Putting the 'role' back in role-playing games since 2002.
Donate to Codex
Good Old Games
  • Welcome to rpgcodex.net, a site dedicated to discussing computer based role-playing games in a free and open fashion. We're less strict than other forums, but please refer to the rules.

    "This message is awaiting moderator approval": All new users must pass through our moderation queue before they will be able to post normally. Until your account has "passed" your posts will only be visible to yourself (and moderators) until they are approved. Give us a week to get around to approving / deleting / ignoring your mundane opinion on crap before hassling us about it. Once you have passed the moderation period (think of it as a test), you will be able to post normally, just like all the other retards.

Lets discuss game save design theories.

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
We've had some thread derail into this over the years, but they usually start on some specific tangent and get dumped because both the original topic and the latest directing get beaten to death. So lets start with something broad here. This thread will be a place to discuss how games (especially RPGs) should handle saving and various pros/cons.

Lets start with some things we can all agree on (I hope.):

The system should allow you to leave at any time, in case you realize you're going to miss the season finale of My Little Pony and need to run off.
The save system should have backups in case of external system failure (Your computer, power, our the program itself dies).


I'm not going to put anything more than those two points up on that category for now. Fortunately, those can both be solved very easily with two things that have no other impact on the game: Temporary saves and temporary autosaves that are only accessible after an improper shutdown. The later are a bit easier to abuse, but I think it's safe to say if anyone feels like going outside the program to accomplish something within the program, no amount of design is going to stop them.

A 'temporary' save, is simply a save that is deleted upon being loaded. In pretty common usage at this point, and also the basis for any kind of hardcore game with a save system.





So, with some basics everyone can agree on out of the way, lets get into some examples we can bash eachother about.

One of the most common things people complain about involving saves is the save/reload spam abuse that break difficulty so easily. I've tried to think of some more elegant solutions (and seen other people do the same) before, but they always seem to encounter major hurdles in the form of requiring a heavy investment of design time to tie up all the loose ends. So I suggest an ugly and brutish solution instead. If the goal is to make the player not reload to accomplish a task simply because it's faster and less grief for him to try again than recovering from the failure, lets simply make reloading slower and cause more grief. Specifically: When you reload a save (that is, a save that has been continued from before, NOT a save that you made when taking a break from the game) you will be unable to do anything with a random variable for a set amount of time. No resting and hoping to not get ambushed by wolves, no lockpicking, no skill testing dialogues if they involve chance, no combat. You could accomplish this in one of three ways: Simply make the game pause for the required amount of time when you attempt to do such a thing while being penalized. Make such options unavailable while penalized (more work, slightly less immersion shattering and annoying overall). Or, finally, apply a negative modifer to all rolls so massive it outweighs even the benefits of rolling as many times as you could before it would have worn off anyways.

The result, in theory, should be that reloading after failure is agonizing. The only way to 'get around' the penalty is to not care about the penalty and do something else, either inside or outside the game. If your game is worth playing, that shouldn't be possible. There's no way in fucking hell I'd be willing to take a 10 minute break while playing say, XCom or Might and Magic 3 as opposed to sucking it up and taking the failure. Of course, this depends on the size of the failure, and games should include failures so complete you cannot continue from them. If the game is interesting enough, and balanced well enough that such failures are infrequent, it'll still be worth playing. Any hardcore game (roguelikes for example) is proof enough of that.

The primary drawback to this, is that it requires you to set the difficulty at a point where total failure is extremely rare. Unless your game is made of gold and shits ecstasy people are going to drop it rather than continue if a 5 minute fight kills them 5 times in a row, when they certainly wouldn't with normal reloading.

Optional parts that will/won't work:

This can't be tailored to the TYPE of failure. If you get a free reload because you died in combat, people will just get killed in combat after failing a lockpick roll.

Works well with soft death systems. If a total party kill doesn't end the game, but takes all your money and sends you back to town, then the player has the option of continuing, despite a loss of maybe 20 minutes of progress, or reloading and waiting 10 minutes before progressing without any of that loss. Some people would rather have the fastest route to new ground, while others would rather backtrack than twiddle their thumbs.

