Feel free to make your own little description of games and add opinions here and there ^^. If this all goes well, I guess this could go on the wiki to make this into a nice document for makers who are wandering about this concept.
++ Replayability And Length
++ True Linearity
A kinetic novel. A single story, and nothing you can possibly do by interacting with the machine to make it steer in any way. A movie or a regular book would be just as linear. Except for interacting with the interface, there's no real *game* interactivity.
++ Perceived Linearity
Most games with a linear story, but a gameplay that automatically makes things rather unique (platform games, strategy games, fighting games... anything, really). Strangely, these are probably the most popular games out there. There's a *lot* of interactivity. While you might not be able to change the story, the attraction is elsewhere. Just hear people playing a VN for the first time while being regular gamers : "Uh?... All I do is click and click? This sucks!". Well yea, the power to change stories comes with a cost I guess ^^.
++ Replayability
Here are examples of games offering replayability (or not...) in various forms. These are analysis of what they offer and how they implemented it. There will be as few spoilers as possible. Since the structure is what's of interest, it shouldn't be too much of a problem. Short summaries of the games will still be given, but nothing more than what you could learn from a review I think.
++++ Indigo Prophecy (Fahrenheit)
*Faking replayability*
It might not be very well known overseas and it's not a VN but an adventure game. You are Lucas Kane, a bank employee (working on computers there) and one night, at a restaurant, you are... possessed, somehow, and forced to kill an innocent man. When you can move again, what will you do? Hide the body and wipe the evidences? Run away as fast as possible? Try to get out as if nothing happened? It's up to you.
Well... that's what the huge ad campaign said about the game. Playing through it the first time was, I admit it, a total blast. The game is really well done, leaning very strongly towards the movie feeling. But when I played the second time... oh dear, what a disappointment! Everything... everytime I had wondered "ohhh... what if I didn't pick this choice?"... was a waste. Because the game tried *very* hard to keep you in its single track.
One of the worst examples I can find is a part where you need to gather clues. Understand that you play the role of Lucas, the "murderer", but also the role of police agents trying to find and arrest Lucas (interesting gameplay element). When you play Lucas, what you do can leave clues of the investigators. When you investigate, you try to pick up these clues (fun stuff). Where it really hurts, though, is that, when replaying, I realized "hey, let's try to leave as few clues as possible". Ok, so I do this, knowing the game well enough. Then I reach the point where I'm supposed to assemble clues to make the final proof. In my first playthrough, I had to assemble 5 elements to make the proof. But this time, I realize I could *never* have more than 4 elements. I kept thinking "no way, I just forgot one combination!"... but no... if you actually played the perfect fugitive, what do you get? Well the investigators only need to assemble 4 elements to make their proof... Neat.
Anyway, I don't want to make this too rant-ish, but basically, this game, while still being one of the best ambience based adventure games I've played, greatly disappointed me with these gimmicks. Whatever you try, the game just punches you in the face and brings you back to its extremely linear route.
It wouldn't have been a problem if commercials and author interviews didn't talk *that* much about "freedom" and all the "what if I did something else back then?" idea, when it wasn't actually implemented. Basically, this game suffered from bad advertisement and needs to be played only once if you don't want the whole thing to break down.
My opinion : the idea isn't bad. In each scene, you can actually do many things in any order, and omit some of them. While not impacting on the story or even the gameplay in any meaningful way, it makes the game feel very life-like (what other games lets you rummage through your fridge and drink some milk, or booze... take a shower... surf on the web, check your e-mails... order a meal at the restaurant, etc...). The fact they don't impact on an already great story isn't bad either... as long as you don't say they do ^^.
++++ Figures of happiness
*Branching stories*
Note : my memory of this game is a bit hazy... feel free to complete and correct this ^^
Figures of Happiness is a hentai VN. You play the role of a guy who quickly ends up as a ghost after making a promise. This game, while not absolutely memorable, still got to me for a few reasons. First, the story is rather interesting, but most importantly, it has a *lot* of branches. I once made a chart to be able to find all the endings and I was surprised to see I didn't have enough room on my piece of paper (a regular letter format). It has a rather long main branch which serves to introduce everyone, the whole problem of being a ghost and so on... This branch includes mostly dead ends (at least that's what I remember), and ways to meet a few girls outside of the two main ones. In any case, you'll reach the central point of the story.
