In games that are non-linear, many gamers save before critical decision-making times, in order to see what effect a given decision will have on the story to follow, then go back and check the other options to see what the "best" choice would be. I know this, since I find myself doing the same thing in some games. This also applies to treasures in dungeons, for example, if you take a treasure and unknowingly block the path to three treasures that are even better, simply load up a save game and get the better ones: no harm done. What I am wondering is if there were a system in place to remove this option of essentially jumping back in time to make better decisions. We don't get that luxury in real life, so to keep the player immersed in the game's world, wouldn't it be better if it were not possible there either? You would play the game making decisions just as you would if you were in the actual situation, and the story customizes to your TRUE decisions, not your IDEAL ones. I also want to raise the question of how this could be done. As far as I see it there are two options, and it ties very intricately to how data is recorded in the game. Either, the game is saved automatically with each action taken, in which case the only real way to roll back to a previous "save" is to copy the save file, and replace it when you want to go back. The other method fits into a system where the game is saved at various save points scattered throughout the game world, similar to many RPGs. In this case, you would have to place important decisions far enough away from save points in order to make the player not want to go back and load a previous save. For example, the player may save at the beginning of a dungeon, do the entire dungeon with puzzles and enemies (possibly saving after some of the more difficult puzzles to remove frustration in the event the player dies), some more enemies, a boss and THEN place a story-shaping decision, so if the player really wants to choose something different, they have a lot of stuff to redo. I'd like to hear some comments and opinions on this type of system. What do you think?

An important part of not being able to roll back descision is to make the impact of the descisions clear to the player. Many times I've seen games where the player is given a "choice", but only the developer (or someone who has played it before) knows the effect of. The simplest ones are the two treasure chest choices, where you don't find out what is inside until after you open it (making the choice pointless unless you intentionally factor in load/save). The more complex ones are where players are asked to make some storyline type of choice, but not given enough information about the story to understand what the consequences are, for example (WARNING! DOOM 3 MINOR SPOILERS AHEAD) when you reach the communications dish in Doom 3, you are given the choice to send the message (which you know is supposed to bring help) or not send the message. However there is no information telling you why you shouldn't send the message (you only find out right after you make the choice that ships will be sent, that could be captured and used to transport evil back to earth). So until you make the choice you don't know what the point of the choice is. Even then it's not really clear how the game is affected.

There is nothing wrong with the 'Save and Exit' strategy if the game is well designed. I think that it would be a flaw in the design to offer the player a choice, which they have no way of knowing the outcome, without letting them go back and make a better decision.

I think Fallout did this well. You could save the game anywhere, and jump back to make better decisions. Sometimes, it would only be a few minutes of game play lost for a better decision, but others could cost you many days of playing (if you still have the saved game).

Another option for the save games is to allow only a single saved game to exist at any time. This way, the player may have to decide between jumping back to before they open a treasure chest, or starting a potentially dangerous line of dialog.

I think such actions are fine by the player as long as he/she is having fun. The primary goal of a game is to be fun, not immersive (which is really just a factor in how fun a game is).

I think the best way to implement this is to just have the decisions' effects occur so far in the future, that the player simply won't want to go back and remake the decision when he understands the ramifications of his decision.

Quote:

Original post by Alex What I am wondering is if there were a system in place to remove this option of essentially jumping back in time to make better decisions. We don't get that luxury in real life.....

Which is exactly why people play games. They want to have fun and explore without suffering the consequences as they would in real life.

People buy games because they want to play them. Trying to force them to only play the game the way you want is a mistake. Many people don't "roll back" but those who do, do so because that is how they like to play. They don't have time to play the whole game through fully in every possible way so they explore and reload. If you stop them playing the way they want to they are less likely to enjoy your game. If they don't enjoy it they wont play your next one (or even the current one).

Quote:

Original post by Obscure Which is exactly why people play games. They want to have fun and explore without suffering the consequences as they would in real life.

Good point...

I guess for the most major decisions in the game, it would be best to let the player know what will happen as a result of each of the choices, like Michalson suggested. You get this kind of info from characters in game discussing the situation, and saying what would likely happen if they took a certain course of action. At least if they know before-hand, less immersion would be lost as a result of saving and loading because you dont know what decision you really want to make.

I also like Numsgil's idea of having the effect of a decision play out in the distant future. Though, I'm not sure if players would find it exciting or frustrating to have something happen because of something they did so long ago. I guess it depends on the how drastic the effect is, obviously they won't be too happy if someone important in the story dies or something like that, but they would be happy if given the opportunity to do special side-quests or something.

Thanks for the feedback everyone!

An extreme example: I come across 2 chests, opening either one causes the other to disappear and one contains the key item for the current dungeon while the other triggers a tough boss fight, following which I would have to trawl through an additional 50 rooms of dungeon to retrieve the item. If I have no warning about all this until after I open the chest (and possibly check an online guide after going through 30 more near identical rooms trying to find the item) then I'm either going to reload a previous save, walk away from the game entirely, or play through the rest of the game with a guide beside me. If you remove one of these options, then I'm more likely to give up on the game entirely, and be very vocal in my criticism of it to everyone I know who even might consider buying it. If, on the other hand I come across those same two chests at the end of a dungeon, know in advance that one of the two contains the reward, and the other punishment, and have some sort of hint available to tell me which to open, even if I miss it or fail to figure it out, I'm more likely to accept the outcome I get stuck with, and probably be moderately vocal towards my friends about how the game gives you enough information to make the right decision... I may also deliberately choose the punishment outcome for the sheer challenge of it.

I guess the point I'm trying to make is that the way to prevent players from save/reloading through "choices" is to make them genuine informed choices rather than random guesses. And if you want to stop save/loading, then track the number of times the player goes back and reloads a save that wasn't a "save on exit" and make sure it's included in the player's statistics.

Hence my suggesting you count the number of loads of non-exit saves and show it as a statistic rather than forcing exit-saves or constant rolling save only.

As a slight aside, I've long since come to the conclusion that "interrupt" or "exit" saves are always legitimate (the player is always able to walk away from the game, and it's your job to make sure he can come back) - after all, not everyone can put together the 6 hours or so required to complete the last level of GTA... "perfection" saves (I've just got through an encounter, but took too much damage or used too many resources, so I'm going to reload) are not legitimate, but can be necessary due to poor game design or low player skill. Finally, "branching" saves (there's a choice here, and I want to know what the story looks like down both branches without having to replay the entire game to get here - eg the three "victory" endings in Deus Ex), while not required in a game, are a reasonable thing for some players to want to do, but open to abuse by powermaxers - so counting the number of non-interrupt loads gives powermaxers another statistic to optimise, while those in it largely to explore the entire story tree can ignore the number and pursue their own goals...