So, someone reply and tell me why my idea is shit and I'll do you the courtesy of returning the favour.
 
Self-Ejected

Kosmonaut

Lost in Space
Joined
Jul 11, 2008
Messages
4,741
Location
CCCP
I'm rather interested in the implementation details of the save system. How do you serialize an de-serialize every detail in your game world? I mean programatically. Anyways, about you saving design idea, anything that makes the saving and/or reloading state of your game a hurdle, will be met with resistance.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
Well, having your party get wiped out is generally met with resistance too. I'm hoping the resistance would be channeled into playing better / learning the game as opposed to impotent raging at the devs. Though I suppose giving them an empty 10 minutes to write shit on the forums works against that pretty hard, heh.
 

Tigranes

Arcane
Joined
Jan 8, 2009
Messages
10,350
It would need to be a pretty hefty time delay to make sucking-up a cost-effective decision for most players, I feel - if you care enough about a game or your character that you're bothered enough to regularly reload, you either larp/get really into the story, powergame/get into your character, or are anally retentive. Sometimes I make dialogue decisions in RPGs and get pissed off that it's not the decision I thought it was, and reload - I would probably do it even with a 10 minute time delay, because in my mind that sets the 'canon' straight. Hell, Gothic 3 upon release + my shit-ass computer = literally 5 minute loading times, but I still did when I attacked someone by mistake or something. If I was a player who was obsessed with maximising quest rewards or something, I'd imagine, again, I'd still reload, and just be more frustrated. But even 10 mins is pushing it - go any higher and it just becomes a hugely frustrating mechanic in its own right, even if it solves this problem. Hell, if you misclicked and reloaded then found out you have to wait 15/20 minutes, it's good grounds for a lot of players to just stop playing such a fucked up game.

The most banal answer remains the best - make options for failure interesting enough. In the case of story-heavy RPGs in particular, making failure have its own consequences means it's easier to integrate that into the story of how you're playing the game, whether at a larp level (my character failed at X and...) or player level (this stuff happened in my game because...). You can't do it for all, but you can do it for big things. Complementing this is that if you do it right, sometimes the player doesn't even realise they could have got a 'better' option. There are better examples, but I remember initially thinking Malik was meant to die in DX:HR because on hardest difficulty, it seemed like the odds were pretty stacked. In short, teach people how to enjoy games where you don't get the optimal outcome, instead of trying to force them.

In the end I think the technical infrastructure shouldn't interfere too much; it's a virtue to have a simple, quick save/load system where you can easily make backups or manage various saves. Then provide an ironman option a la TW2 or Dungeons of Dredmor.
 
Self-Ejected

Kosmonaut

Lost in Space
Joined
Jul 11, 2008
Messages
4,741
Location
CCCP
[snip] If the goal is to make the player not reload to accomplish a task simply because it's faster and less grief for him to try again than recovering from the failure, lets simply make reloading slower and cause more grief. Specifically: When you reload a save (that is, a save that has been continued from before, NOT a save that you made when taking a break from the game) you will be unable to do anything with a random variable for a set amount of time. No resting and hoping to not get ambushed by wolves, no lockpicking, no skill testing dialogues if they involve chance, no combat. You could accomplish this in one of three ways: Simply make the game pause for the required amount of time when you attempt to do such a thing while being penalized. Make such options unavailable while penalized (more work, slightly less immersion shattering and annoying overall). Or, finally, apply a negative modifer to all rolls so massive it outweighs even the benefits of rolling as many times as you could before it would have worn off anyways.
But how much time is how much? I mean how much time does it have to pass to consider a load just the continuation of a previous session, and not a reloading spree? This is very arbitrary and I think that every time chosen will be unreasonable for somebody.

