This week, I finished Zero Time Dilemma, the third and presumably final game in the Zero Escape series, and it got me thinking about one of my favorite subjects: the difficulty of pacing in games. Zero Escape isn’t the most niche series, but it’s fairly obscure so I’ll go through a brief introduction.

The ZE games are often classified as adventure games because they feature puzzle rooms, but fans know the truth. They’re visual novels. In fact, the core mechanic that sets the ZE series apart from other adventure games is a play on the core visual novel mechanic of branching story lines.

The core story of every ZE game is the same: nine people are abducted by a hidden figured named “Zero” and forced to play a game to survive. They are trapped in various rooms with various other team members and must solve a puzzle to survive or escape the room. Occasionally they are forced to either turn against each other or risk betrayal to stay loyal. Think Saw. Someone in Japan really liked Saw and decided the conceit needed time travel and telepathy.

Like most visual novels, ZE games multiple paths and multiple endings. Most paths lead to death, and it is impossible to get the good ending (where most people live and the mystery is solved) on your first or even second, third, or fourth playthrough. The reason for this is the central conceit of ZE–your character needs information from multiple time lines to fully unravel the mystery. ZE protagonists all have the ability to either “SHIFT” or send information through the “morphogenetic field.” These are two different things in-game, but they both amount to the same idea–once you see the events of one ending, your character knows it when you re-play the game or go back and make a different decision to try and get another ending.

To give an example, your decisions may end up with your character trying to defuse a bomb. The bomb requires a password, which you don’t know (you as the character or you as the player, the game actually does both). Your character dies. You go back and make new decisions to get a better ending. You still get a bad ending–everyone gets sick with a time-accelerating disease–but along the way you learn the password to the bomb. You then go back to the timeline with the bomb, defuse it, and advance the story.

This happens a lot, especially in the second game, Virtue’s Last Reward, which introduces a flowchart that allows you to jump around to all the different decision points you’ve reached and skip dialog you’ve already heard. It plays faster than it sounds.

The story, and how the player affects the story via seeing other time lines, is the core appeal of the ZE series. Everything else is a bit underwhelming. Each game has a few characters that are underwritten or dull. In general, the cast pales in comparison to the very adjacent Danganronpa games. The delivery of the (otherwise good) story is bogged down by pseudoscientific tangents that would make Hideo Kojima blush. It’s fun the first time someone waxes on about how the brain could be like a computer terminal tied to a central server, but these tangents are often repeated ad naseum and there’s no way to skip dialog you haven’t heard–even if you just heard another character saying the same exact ideas. And the puzzles range from stupid easy to bizarrely obscure. What really makes the ZE games work is the narrative and the way it plays with the visual novel convention of the branching path.

This brings me to the conceit of Zero Time Dilemma and why it ended up something of a disappointment.

Don’t get me wrong. ZTD is good as the third game in the ZE series. It was a game we absolutely needed after Virtue’s Last Reward and does an admiral job of tying up a ton of the series’ mysteries. A lot more was resolved in the story line than I expected, and better than I expected. The only problem was how it was delivered.

In 999, the first ZE game, every new playthrough was just that–a new playthrough. Once you got an ending, you had to start all the way over and do it again, making different choices along the way. You could speed through dialog you already saw, but it was a tedious process. Virtue’s Last Reward added the flowchart mentioned above.  You could go back to any decision point in the game you already reached and replay from there.

Zero Time Dilemma takes it one step further. The nine abductees are divided into three teams and from the start you can choose who to jump to. You can also choose when to jump to. Multiple scenes with each team, scattered across the various timelines on the flow chart, are available from the start.

How does this make sense in the story? Well, at the end of every segment, all the characters are injected with a drug that makes them forget the events of the segment. So it doesn’t matter what order you play in. Each time they wake up, the last thing they remember is the end of the first segment. They don’t know what (if anything) happened in the interim. Each section plays out like its own little episode, and you as the player have to piece together how they all fit in each respective timeline.

If this all sounds confusing, I’ll boil it down simply: outside of the very beginning and the last ~sixth or so of the game, the player decides the order in which the story is presented. And the game goes out of its way to obfuscate when (and on which timeline) each segment takes place, almost as if to prevent you from playing in any particular order.

It’s a bold move. I loved it at first, for audacity if nothing else. But then I started to get annoyed. The format basically prevented any character development. Scenes were robbed of drama because I didn’t understand what was happening or what had happened before. That confusion was intentional. It just sucked.

The biggest problem games face as a storytelling medium (outside of developers writing with no experience) is pacing. I know I’ve written about this before, but controlling pace is nearly impossible in games for a writer or developer. And controlling the player’s focus is almost as hard.

