Prompted by a bit of fantasy-novel induced research about fiction writing, I went and tied up some loose threads in the SJ plotline. There shouldn’t be so many loose odds and ends anymore. Some things had to be a little better explained, some characters’s storylines had to get a sense of closure and some conflict was added for heightened tension. Inasmuch as a videogame plot allows for that, at least. It’s a three-act structure with between nine and twelve noninteractive scenes each, plus a bunch of interactive ones that don’t take control away from the player.
I thought this might be too much – that it would perhaps drown the interactive part of the game in heavy storytelling – but I looked at some other story-driven games and found that they tend to have up to five hours of noninteractive content. That’s ridiculous. SJ shouldn’t nearly reach that amount of cutscenes. A lot of the ones it does have aren’t very long. I currently estimate something like 55 minutes of cutscene per act, but that includes a few pretty long ones such as the introduction, the key story sequence in the middle of the game, three miniboss/boss fights, and a long ending sequence. The rest are one- to four-minute scenes. The game will allow you to skip cutscenes.
Gameplay is roughly estimated to be minimum 5-6 hours if you know what you’re doing and only barge through the main storyline without ever taking any quests, not bothering to build up Scout’s skill tree or to revisit any of the semi-sandboxy game environments. If you really relish in it, it should be three or four times that, if not more. And since first-time players usually take some more time (and the game isn’t meant for a habitual shooter audience), these estimates might be on the low side.
So I think I’m fine. Which is a nice surprise.
It was originally structured “early game” – “mid game” – “end game” but I noticed that the early game was too long. When I tried to apply the classic three-act structure, it fit immediately in a very natural way, as if it had always been there, which was a happy coincidence. I’ll take that as a sign that I did something right.
Each scene has already been matched to a certain spot in the game’s progression. A lot of them take place in faction camps that are situated a bit out of the way of the gameplay. Others are tied to major events or to things like picking up a key or accomplishing goals in the main questline, so a good number of them should come as rewards for the player. The placement of triggers in the levels has largely been finished, apart from one level that is concepted but needs to be blocked in yet. It’s pretty tight but there is a lot of room left to gameplay and simply scavenging around or taking quests. Roughly one scene per major new area, which means as long as you don’t proceed down the main line, or spend time scavenging or questing in already-known areas, you won’t see any cutscenes until you choose to continue the main quest.
spawnhost 
July 29th, 2015 at 11:44 am
6 hours to beat the game when rushing? You’re making the real deal here! I think the longest game I made took the average tester 10 minutes to beat when given infinite lives. I imagine you’re pouring over a thousand hours into making this.
Thanks for sharing your experiences with setting up your cutscenes. It’s not something I’ve had to deal with yet but I learned a couple things I can take with me. I’ll try out that 3 act structure on my first game with a story.
July 29th, 2015 at 6:34 pm
It’s a learning experience for me as well, after all I’m not a fiction writer by trade. The story really began with some characters I had invented for the game, and these sort of had stories attached. Before that, it was a very brief story (half-life type of plot.)
Regarding length of play, well, I’ve made large single player levels before that took people 30-50 minutes to play through despite being simple FPS levels. You can do the math, if you take 5-6 of these levels and make them slightly bigger and add a story and a main quest line, the player will have to go back and forth a lot more and much of the playspace can be cleverly reused (remember Nethack, you had to go all the way back to the beginning after getting the amulet, to access the endgame, and you can do this in a more concealed fashion in other games as well.) Resource management is another thing that increases play time, if the player has to go and find a medkit before pushing on it also takes more time. SJ has a full loot system, so the player will have to scavenge quite a bit for necessary resources, and finally, the stealth and exploration mechanics also make a game much slower than if it was a fast shooter.
Lots of things contribute to play time. Basically the player’s path needs to be looped through the environment, with some reuse of certain areas, instead of being all linear. Many things can slow the player down. And that’s before side quests enter the picture.
Also, my levels are just physically big to start with, which halfway makes up for the game only having 5 major areas. If it was a linear shooter and the player moved at Quake speed, you could probably blast through it in 3 hours max. But with the stealth etc, it’ll be closer to 6. A lot of sneaking and checking out dark corners for loot.
I’m glad if my learning process is interesting for others.
August 8th, 2015 at 10:39 am
Very interesting stuff. Something for me to seriously consider as I approach when I’ll go for a full-length project. Thanks again