Also, removing, crippling or affecting negatively something will not only break the immersion, but also the balance of the game. I'm not talking about difficulty, I'm talking about a steep change in how the game is played, in a negative way. And also, how would you test and balance something like that? Think that the saving penalty can occur randomly, so this will be very difficult to balance. And I fear that could introduce game breaking bugs or something.

The result, in theory, should be that reloading after failure is agonizing. The only way to 'get around' the penalty is to not care about the penalty and do something else, either inside or outside the game. If your game is worth playing, that shouldn't be possible. There's no way in fucking hell I'd be willing to take a 10 minute break while playing say, XCom or Might and Magic 3 as opposed to sucking it up and taking the failure. Of course, this depends on the size of the failure, and games should include failures so complete you cannot continue from them. If the game is interesting enough, and balanced well enough that such failures are infrequent, it'll still be worth playing. Any hardcore game (roguelikes for example) is proof enough of that.
No, you are being overoptimistic. And also, how do you know what can be considered a good save spamming deterrent, and what can be a fucking annoyance? I personally hate when games take control of my character in this way (specially in RPGs). You are not just doing reloading agonizing. You are just outright making your game unplayable.

[snip] Works well with soft death systems. If a total party kill doesn't end the game, but takes all your money and sends you back to town, then the player has the option of continuing, despite a loss of maybe 20 minutes of progress, or reloading and waiting 10 minutes before progressing without any of that loss. Some people would rather have the fastest route to new ground, while others would rather backtrack than twiddle their thumbs.

But there's a difference, with the soft dead you are just getting back in time, so to speak. Losing some XP points, money or some party members is a given, and I think is something more tolerable than some random penalization like the one you are proposing.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
I find it funny you mention those two as an example, as they obviously didn't add ironman at the same point in development. Adding ironman mode to a game balanced for reloading is nothing but a cheap afterthought only good for giving people extra challenge on a replay. You might as well add 1/4 leveling rate mode, or unable to deal with merchants mode. The entire reason there IS a need to implement a better system is the fact that playing a game with more or less reloading than it was balanced for breaks it and turns it to shit. I'd rather have everyone playing a game thats 20% shitty because of the gimmicky save cludges (whatever the best one may be) than have half the players playing the game at full potential and the other half playing it as 80% shit because you're not supposed to save as often as the last game you played.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
As an interesting thought; imagine the system works in the opposite way: whenever you've freshly reloaded a save, you are obscenely lucky for a while and win at everything. Critical hits, awesome loot drops, success at every skill.

That is, essentially, what reloading does anyways, without the additional agony of having to reload more than once. Is this effect really less destructive to the play experience and overall value of the game than pissing off the player?
 

Tigranes

Arcane
Joined
Jan 8, 2009
Messages
10,350
I'm not sure which 'those two' you mean, but if you mean DOD & TW2, for the latter - yes, obviously Ironman is pretty stupid for something where you have to wade through five thousand cutscenes, but the former is a roguelike lite, so uhm. Sure, just chucking on Ironman isn't great, but it's an easy option that does make a lot of things fun - hence all the Ironman house rules for IE games, etc. Permadeath is also a rare enough preference that it's not viable to devote too much on it.

Back to save/load in general, though, nobody disputes (here, anyway) that save scumming sucks the fun out. The point is that your proposed solution, and in fact, most solutions raised that I've seen, actually do more harm than the problem they are meant to fix. Why would I choose a game where I have to wander around thinking "Man I can't do shit for the next 10 minutes"? Because no matter how you implement something like that, it will never be precise enough to target only save-scumming, it will always annoy you when you load innocuously.

I forgot about this before, but one partial solution I found very effective was Alpha Protocol - the game autosaves immediately after key conversations/decisions (something DXHR copied in places). I remember being tempted to reload then finding it wasn't possible. In later playthroughs, where I was trying out different dialogue options, it really was quite hard and annoying to break the system and reload. Of course, this also required the checkpoint system, which again had the same problem of causing inconveniences and detracting from the fun in itself for many people.
 
