Josh Sawyer On Making A Detective Game Where You Never Solve The Case
Pentiment is a game where you never solve the mystery. How did the creative director of Fallout: New Vegas and Pillars of Eternity make it satisfying?
Josh Sawyer views his games’ career in a pretty holistic manner, and recognises how difficult some of the circumstances were under which he created so many genre-defining hits. “For the first half of my game development career I was developing games in the way that
Fallout: New Vegas was - with someone else’s technology, someone else’s assets, and very fast.”
Despite being something of a modern classic in 2023, when New Vegas first came out in 2010 it's safe to say its reception was mixed. It had an extremely tight development cycle of only 18 months, and while it reviewed well, it was slated for performance issues and dated tech. Sawyer says this is down to using an “old engine”, as well as admitting “we were not as experienced using [Gamebyro]. How could we be as experienced using the engine as
Bethesda?”.
These short time frames were nothing new to Sawyer. Two of the most well-regarded games he worked on before New Vegas were Icewind Dale and Icewind Dale 2, both of which used
BioWare’s Infinity Engine, and came together in 14 and ten months respectively, “which sounds crazy, but it’s true”. A far cry from his latest game,
Pentiment, which spent nine months in pre-production alone.
Sawyer doesn’t talk much about how these ludicrously short production timelines probably led to a lot of crunch, but he does touch on the lessons learned from them. How these constraints “teach you to be extremely focused on making content very fast and being able to be very creative with what the engine can do”, or in other words he says learned to not fight against his limitations, “because you will die”.
It’s a marvel just how much content, be it lines of dialogue, quests, interlinking systems, exist in these games considering how little time they had to make them. However, Sawyer thinks that choice-driven RPGs like Fallout don’t lend themselves well to the reviewing experience. Things that
Obsidian considered valuable to the core of these games are just not things you encounter naturally on a review deadline. You’ll miss the experience of roleplaying vastly different characters and seeing how they impact the world, and vastly change the journey fans would eventually embark upon. Instead, most reviewers could only “play through the game once and didn't talk to anyone else about it”, meaning they missed out on a lot of why people love these games so much.
Sawyer thinks many playing it just before release would have seen New Vegas “in the worst light, you’re seeing it as a game that looks old when it comes out. You’re seeing it with a lot of weird bugs. You’re seeing it made by a team that is not as good at using [the engine as Bethesda]... And you’re rushed.”
Most controversial about New Vegas’ critical reception was its Metacritic score of 84, which reportedly was one point below Bethesda’s target for the game and enough to withhold bonuses for Obsidian. For all the bluster around it at the time, Sawyer neither confirmed nor denied if the story was actually true, just saying that it was something “the fans cared more about than we did.”
For all the game’s quirks you can pick apart, Fallout: New Vegas is recognised these days as a masterpiece. However, Sawyer thinks the skills he’s learned from years of getting games done in short timeframes has backfired in recent years. “You can find yourself self-sabotaging, saying, ‘well, let’s do everything in the most efficient and simple way because we don’t have time’, [but you have to say to yourself], ‘No calm down. You have time to figure it out, use pre-production. Actually solve problems.”
Even then, Sawyer says you can be a bit lost. “You have to know what battles to fight and what not to fight. It takes a little bit of time to understand, ‘is this worth spending time on?’”. While it might have been freeing to have more time, the pressure of working on two Kickstarter-funded games,
Pillars of Eternity and
PoE: Deadfire, at a then-independent studio seemingly took its toll.
When news broke that Obsidian was being acquired by
Microsoft in 2018, Sawyer seemed to jump at the chance to work on something very different. Obsidian had been around since 2003, and he’d been there since 2005. After all those years Sawyer wanted to do something radically different and personal to him where he could draw on his art history background. “It wasn’t like making an offer he can’t refuse,” presumably referring to Obsidian’s CEO Feargus Urquhart, “but I said, ‘look we’re being acquired by Microsoft. I’ve done a lot of big projects for the studio. Really, I just need to do this.”
Sawyer is frank in saying that he was burned out after Deadfire but tells me he said, “I don't need a big team, I don’t need a lot of money. But I do need the freedom to do this because I don’t know that I can direct anything else.” He was, however, quick to make it clear that this wasn’t an ultimatum but “if [Urquhart] couldn’t have been able to support it, I don’t know if I would have stayed, because I don’t know what I would have done.” Sawyer adds, “I couldn’t go back to shipping a big game at that point. Maybe now I could, but it wasn’t possible for me then.”
Pentiment is an original narrative adventure that focuses on medieval art, class boundaries, being a really crummy detective, and the stories we tell to the next generation, and launched last year to critical acclaim.
It wasn’t the easiest sell - it had no established foothold or niche, it asked for a lot of research and depth, and Sawyer even admits that if art director Hannah Kennedy wasn’t aboard, he might not have made Pentiment at all. Once she was, Sawyer was able to convince the rest of the team of its narrower historical setting, the detective concept, thanks to the striking period-appropriate art the game offered.
That’s not to say development was easy. By the time they were in the depths of production to nail the historical accuracy of the time - Sawyer found himself translating “line by line” a book from 1910. This wasn’t even for the main plot but for background about hunting in the Late Middle Ages which was not only written in German, but in Fraktur, a calligraphy handwriting style which would just look like fancy scribbles to most of us.
Sawyer’s work on Peniment sounds methodical and focused in ways that would have been impossible on previous projects. The question is now that he’s wrapped up his dream project, what’s next? Sawyer is still at Obsidian, working on something, so I had to ask him if getting to take the shackles off and make something so niche and personal would affect his future work. “The important thing is that you really need to believe in the thing that you’re making, and compromise as part of creation.” The way he explains it is that “you can still retain unusual things, you can still retain outright weird or off-putting things, but you have to put them in the overall experience in the content.”
Penitment is notable as a detective who-dun-it with no canonical murderer. Several people could have done the deed, and you will condemn someone to death no matter what, but it offers no canonical right outcome, and none of your choices are necessarily correct. Hell, none of them even leave you feeling particularly good about the role you played. It’s an experimental game at its core because you will never know everything needed to solve the case, no matter how many times you play. It’s a far cry from Fallout or pretty much any other mainstream RPG, where if you have leveled enough into a certain skill you’ll get the best outcome.
If Sawyer was to return to a bigger type of game, would these elements remain? It’s one thing if a reviewer or player can’t appreciate all the content in one playthrough, and we only realise how in-depth it is thanks to years of hindsight. But how would a modern triple-A game be received if you were left with no right answers?
“We’re not shipping our heads. We’re not shipping ideas. We’re not shipping design documents. We’re shipping a whole experience,” he says. “Sometimes you get to put in all your cool things, and sometimes it just doesn't work [because] your audience’s either too broad or there are too many other limitations.”
Does that mean the next Obsidian game will be less sprawling and ambiguous, in the name of your first playthrough showing you everything or having a definitive outcome? Or does it mean Sawyer is looking to find ways to leave questions unanswered and leave the player still satisfied?
“You don’t like hearing anything [you worked on] trashed,” Sawyer says, but he’s glad and surprised a few years down the line a game can, “become really beloved and people do see all the attention and care that we put into all the different ways that you can play.” Maybe Obsidian’s games just defy the review framework, and as they become less interested in the best solution or correct outcome we will only really be able to appreciate the studio's brilliance the more time passes.