You'll quickly have to choose between two new main paths, each focusing on a couple of girls. The good point in replayability is that each of these main paths is *very* different. It really focuses on a small part of the cast which lets a lot of things to discover the next time. Within these two branches, though, it's mostly a matter of making the right choices for the girl you want. They also include unique choices that lead you to special endings (with the supporting characters). All branches don't have the same length. Sometimes you'll end the game early in this second part, sometimes it will really last for a long time.
My opinion : when I played it, the game seemed really huge and with a high replayability factor. I hope my memory isn't playing me tricks on this one ^^;. The charts used were really tree-based. I don't think there were that many branches coming back to a main one. It was certainly a lot of work to make such a story, but it's probably the most enjoyable when you play it again since once you start seeing new stuff, it's up to the end of the game.
++++ Gloria
Gloria is an old hentai game taking place during the beginning of the last century. You play the role of a MIT (int the english version at least ^^) student being hired to tutor a girl of the Gloria family. The beginning of the game is rather linear (going to the mansion, meeting everyone) but it still offers unique scenes depending on your choices. The earliest ones can give you a quick game over, but I still found them amusing (those who played might see I mean). Once you're in the mansion, you start earning love points with the girls depending on how you talk to them. It doesn't change things right away, but it will definitely come back later.
The main thing with this game is that it's *very* difficult to know exactly what the consequences to your choices are. Fortunately, it's rarely frustrating and illogical... but at times, you'll get very unique scenes that do make sense, but can't really be predicted... and might not even be reached voluntarily. I'm pretty sure the game uses either a lot of variables, or a *very* detailed events tree.
Quickly, you'll have to choose which girl of the family you want to tutor. There's a bit of a fake choice here since you can only tutor two of them, but I think choosing one of the "impossible" girls first might still have an impact later. So, that's two main branches from that point. Each has a unique storyline and gives access to its own unique scenes. Sometimes you'll be able to woo one of the other four girls for example.
Another interesting point in this game is the ending. There are two parts : one deciding who you end up with (if anyone), and what the mansion you're living in becomes. It all depends on your choices. Both are not *totally* independant though. Basically, you can end up with any of the girls... or alone... and things can go well with the manor... or not... or even really well... And sometimes both parts are linked together (things go well and are related to the girl you ended up with). According to a FAQ, there are 48 paths (not counting very small variations that would also lead to seeing everything). I'm pretty sure each path offers at least something new... either in the form of an ending, or in the form of a new scene during the game.
My opinion : while I'm now somewhat bored of this game, I'm still amazed by its complexity. But it comes with a price: except for the first, what, 5 playthroughs, you won't really see much new stuff anymore. As I said, one or two new things per playthrough. It's a good thing because chances are high that you *will* have a little surprise with each game, but it's bad because you'd mostly be skipping through 99% of the whole thing.
Interesting interface note : my version doesn't have any skipping functionality. It can be somewhat emulated through keeping the Enter key pressed, but it will even validate choices this way and is a bit dangerous.
++++ Princess Maker 2
This is an upbringing simulation where you raise your daughter, given by a god/goddess, and try to make her a fine lady.... or not ^^;.
There isn't much story in this game, except maybe for the RPG part of it. Replayability comes from the large number of possible endings (54), each being a possible future for your daughter. Naming a few : cook, artist, nun, evil queen, great adventurer, magician, teacher... etc...
The way you raise her, what you make her do throughout her teenage life, decides what she becomes when she reaches her 19th year. Sadly, I find that she will mostly end up as a house wife, because you tend to have her do housekeeping jobs to startup.
The good point in replayability, apart from the many endings, is that the game has quite a few unique "domains". The most obvious one being the RPG part. Instead of sending her to school, making her work, etc... in the rpg part, you control her through forests and dungeons and have her fight monsters. Quite a change from her usually quiet life
. In the many times I played the game, I only tried the rpg part the last time. And I didn't go far because I didn't train her enough and didn't *really* want to go down that path. Just saying that, according to faqs, there is a *lot* to discover in that part. Little events, objects, and so on...There is also the "social" part of the game where you chat with people from the royal guard up to the king. This gives access to endings such as "the princess". It's not as big as the rpg, but you have to focus on it if you want results too.
My opinion : while not really a VN, of course, the game shows that many endings isn't forcefully good for replay value. 54 endings looks cool... but really, you won't play 54 times. Most importantly, without a FAQ, there's no way you'll get them all in only 54 playthroughs. I don't think anyone has that much time to waste nowadays?