Self-Ejected

Excidium

P. banal
Joined
Aug 14, 2009
Messages
13,696
Location
Third World
Back to save/load in general, though, nobody disputes (here, anyway) that save scumming sucks the fun out. The point is that your proposed solution, and in fact, most solutions raised that I've seen, actually do more harm than the problem they are meant to fix. Why would I choose a game where I have to wander around thinking "Man I can't do shit for the next 10 minutes"? Because no matter how you implement something like that, it will never be precise enough to target only save-scumming, it will always annoy you when you load innocuously.
And that's why I'm against any sort of save restriction. If a player wants to be cheap and abuse save scumming, let him. It's only his own fun he's ruining.
 

Skittles

He ruins the fun.
Joined
Apr 20, 2011
Messages
983
Why introduce this time delay mechanic to deal with save scumming anyway? Can't you just save the seed so results of random actions are the same upon reload? I feel like JA2 did this pretty easily.
 

Majestic47

Learned
Joined
Nov 9, 2011
Messages
432
I just brofisted Excidium - actually this isn't an issue.

Go ahead, save scum. If that win is so important, so be it. The developers can't be faulted if players wanted to do it this way. People say, 'save scumming removes the tension.' well, put iron man mode in.
Get tense all you want. Perhaps reward the iron man players with additional gold / loot / exp for the risk he took and achievements.

It's those little things that made people consider taking additional challenge is worth it.
 

janjetina

Arcane
Patron
Joined
Mar 28, 2008
Messages
14,231
Location
Zagreb, Croatia
Torment: Tides of Numenera
Needless complicating and making game a chore to play will get you nowhere. You can prevent save scumming by keeping a random seed* (or a table of random seeds if employing more than one random number generator) along with the rest of the game data and saving the seeds on exit. Seeds are not generated at loading, but at certain defined points of the game.

A rational approach would be to make this optional. Let those who want to save scum save scum, let those who want challenge have their challenge.

For those who want the ultimate, hair tearing challenge, there is an optional ironman mode (with tempsaves and saves on exit only).

Gradation and customization. my brothers!

*actually not a seed, but a random number generator state, but seed is shorter to write down
 

Alex

Arcane
Joined
Jun 14, 2007
Messages
8,750
Location
São Paulo - Brasil
Well, the thing about the iron-man mode is that it works very badly for some games, while it works very well for others. Take Crawl or Adom for example. Two roguelikes specifically made with iron-man gameplay in mind. In those games, dieing a lot and restarting from the beginning works very well. There are a lot of different gameplay options, so if you gt tired of playing a character type, you can play another for a while. The starting locations always change, so it doesn't feel like you are going through the same thing every time you restart your game, and so on.

And that is the thing I think is worth discussing, how to make your game work well with various different save structures and what kind of save structures are good for your game. For example, Damned Registration's suggestions probably have in mind a game full of interesting failure results, but without the random level generation of roguelikes. In that kind of game, you want to clearly communicate to your players that they don't want to save scum. Of course, they may do it anyway. In the end, it probably comes down to find the files and keep backups, and if someone wants to do that, it is their loss. But if you make it clear in your game that doing so isn't a good idea, then more people will play the game the way you want. Well, at least, it worked for me and roguelikes.

Anyway, about the idea itself, what I don't like about it is that it tries to annoy the player into following the rules. I think all systems that rely on the player's patience. I really think that game mechanics should never use the player's time as a game resource directly. The player's patience to wait for something shouldn't be something that is tested in game. I would instead suggest the following system: have a few "checkpoints" in your game, places that mark some important change in it. Whenever you reach a checkpoint, you game is saved, and you can restart from there, exactly as you were when you got there. You can also save your game outside the checkpoints, but it works like a roguelike save, without the ability to restore it twice. This would work well, I think, if each checkpoint area had various different options you could explore, probably not as many as a roguelike, but still with a lot of replayability. I think that doing so clearly communicates: "Look, this area is full of cool stuff to explore. You should probably replay it some times, maybe arrive here with a different character to test other options".
 

