Wednesday 1 July 2009

A History of violence. Part 3

In this blog post I will focus an underused combat mechanic: Chase Sequences. This type of "combat" is very common in horror movies, but quite rare in horror games. I will briefly discuss how we used it in Penumbra, problems it causes and how some other games have implemented it.

In Penumbra we used chase sequences on three occasions in the entire series and and each time it was a highly scripted event. There was always little room for the player to move in and mostly a very clear path. The first two chase sequences were in Overture and both involved being chased by a giant worm. In both of these the player had to complete a specific obstacle along the way and failure resulted in death and a restart of the chase. While a lot of people said that they liked this, a large amount also disliked this portion of the game because of its trial and error nature.

In Black Plague the chase sequence was a bit different. Here the player had more space to move about in, the goal was not as clear and we also deliberately tried messing with the player's head. The Overture chase sequences focused more on creating fear, while this had a larger focus on confusing and disturbing the player. Like the worm chases, some players really liked the sequence while other did not liked it all. Here the problem was that some people did not found the right way quick enough and the terror experienced was transformed into frustration.

The reason for people not liking the chase sequence is pretty much the same in both of the above instances, and probably also the main reason why this type of combat is so underused: It requires a very scripted environment, has a very strong trial-and-error feel and loses much of its impact and fun-factor after the first try. In order for it to be entertaining, the player must not repeat the sequences too much.

Some example from other games with chase sequences includes Clock Tower and Beyond Good and Evil (which is not a horror game though).

In Clock Tower chase sequences replace normal combat to a certain degree. When an enemy appears, the player must either hide or escape by using scripted interactions with the environment (throwing plants, etc). This is probably the most fateful adaption horror-movie chases I have played in a game, but it comes with lots issues. First of all, the chases loose their appeal very quickly as there is no real reward for avoiding the enemies, so enemies become more of a chore as the game progresses. Also, to avoid players from repeating hiding/avoidance patterns hiding places and offensive environment pieces stop working after some uses, sometimes making it very hard to escape an enemy and leaving one running around in circles trying to avoid the threat. More problems exist, but to sum it up the main problem is that it is very hard to keep this kind of generic chasing interesting and it is therefore not a very good mechanic.

Beyond Good and Evil's chase sequences are very scripted and at times similar to quick-time events. The sequences are also very much based on trial and error, but have very frequent check points removing some of the frustration. The chases are quite rare too, so they never get too annoying and always seem fresh. They can still get very frustrating though, and like the chases in Penumbra, enjoyment depends on how many tries it takes to complete them.

Finally, a Silent Hill 1 remake is going to be based around some kind of scripted quick time chases instead of combat. It will be interesting to see how this turns out and how they will manage to keep the chases from becoming frustrating trial and error based Dragon's Lair like gameplay. Judging from previous chase sequences in games, it seems like a risky decision, but it is always fun that developers dare to try new things.

What is your take on chase gameplay? Did you find the penumbra chase sequences fun or just frustrating?


  1. I dislike chase sequences in general because of the trial and error gameplay. Dying breaks the immersion, removes any fear, and gets frustrating very quickly as you play the same bit over and over. The ones in the Penumbras were all that bad (I only died a couple of times each), but I think I would've enjoyed the game more if they weren't in it.

    I believe the problem lies in the strict scripting of the sequences. Players are going to be advancing at different speeds and not everyone is going to figure out what needs to be done.

    Experienced players might be moving so fast that the chaser seems like a snail, but you can't make it faster because then unexperienced player will have an extremely hard time beating the sequence. This problem I'd probably solve with a mechanic familiar from racing games: the rubberband effect. Default speed is rather slow so that unexperienced players will not have too hard a time, but if you get too much ahead the chaser starts catching up. Admittedly this includes the danger that some players will figure out you don't actually need to go as fast as possible, but these are the people who would've been leaving the chaser in the dust anyway. It's not a perfect solution, but I believe it's better than using the same speed for all players.