My go-to example of a pacing problem is The Last of Us, in my opinion an excellent game that manages to have an excellent narrative despite itself. The story in TLoU is broken up with various enemy encounters with a wide range of difficulties. There are a handful of encounters that are serious difficulty spikes, which can lead to several deaths and forcing the player to repeat the same bit of game play over and over again. This completely strips the narrative of momentum and the repetition of a sequence can sour any emotional response evoked by the story surrounding it.

Uncontrolled focus can also ruin pacing. Optional objectives and gameplay freedom will destroy any sense of immediacy or drama unless tightly controlled. At the end of Final Fantasy VII, a giant comet is about to smash into the world and Cloud is the only person who can stop it. However, I managed to breed several generations of Chocobos in that time, which took several real-world hours and presumably at least an in-world decade. That’s an extreme example, but illustrative of a problem any kind of game can have, as long as that game allows players to decide where to go next.

Visual novels typically eschew giving so much control to the player that pacing and focus aren’t a huge problem. On one hand, Zero Time Dilemma is relentless in this effort in-game, giving players no way to even skip through dialog if they read faster than the voice acting. On the other hand, the whole ZE series interjects a bit of pacing concern with its puzzle rooms. A frustrating puzzle or riddle can stop momentum just like a frustrating action sequence. But that is completely overshadowed by the way ZTD in particular gives up control of its pacing and focus with the extra layer of non-linearity I described above.

For a game that entirely relies on its story, this is a problem. The game gives no indication of how long each sequence is, how important it is, or where it fits into the narrative. You can stumble into an important, dramatic scene. Or you can run into a whole stretch of disjointed sequences. Outside of choices that lead to immediate death within a sequence, you rarely get to see the consequences of your actions unless you happen to pick two sequences that happen in the same timeline.

I don’t have any problem with nonlinear storytelling in general. And I also don’t mind games that let you switch between teams or characters with interwoven stories. But combining the two–and not even curating the nonlinear order at which players see each team’s sequences–abandons far too much control over the story, especially when there are sequences that are less interesting, and sequences that go nowhere without a password from another sequence.

Most of the time, I am loathe to use movies to explain games because I think that’s a disservice to both mediums. But I can’t help myself. Imagine if Pulp Fiction’s sequence order wasn’t curated? If you could accidentally start with Butch and Marcellus’s imprisonment at the gun shop? That section of the film barely works as is, but without the context behind how the two characters ended up together (and the time we’ve spent with the more interesting characters) powers us through.

I understand what ZTD was trying to do. I’m always trying to think of ways to merge interactivity and storytelling, and the ZE series has been pushing that forward with each iteration. But unfortunately, this newest elaboration undoes a lot of the pleasure of the series. Not only do plot points lose impact because they aren’t deliberately paced or sequenced, but the fun of experiencing the branches and changes in the story are gone because early ones are robbed of context.

The memory wiping of all characters between sequences also frustrates any character development. This exacerbates the series’ long-running problem with the non-central team members being rather dull. Even the interesting characters (the protagonists of 999 and VLR) aren’t allowed to develop further. All of this put together makes the early-middle portion of the game feel rather meaningless, even if you know by the end you’ll have enough to piece the story together.

In the end, I still enjoyed my time with Zero Time Dilemma but mostly because of the final portions of the game that (by necessity) are intentionally sequenced. Characters don’t lose their memories between sequences because one (or more) is shifting between timelines intentionally. The story gets wrapped up, there’s the usual insane viewpoint twist, and all the right pieces fall into place. But I can’t really recommend ZTD. Why? Because if you’re already into the series, you’ve already played ZTD or you’re going to play it no matter what. You need to play it. You need to see how the strange stuff at the end of VLR ties back together.

If you aren’t committed to the ZE series? Zero Time Dilemma is going to make literally no sense. It completely assumes you’ve played the other two games. In fact, context from the other games is especially necessary because of the nonlinear gimmick described above. Without that context, all the characters will seem like boring ciphers for the first part of the game rather than just a few of them. I don’t think I would have made it past the the long stretch of out-0f-order sequences if I didn’t already have a reason to care about what happens to Sigma, Phi, Junpei, Akane, and Diana.

The good news, though, is now I can fully endorse picking up the series, because (outside of the iffy puzzle design) my only real caveat on suggesting it was the lack of an ending. VLR ends with way too many mysteries for a game that didn’t definitely have a sequel in the works. Fortunately, for all its problems, ZTD provides a satisfying conclusion that wraps up almost everything in the series.