Phelot

Arcane
Joined
Mar 28, 2009
Messages
17,908
In my opinion, Excidium is correct. I don't really like any restrictions on saving mostly because at times I need to be able to get off the computer and do something else at a moments notice. So, even some of X-Com's longer battles would be a pain trying to find one last idiot alien so I can save and get going. And I also believe that it isn't a game dev's responsibility to account for every little exploit of otherwise mundane features.

That said, I will also admit to save scumming at times, but only in certain games and here's where I think DamnedRegistration might be on to something. While I don't like the idea of a forced penalty, I do notice that a few things prevent me from being tempted to save/reload:

In ToEE Co8 mod, I never save scum other than if a party member dies in a battle they probably shouldn't have died in which isn't really scumming anyway. I don't reload often for a few reasons. The first would be that saving and reloading takes a bit of time in ToEE. It's not instantaneous like, say Jagged Alliance 2. Also, consequences tend not to be as severe in ToEE. Taking a hit is certainly not the end of the world and even being knocked unconscious. I'd rather play out a turn full of misses, frustrating as it may be, then reload. That isn't to say the game isn't challenging at times, just that overcoming a challenge typically means my party needs a bit of rest and a few spells to be ready for more.

In JA2, that isn't the case, so save scumming becomes a problem, especially with things like the 1.13 Drassen Counter Attack. Receiving a single hit on a merc, even if not fatal, negates that mercs usefulness instantly. But maybe that mod is a bad example since there's a lot of cheesy difficult situations (6 mercs vs 80 enemies, yeah right...) But as I mentioned, the reason save scumming is so tempting in JA2 is because it takes less then a second to load a saved game. None of this stops the player from saving a game that will develop into a doomed battle regardless of save scumming.

I guess if you really wanted to prevent save scumming would be to cause some kind of lock down on random rolls after a few reloads. I've noticed in a few games, JA2 vanilla included, that after maybe the third reload or so, you always get the same result when trying to target an enemy. I may be crazy, but I've seen situations where one reload, I hit for little damage, next reload misses to the right, next to the left, then every reload after that has me missing at the same spot. I could swear I've seen this happen in other games, but I can't really confirm it. In any case, if there was an easy way for devs to allow for maybe a limited number of random seeds each time you attack or whatever, maybe that would encourage people not to bother save/reloading.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
Checkpoint system could work, but I think more because your number of hard saves is limited. Distancing the save point from things a person would want to scum certainly works as a deterrent, but it works in the same way as my idea, with a layer of something to twiddle your thumbs with while you wait. But it's limited because unless you make your checkpoints in the ass end of nowhere, there'll be stuff worth reloading for nearby. If I got a new hard save at every new town I entered for example, it'd be a pretty strong encouragement to use the opportunity to savescum at least a few times while trying to pickpocket/lockpick in that town.

Saving the RNG seed doesn't really work, because the player can just change the order of his actions to get a different state out of it. So instead of Lockpick-Fail-Reload-Lockpick-Win you get Lockpick-Fail-Reload-Lockpick another thing-Lockpick target-Win. Or get into combat with an enemy, or whatever. Which again, is just a minor time penalty on the player. You'd need a separate seed set for every RNG target in the game as soon as possible (preferrable when a new game is started), which seems like a lot of extra shit to do, both programming wise and system resources wise. It also does absolutely nothing to resolve the problems of things like walking into an ambush, reloading to 5 seconds ago, and fireballing a half dozen snipers from off screen for free xp and loot.

Go back to my example with a reload system that gave you cheaty uber success for a while after reloading. Would you honestly play the game properly with that there? I think most people would assume they were meant to abuse it to succeed at nearly anything. Especially if success was difficult to begin with. If this system isn't better than a normal system (I hope everyone agrees that it isn't) why is the default system better than mine? They're just moving along the exact same gradient as far as I can see. You're trading player annoyance for enforcing intended game balance. It seems silly to assume that the amount of annoyance you get from reloading normally in the first and most obvious system anyone ever designed is the perfect sweet spot for creating perfect gameplay.