On the other hand, there's this concept of having a few "domains" in the game. We could see these as "main branches" with the difference that you can go to any branch, whenever you want. A good VN example of this would be having a main branch per girl and allowing the player to choose who to talk to, quite often. After enough times, the relationship would reach a point when it can go further into details, and you start going down that girl's path. You could keep checking out the other girls' paths, but you probably wouldn't have enough time for everyone (this is assuming the game puts some time limit to avoid two-timers ;p).
++ Replayability... Mandatory?
It could have been the very first question. Is replayability even a good thing? Is it necessary? Does it bring something that good that it requires so much effort to go through?
It probably depends on your motivation and your resources. Big companies, having enough money, could hire more writers, more scripters and be able to add many branches and test them. If you're a lone creator, this could take many years and still not be that great.
What I realize is that many kinetic novels or games offering fake replayability are actually pretty good nonetheless! They might not be very interactive but they still tell a good story and even have a greater strength since their story can be very detailed and always work.
There's no definite answer to this. Replayability is fun. Wether you actually replay and see new things, furthering the experience... or you just play once but think "hey... I shaped -my- story and it was cool". But of course, replayability is hard and you'll often have to reduce the size of a single playthrough or have things be less detailed.
==== SECOND PART ====
Adding a few games, maybe the most important ones, before diving into the depth of replayability mechanics.
++++ Ever 17
While I haven't played this one, I think I understand the concept : you can play through once, but you just won't really be able to understand the whole mystery. Now here's an incentive to go back at it... especially since you can pick two very different point of views right from the beginning (not even a main branch?!). Each protagonist gives access to different girls, hence explanations... And then, once you've gone through all the possibilities, you unlock a final explanation which, I hope, leaves you with a great feeling ^^.
I'll be playing this during these Christmas holidays and I truely hope I'm not making it more than it is... but I think it's probably the game closest to what replayability should be... Replayability as a part of the *whole* game... not just as a gimmick to lengthen the game. Every playthrough adds something to the story and is actually needed for the full experience.
I think we're close to what Fate/Stay Night does, except you don't *have* to play through three sequential parts in a fixed order.
++++ Fate/Stay Night
Here again, I haven't played the whole thing but there are interesting things in there. The game plays more or less like a VN with fake choices, meaning lots of dead ends. BUT... Each dead end is actually well done! You sometimes have 5 or more minutes of reading before it ends with your death... and you can see why it happens. You don't really feel cheated by a one-liner death that has you crying because you forgot to save. I could actually see myself replaying such a game just to see what happens when you mess up. But wait, there's more! Everytime you die, you get a comical scene with the characters of the game, chibified, explaining why you messed up! It's not like we actually need advices, since all we need to do is take the other path, but, from what I gathered, they're pretty funny and are worth dying for
.It's not over yet, though... Once you finish your first game by avoiding all the death traps... You can restart again. Except this time, the story changes. Same protagonist, same overall story... but some elements are different. The story will focus more on a character which was only secondary before. I don't know if you actually go through some identical scenes or not, but even if you do, I know that at least a big part of the story is totally different. This is probably similar to Ever17 in that regards : you haven't truely finished the game until you've played through three times.
++++ Ever17 & Fate
It's really an interesting take on what the interactive medium can give. While movies could do that in a linear way, they would never be able to offer any actual choice. Here, the player controls where the story goes and always gets to see something interesting, even when it goes bad. Also, with the gimmick of restarting the game, the authors can show different point of views or wholly different takes on the same universe or theme.
=========== THIRD PART ============
++ Ways to implement replayability
+++ Story flow
Before really looking at how to make a playthrough different, let's look at the way the story *can* be different.
In any case, the difference between the kinetic novel and an interactive novel would be the fact it doesn't always follow the same path. Using Ren'Py, I can see two ways to do it :
++++ Branches
Following the main branch, you will give a choice to the player at some point. His choice will make him take one possible branch among many. Branches are neat in the sense that they allow you to draw charts rather easily. It helps to follow the story flow and detect inconsistencies. It doesn't mean it's not powerful since there are many ways to use branches. You can branch out, then branch back to the main storyline, and so on...