    If you need to solve puzzles during the sequence, there's a very good chance some players will get stuck. It's annoying enough in normal situations, but having limited time will multiply the frustration. This problem I'd solve pretty much like any other situation where the player can't figure it out. Solution 1: Make the puzzle optional. Or, if there's more than one, make some of them optional. Solution 2: Give the player multiple ways to solve the puzzle. Solution 3: Make them encounter a similar puzzle earlier without a timelimit so they'll have a better idea what they should do.

  2. Well, as you have said, the key is to make it a scripted sequence, and to create more of a film-like experience. The player doesn't even have to be in any real danger (unless he/she decides not to take part in this roleplay-like gameplay, and just stands there). The player should mainly act as the game hints him/her to, run away, push some buttons to fight off monsters, trow things behind him and such. It's up to the dev team to devise an interesting sequence. This is, in a way, a guided play, though the player still has all the freedom. This ingame hints of what the player should do can be something subtle, like the arangement of the environment, or something less subtle like flashing doors (more subtle ways being preferable). The players have a tendency to follow these hints, because thats basically what they do most of the time while playing: a lit computer monitor in a rum will, for example, make the player check the terminal for important files. Why not use this in a more creative way with scripted sequences?

  3. Chase scenes in Overture was my favorite part of the game. It was great dramatic climax with lot of adrenaline. When I played it first time I did everything right so I wasn't frustrated by re-trying. When I played the game 2nd time it wasn't that great. I would like to see more of this. What Razalhague said could make chase scenes less frustrating to other players.
    In Half-Life 2 player always have a chance to explore combat area before combat starts. I think something similar should be added to the chase scenes in horror games. Player should solve some puzzles and explore area before something starts to chase him there.

  4. Wait, wait, wait, wait... I do not believe people actually disliked the Penumbra chases. Well, considering the wide variety of inteligence in the human race, I actually do now...
    Anyway, for me, the Penumbra series chase sequences were incredible partly due to the fact that I completed it on the first go. I'm guessing that the players who told you guys that the chase sequences were "pure trial and error" forgot to mention the shock they got at first, which, if they were like me, scared the hell out of you and made your heart thump. Most games get frustrating when you have to do the same thing over and over, it doesen't get better or worse with chase sequences.
    In Overture, the worm scenes were incredible: they came out of nowhere, they made you jump, they made you run like hell and they made you nervously fumble for your pickaxe to try and break down the obstacles. That first jump made me consider the whole thing a work of art.
    Now, in Black Blague, the unique chase scene was another piece of art: confusion, fear, near-death... It all mixed up into one single purpose: run! Like I said, even if people say they found it frustrating, I bet they didn't expect it and, in a certain way, I can see that it needed to be frustrating. If not, how were you going to create disorentation?

    All in all, what I'm trying to say is that chase scenes don't necessarily have to be fun to repeat more times or extremely easy. You did an incredible job when you created those scenes because you did not expect it and the first try was always in the bullseye of the emotions you wished to generate with those scenes. The second try maybe not but that's because Penumbra is just like a horror movie: when you know what's coming next, you don't get scared.

  5. Windex here, and I loved the chase scenes. They were a great change in pace between the repetitive levels, however I felt Black Plagues chase scenes weren't as good.

    In my mod The Shaft I dabbed in 2 years ago, I had a chase scene being made in one level. The player is being chased by a wolf, who is breaking down doors behind you. Eventually you find a dead end - a random office room.

    The wolf eventually gets to your door, and start ramming against it. The second time it does so a vent cover breaks off in the room. The player would have to move objects to get into it, which worked really well.

  6. I have not completed Black Plague yet, but I did find the second Overture chase scene to be slightly repetative after a few tries.

    In contrast to a number of the above comments, I would suggest moving away from static, scripted chase scenes. Not that every enemy encounter needs to become some sort of run-for-your-life experience, but perhaps let the player have some other options than pre-scripted events.