The only time people seem to honestly respect the intended difficulty of a game is when you blatantly spell it out for them, like wizard mode in a roguelike telling you you won't get a score in all caps. People seem to think it's ok to reload when half your party gets killed by a trap, because that's really hard to recover from. Even though a different party would have survived the trap easily, and had a lot more trouble with combat than your party of 6 wizards. But for some reason they don't think they should deal with the cons of their party because they're easily mitigated by reloading, while the cons of a party of 6 warriors (not enough equipment to go around) is something you're supposed to deal with, because there's no way to cheat it.

Edit: People seem to be misinterpreting part of the idea. I'm going to go all caps rage monkey on this in the hopes it gets across. YOU ARE NEVER PENALIZED FOR TAKING A COFFEE BREAK. RELOADING FROM THE LAST SAVE YOU MADE WHEN YOU STOPPED PLAYING WORKS NORMALLY. YOU ARE ONLY PENALIZED FOR RELOADING A SAVE YOU ALREADY PLAYED AHEAD OF.
 

Tigranes

Arcane
Joined
Jan 8, 2009
Messages
10,350
I actually disagree with the whole "it's their choice, whatever" argument. It's sensible to a degree, but hey - humans are learning creatures, and that applies to 'leisure' activities just as much as anything else. We are trained to enjoy what we enjoy in the way we enjoy it. Save scumming itself is a result of games in past eras training us to reload. Ever had a case where you 'lost' a battle as part of a scripted event (happens a lot in JRPGs), and you reload because you think it was just you? We've been trained into this feeling that (1) we should be able to overcome every challenge within a game, and (2) we can reload and make it all better. So, if possible, should games and especially RPG devs look for ways to retrain gamers to stop wanting to save-scum so much? Absolutely. That's part of designing a game to make it 'fun', in the long run. I'm with the OP there.

I brought up AP's checkpoint mechanic, and I said it did well to stop save-scumming, but this is going to be limited to games with, well, fairly linear level progression. AP was extremely nonlinear in its plot branching but very often linear in the actual level design - so it was easy to put in checkpoints. You never really had reason to go backwards (so much so that the game actually destroyed the world behind your back to save resources). Putting aside whether AP was good or not, it gets at Alex's point that different games just can't handle different solutions like checkpoints or Ironman.

In strategy or simulation games, the random roll lock is a good idea - another thing is that random 'events' can be determined some time before they are presented to the player. I've only seen this in a few, weird situations (like Football Manager), but imagine, say, in Medieval: Total War, the turn where the Mongol horde appears was actually 'rolled' 20 turns before. Reloading to your last autosave to prevent them appearing would not work, because the game already decided this, and you're not sure when that happened.

Finally, re. the time delay idea, again, DR, you are never going to get it precise enough to target only save scumming. E.g. what if you were playing a bit, then alt-tabbed because of a visitor/your boss/bandit attack, then reloaded because in the meantime you missed something? What if your kid/cat/pet elephant fucked with the keyboard? What if you got a weird bug while playing on? What if players made the wrong decision because of a misclick or because the UI did not give them all the information? There are so many reasons that you are almost certainly going to have a lot of pissed players. Hell, it's like DRM!
 

Skittles

He ruins the fun.
Joined
Apr 20, 2011
Messages
983
In JA2, that isn't the case, so save scumming becomes a problem, especially with things like the 1.13 Drassen Counter Attack. Receiving a single hit on a merc, even if not fatal, negates that mercs usefulness instantly. But maybe that mod is a bad example since there's a lot of cheesy difficult situations (6 mercs vs 80 enemies, yeah right...) But as I mentioned, the reason save scumming is so tempting in JA2 is because it takes less then a second to load a saved game. None of this stops the player from saving a game that will develop into a doomed battle regardless of save scumming.

