two more were really cool, but the leads all agreed that they would cause the least damage to the zone if cut, so they were marked as C.

RPS: How much do you look to PST's areas specifically when working on your own designs? The influence was clear in Neverwinter Nights 2: Mask of the Betrayer [which Ziets was creative lead on], but how conscious was that influence? How much did and do you say, "Let's make this kinda Planescape-y"?

Ziets: This may be surprising, but on Mask of the Betrayer, I think I was more consciously looking to Baldur’s Gate 2 for inspiration, at least in terms of area and quest design. I’d say the Planescape influence was stronger for NPC dialogue, though.

For this game, the influence of PST is very conscious. We’re specifically making a thematic successor to the first game, and it’s important that we evoke the same weirdness of PST. So while I was designing the Bloom zone, I was also re-playing parts of PST… not to borrow specific elements, but to tap into the overall mood of how characters and quests should feel. I sometimes started my day by playing a few minutes of PST, and if I felt like I was starting to lose the Torment vibe, I’d jump back into the game for another ten minutes or so, just to get my mind into the right space. It’s not too dissimilar to previous games where I’d listen to certain music while designing a level or writing dialogue, just to put myself into the right frame of mind.

Fortunately, this process seems to have worked, at least for the Bloom. When others on the team read through the zone documentation and played early versions of the zone, “this feels really Tormenty” was a frequent comment. So… mission accomplished! Now we just need to make sure we execute it well and do the same thing for the rest of the game.

RPS: When are you hoping to enter production? How long do you think it'll take?

Saunders: Well, I adjusted the December 2014 delivery date even before the Kickstarter campaign ended. It was clear even then that, with the generous support we had received, we’d want more time to complete the game. In general, spreading development over a longer period of calendar time yields a better quality game than having a larger team for a shorter time. Since the “delay” was obvious, I wanted to be certain to tell the backers while it was still easy for them to retract their pledges if they were unhappy about that. Besides just seeming like the right thing to do, it was also a statement that we would be placing higher value on Torment’s quality than on its timeliness.

Having the opportunity to iterate on Wasteland 2 for longer has meant a longer preproduction period for Torment, which has been only a good thing for everything, except the release date. We’re in a “limited production” phase now, with real game content being created, but with our focus still on increasing efficiency and not generating quantity of content. People from our production team are gradually shifting over to Torment from Wasteland 2 and we’re beginning to gain more momentum in terms of making things. We recently announced to the backers that we now expect to release Torment late in 2015 – overall the response has been favorable, with people caring more about the final quality than the final date.

RPS: I know this is a long way off, but do you think you'll support Torment with bonus content after you release it? Or is your goal to create a complete experience first and foremost?

Saunders: Our goal is to create a complete experience. We do plan to support the game after launch, though that is more likely through ongoing bug fixes and minor improvements as needed. Conceivably we might add back in any B-priority content that didn’t make it and C-priority content that still feels like it would improve the player experience of the completed game.

RPS: What does one really long interview matter?

All: *tacit acknowledgement of really dumb joke*

RPS: Thank you for your time.