    For instance, instead of having the player simply die when overcome by a monster, have them go into a 'struggle' sequence where they might hit the spacebar a number of times to throw off and stun the creature, giving them a moment to run away. Aside from preventing the player from repeating parts of the game, it also prevents them from dying, which I believe drains a significant amount of fear from the game.

    Of course I'm just theorizing now, not having made a horror game myself, but I would expect the greatest deal of fear comes from the FEAR of dying and the immersion of the player thinking "*I* am going to die", not "my *player* is going to die." When the player dies, it may pull you from this illusion, because you don't die in real life. I also viewed the death-screens as a kind of let-down, because the black screen and silence kills any adrenaline you've built up.

    Anyway, I've rambled enough for now. I'm still enjoying Black Plague, and I hope my comments above haven't made it seem otherwise! I don't claim to be an authority of any sort of horror - just well-opinionated. :P Thanks for making such scary games. :D

  7. Another game with this mechanic is Haunting Ground (made by the same people as Clocktower). It is done rather well in that game, and I liked it better than Clock Tower 3.

    The chase scenes were fine, but I don't like trial-and-error mechanics in horror games, since doing the same thing over and over again kills the horror. If I have to do soemthing 3-4 times, when I load the game, I lose the immersion, remember that I'm just playing a game and just try to get through the sequence.

    The enemies that you had to run and hide from were good because they weren't scripted and there wasn't as much trial-and-error with them.

    A balance is good. Frustration is bad and should be avoided, but tension, fear, and creepiness should be nurtured in players.

    You guys are one of my favorite horror game developers (probably second only to Team Silent). You really seem to understand horror and know that it is more than adding 'scary' stuff to an action game.

  8. Note: There will be Penumbra spoilers in this comment.

    Chase sequences absolutely can work well, but I've yet to see a really good one show up in a game. I think that, if you're going to stick with the heavily scripted approach (which is probably the least work for the best impact) the trick is going to be to provide multiple solutions to every challenge point and to vary the speed of the pursuit to try to keep the player from failing; or else, to make the consequence for failing nonlethal.

    The idea behind the former is that, while players are good at spotting clues not everyone is going to see the right thing to go for right away unless your environments are really boring. The second worm chase in Overture, for instance, when you reach the room with the support pillar you need to take out; I went for the wheel that opens the door my first try, and the chase was tuned tightly enough that by the time I got Philip's message about it taking too long, there wasn't enough time left to try anything else. I died and had to repeat a bit of the chase to try another solution. It's the sort of trick that feels a little bit of shallow, but I think what I would do is make sure that every or almost every object the player can try to interact with during a chase would play into the scripting as a viable solution. As long as it keeps the player from repeating the chase they won't notice that any path would have taken them through, and the tension is preserved.

    Really it's the same problem all combat in horror games faces; the absolute worst thing you can do is actually kill the player, because it doesn't really cost the player anything to die and try again. The only way I've found that you can effectively hurt the player in a horror game is by depleting some sort of resources, because as soon as they die they hit continue and they are actively reminded that there's no real risk.

    Probably the real key is to keep exploring resources you can deplete without actually ending the game; yet at the same time you probably want the game to get easier with fewer resources and harder with more, so players of different skill levels don't become frustrated on the novice end or bored on the expert end. My best idea along these lines so far is metagame type rewards; access to new gameplay modes or models or somesuch for conserving your resources, but this feels like a pretty bad solution for a number of obvious and not so obvious reasons.

    Sorry for the long (and semi-tangential) response; I obviously spend too much time thinking about these things, and just finished Black Plague this weekend so it's all still fresh in mind. In short: chase sequences are good flavor when used sparingly, they were enjoyable but short of perfect in both Overture and Black Plague; I think the latter was a better design, but was not implemented quite as well. Consider providing multiple routes to success, to minimize the chances of a player spotting the wrong game object first and so having to repeat the sequence. Enjoy your work, and look forward to seeing what you come up with in Unknown.