I'm pretty sure JA2 uses the seed saving technique. I just tested it--you can load and take the same shot dozens of times and there seems to be almost no variation whether it hits or not. Damage seems to fluctuate a little, which is interesting, but hits don't. You can vary the number of AP you commit to taking the shot, but it seems like that only matters if you barely hit or missed in the first place. Same goes for shots taken at you, by the way.

What DamnedRegistrations says about varying actions is valid, though is difficult for a player to apply in combat like JA2. First, a shitty roll seems to apply even to shots taken at different targets. Second, you could opt to run for cover or something rather than taking the shot, but you don't gain the benefit of success in the action you're trying to perform, really.

Of course, it seems I'm talking about in combat whereas he's talking about out of combat situations. But it seems like if designers are intent on stopping save scumming in cases like that, there are thresholds, there's seed saving, there's creating mechanical or narrative reasons to continue playing through failure, like Tigranes seems to be suggesting, and so forth, without artificially punishing players by using the save system against them.
 

BirdsCanFly

Educated
Joined
Dec 8, 2009
Messages
337
Just take Total War Shogun 2, you COULD reload, but the loading times are so abysmally high that I think twice before reloading a save.


On a more serious note, I do think it makes for a way more interesting game if you don't reload. I think the main problem is that in most games, when you reload, you died. It would be way better if the punishment was not death, but something else. For instance if you lost a battle in an RPG, it should not instantly kill you, but if you got defeated by bandits they just rob you and you have the chance to get your stuff back, adding to the narrative and so making for a more immersive gameworld. This was actually said in Sengoku Rance, where the developers said they encourage the users not to play a "perfect" game, by reloading after every "mistake", but going along with it, which I think is quite admirable, if people would actually do it.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
I think a big part of the problem is that back when games were first being made, there was some sort of scoring system implemented that made even absolute failure (starting over) not feel so absolute. Even if I die in the pits of hell playing a roguelike, part of the reason it doesn't feel like a total loss is that scoreboard where the character is immortalized. At least I have a record that he got THAT far.

Once games moved away from the arcade roots though, the only thing being immortalized is your current save, which makes accepting any sort of failure feel really bad, because you didn't just miss out on your current score, you kneecapped your potential score. The game is just plain less fun to play if you got your rare purple dildo artifact stolen and won't be able to have the complete collection at the end of the game. And unlike games with a scoreboard, starting over just to beat your current game doesn't feel like it's worthwhile. If I replay a crpg a second time, it's not because I want to do better than last time, it's because I just want to play.

I can't help but wonder if player behaviour couldn't be sufficiently influenced just by giving them a scoreboard for their finished games. Just having the note on your finished game file listing your 541 reloads might be enough to shame people into not doing it anymore, at least for a lot of the small stuff.
 

Phelot

Arcane
Joined
Mar 28, 2009
Messages
17,908
I suppose you could award the player for not saving so much as long as it's not a retarded achievement. I don't know, maybe you get some special item that might help you or get you into a special area if you saved under 25 times (or whatever is appropriate) for one PC. The save limit would be generous, but not if you expect to do some serious save scumming.

But, all these little tricks and stuff can still be exploited. No matter what you do, the player will find a way. They always do, those damn bastards!
 

Skittles

He ruins the fun.
Joined
Apr 20, 2011
Messages
983
I think the least obtrusive solution if you really want to mess with peoples' ability to save as they like is to simply have only a 'save and quit' option. Tigranes' criticism that it'd be obnoxious for people who make fuck ups because of acts of God, etc. stands, and I'm pretty sure a dedicated save scummer could get around that, but it'd address your concerns about save-scumming without turning how you save and load into part of the game. That'd just be awful.
 

Duckard

Augur
Joined
Aug 14, 2010
Messages
354
All in all, I prefer just using the save anytime reload any time system for the sake of convenience. Punishing player for reloading just becomes annoying, and the player that really wants to savescum will do so. Maybe the game could display how many times you reloaded in your stats as a sort of mark of shame, like ToEE did with stat rolls. At the end of the day, you can add little things here and there to gently discourage savescumming and encourage accepting the failure without being too intrusive, but some folks just wont care.