(this part needs more details about the many possible charts genres)
++++ Variables / Story Allocation
Variables allow you to make things a bit more unpredictable in the sense that you can react to things *later*. You just remember things that happen... choices that the player makes, and then, whenever you want, you can pull these memories out of your bag and react to them.
An example is letting the player choose wether he wants to help a poor girl, as a kid. Then, later in the story, as an adult, he meets that girl again. You "remember" wether or not he saved the girl and react accordingly ("oh, I remember you!").
Variables can make things very complex, but also a lot more lifelike and unpredictable for the player.
++ Replayability tools
How to make a replay interesting : possible solutions...
++++ Flavor text
This is probably the easiest to implement... but it can quickly become the most daunting task of them all.
The idea is to react to things that have happened earlier in the story. It works with variables. Basically, you remember that the protagonist learnt something about flowers... then you can have him say something charming to the girl about them.
Maybe the protagonist lived some particular event with one of the girls. Later on, he could make a reference to this event.
Flavor text never really chances the course of the story. It just proves to the player that we at least remember his actions and we react accordingly.
++++ Reacting to the past
This is kinda the same as with flavor text since we remember the past and react accordingly. The different is that what you did in the past will actually effect the course of the story. Did you help Mihoshi with her waitress job? Then later on, she will still be working there... otherwise she won't and might be in a pinch with money and all.
++++ Role system
An idea that would make things more "technical". Instead of basing scenes on a unique character, you actually assign them to a "role". For example, we could have a scene that needs the protagonist (probably mandatory in such games ^^) and a strong person. Each character would have a "strong" flag set or unset. If we happen to know Najika, the tomboy, she will be taking part in the scene. Otherwise, the next one might be one of your male friends.
The problem with this system is that scenes might lose their uniqueness because they would rarely be assigned to only one character. It would also force the author to make things more generic, less personal, etc...
The advantage is that you might not have to write as many scenes as usual since multiple characters could fit. I think this would work better with a game where you don't have to meet everyone on each replay.
++++ Reacting scene
One very classic tool, not a new one : have a scene after choices that shows an immediate result of your choices. Just thinking of a game that would follow a linear path until just before the end but would give choices here and then and provide one unique scene for each, you can get a game that can be replayed at least once, maybe twice. An example of this is Quartett by Littlewitch.
In Quartett particularly, the choices also made love meters go up, and when you reach the last part of the game, you would follow the (long) ending of the girl you had the most points with. Oh and, once you had all the endings, you would get a finale ending (played immediately after the last unlocked girl's ending).
++ Elements influencing replayability
++++ Skip function
While the skip function is of course something mandatory nowadays, especially for those with not enough time on their hands, it has a flaw : it destroys the mood on a replay. I'm basing this opinion on my Gloria games where skip isn't an option. Instead of skipping, I'd simply click-click-click or use the Enter key to go quickly through known scenes. But without the safety of the "skip seen messages only", one is forced to be careful and still read here and there. This helps keep track of the story until you reach an unknown part. With the regular skip function, you just hit Tab and don't care much what's going on, but when you reach the unknown part, you're suddenly a bit lost as to which context you're in.
This is nothing really big, but I think it makes you lose some value on the replays.
++++ Length of a playthrough
The longer a playthrough, the less likely a person will want to play the game again and the more time they will wait before actually doing it.
While long games are good because if you're enjoying them, it's for quite some time, it is still a problem with replay motivation. It's something to keep in mind when trying for a replayable game.
Making a short linear game will make people remember that single occurence. It's not bad at all. Making a short replayable game means you actually worked on a middle-length game. People who liked the game are most likely to return to it and squeeze a little more entertainment out of it without thinking they should really start doing something else.
Making a linear long game might turn away some casual gamers, but those who actually play will probably get a lot of enjoyment out of it. Given the length, there shouldn't be a need for more when reaching the end. On the other hand, there's no real incentive to "replay" since there's only one story. It doesn't mean one won't take it again later, as one would do with a good movie.
Finally, a long replayable game equals the work of a huge game and people reaching the end once will probably keep it aside for a while because they've just had a lot of it already. If they do start a new game immediately, chances are the new game won't last for long (well, if you're like me...). Longer games, replayable or not, are just better taken a while later, when memories get a bit fuzzy. Of course, when that time comes, the fact you can get a different experience out of the game is really a great thing.
In the end, it's a matter of how much work you can put into a game and your target audience.