  9. In my opinion 'static' or 'scripted' chase sequences should be avoided, as it changes the type of gameplay too dramatically and can have all the negative effects mentioned above, i wont list them all :p

    It would be much easier and more effective to spend time working on making more dynamic chase sequences, kinda like the dogs in the tunnels at the start of penumbra. If you made a wrong choice or got seen/made a sound then the dogs would start chasing you, then youd turn and run, and effectively that would be a dynamic chase sequence, which had many different types of endings depending on what the player felt like doing, the items they had, and the environment overall etc. That is a true dynamic chase sequence, and it worked very well. Several times, in black plague as well with the zombies, it got so frighting that I had to quit the game. I cant actually walk around in the tunnels at the start of Penumbra as well now.

    Thats what I call horror :P

  10. I liked the chase scenes in Penumbra, but yeah, the trial and error kind of kills it. I thought the same thing in Penumbra with the Giant Worm chase: after the first time, I thought "well now it is not as scary".

    I still had the adrenaline but not the fear.

  11. I really like the second chase in Overture. It wasn't that hard to figure out, and it was intense. However, I hated the spider chase. It took me tons of tries to beat, due to the luck element (oops, the right path leads into a spiders nest, too bad) and the strict timing. Also, getting repeatedly mowed down by 4 spiders is really annoying.

    Black Plague's was really cool (the one where Clarence is disorienting you, right?). I know there's enemies chasing after me, I'm completely confused as to where I am, nothing is familiar, and I felt lost and helpless. It was awesome.

  12. I had thought the worm chase in Penumbra Overture was great and exciting, but the one problem i had in it was knowing that you were supposed to hit the board-thing down with your pickaxe. The first two tries i thought i was supposed to turn the wheel and run through the door before the worm gets you, I think more emphasis upon the weakness of the board-thing might fix that problem (if it was a problem for other players anyways).

  13. I've only started playing Overture, so I don't know how well the chases were implemented, and it's pretty much the 1st horror game I ever played. However, due to my experience in GTA: Vice City I tend to agree with Alex Delderfield. Also, I think that high-power weapons like shotguns aren't necessarily bad, if the enemies were more resilient. Take the fantasy novel Brisingr as an example: the enemy soldiers were modified so they didn't feel pain, so they were able to continue fighting long after a normal person would have died. Shoot them with a shotgun, and unless you hit the brain or spine they'll continue fighting for another 10 seconds to 10 minutes, depending on what you hit. Do something similar with the zombies, and even a machine-gun won't seem all that effective, even though if you shot yourself, or a steel barrel, you'd see that it does have the expected firepower.

    Also, there is no need to make all enemies of the same type equally strong. For instance, if two crocodiles come to eat you, and one is clearly stronger than the other, the weaker one would back down, leaving you to fight only the stronger one. However, the weaker one would still hover around, and should you hurt the stronger one badly enough, it will attack you, while the stronger one sits back and waits for his chance. In the case of wolves, you could have coordinated attack from multiple directions. If the enemies are cannibalistic, then killing an enemy could attract more of the same kind, and will definitely attract enemies of another kind. Say you kill a zombie in an area that has more zombies, spiders, and wolves. In that case you'd have members of all 3 kinds swarming in to eat the corpse. Some will get a bite or two before a really strong monster comes and claims the whole thing to himself, and then all the rest of them will look for easy prey: you. Also, if they pass you on the way to the corpse, they will attack you. Since even powerful weapons like a machine-gun won't be able to save you from dying at the hands of 3 zombies, you'd have to run or hide if you want to survive. As you pass more areas, you'll be noticed by more creatures, thus making escape even harder, until you reach safety. This also means that killing, or even just heavily wounding an enemy will actually in many cases be punishing since it will attract more enemies who will try to kill you. This will also create chase scenes that are better suited to open environments, and with many possible outcomes, including ones that advance the storyline.


Note: only a member of this blog may post a comment.