It could be nice to reward the player something if they fail a lot as a sort of bonus for "taking it like a man". Something like the fate points in Arcanum could work for that. So if you jam, say, 10 locks, you get a "fate point" type thing that can unlock any lock, even if it would be normally above your skill level. The problem with that is it becomes a bit too game-y and easily exploitable. Better yet is to just leave it alone and just make sure that failures are not such a big deal. You fuck up and you suffer a bit, but you also learn something from it and accept it. When you fuck up and suffer a lot (eg: party member dies), you're probably going to reload. Of course, then there's the issue that there's not enough punishment for failure and stuff.
 

Alex

Arcane
Joined
Jun 14, 2007
Messages
8,750
Location
São Paulo - Brasil
I don't think it is really important to make sure the player doesn't abuse the system, but to make clear that such abuse is counterproductive, and to make it so that players that play in "the proper way" can actually play the game. I have frequently seem games where failing a skill roll could be interesting sometimes, but most of the time, you would be hit with a penalty so stiff you would have no reason to sick around.

For example, take Daggerfall. Daggerfall has a "learn by doing" system where the more you use certain abilities, the better you get at them. However, because of the way it is implemented, the most obvious action is to grind those skills safely, away from the dungeons. I ran and jumped everywhere I went in that game, and to train magical skills, I made very cheap spells that I would cast again and again while on a rented room, where I could sleep safely. You could play the game so you only use your skills while in a difficult situation, where they really matter, but the game doesn't, in any way, make this the most obvious course of action. Even, worse, doing so could make the game much slower and hard.

On the other hand, take Burning Wheel, a pencil and paper RPG. It too has a learn by doing system. However, because of the way skill tests are set up, you never have an empty skill test, a test where nothing is at stake. The game enforces it so that you train skills by using them where they matter. You don't get a "skill check" (think of it as skill experience) unless the use of the skill has something important riding on it. Of course, any computer game built using this system could still be exploited. A player might go after skill tests on stuff he doesn't care about, so he has a very good skill once he starts to tackle the stuff he really cares. But doing so in counter-intuitive. If the player tries to ruin his own fun doing this, then sure, let him go ahead. The important thing is that it is clear what is the right way to play the game and the right way is actually fun to play.
 

shihonage

Subscribe to my OnlyFans
Patron
Joined
Jan 10, 2008
Messages
7,157
Location
location, location
Bubbles In Memoria
I pre-roll the outcome of player's attempts to lockpick every object in the world. The outcome may still be changed if the char's skill rises above a certain threshold as he progresses. This system may be altered later, it's WIP.

In hindsight, a possibly better approach would be to retain Fallouty lockpicking system, but spawn an enemy every time you fail, "alerting the guards".

Also, I forbid saving in combat. If you want to keep trying that critical shot, you better start the fight from the beginning.
 

Damned Registrations

Furry Weeaboo Nazi Nihilist
Joined
Feb 24, 2007
Messages
14,983
If the outcome is pre rolled, you might as well not have a roll at all. Otherwise all you've done essentially, is assign a bunch of difficulty levels on locks, then randomized them all for lulz. If it rolls high on this lock and low on that lock it's effectively made a little girl's diary more difficulty to break into than a bank vault, which is clearly not what you intended.

Failure to pick a lock never really made any sense to me to begin with. It's not like a lock is some whirring contraption of doom. It's an innanimate object. The only way you'll break a pick is if you press hard enough to break it, which anyone experienced enough to pick a lock in a reasonable amount of time obviously wouldn't do. Aside from that, it's pretty much a matter of time and skill relative to the quality of the lock. But if you can pick one locked door in a building, you can probably pick all the others too.
 

As an Amazon Associate, rpgcodex.net earns from qualifying purchases.
Back
Top Bottom