next »

[Hangedman]

So, first up, reading comprehension.

I was intentionally drawing attention to that

[DavidCaruso]

Assuming you were responding to my post:

If someone is having fun by quicksaving and reloading to get the optimal solution, why deny it to them?

My issue with these types of systems isn't that people abuse them. Even if I think they might be robbing themselves of something by doing so and would try to inform them if they told me about it (like I said to 1984), if they want to then they can. But not including a feature in the first place isn't "denying" anything to anybody.

Also I am not sure you understand grinding. Quicksave abuse is savescumming, which isn't really grinding so much as rigging the slots in your favor.

They both stem from the same design "principle" on a larger scale. In autosave-only games (or games with bad checkpoints), the developer decides to "balance" the save structure for "everybody" by making the players determine it, instead of making the decisions about pacing themselves. This is retarded, because how is the player supposed to know how to pace himself in your game to make it balanced and engaging, especially when he has the option always lingering to fuck up that pacing at any point if he feels too overwhelmed? Similarly, in games with grinding, the developer decides to "balance" the difficulty for "everybody" by putting in an unrestricted leveling system instead of implementing a proper skill-based difficulty progression themselves. This is also retarded for similar reasons (though I think grinding is a far worse design flaw than save-anywhere.) Like I said, if I ever fail in a grinding-heavy game I feel underleveled, not outstrategized. In both cases, the player is forced to determine a fundamental element of the game's system that the developer should have already taken into account but failed to, and in both cases it creates a sense of uncertainty about the game's balance which the experience would be more enjoyable without.

When you start saying that users shouldn't be allowed to save anywhere, you are outright dictating the pace of the user's experience as well as what session times will be. Users hate this.

Well, yeah. Is there something bad about forcing users to adhere to certain restrictions? Most, if not all, good games do this. And the session times would be like 10 minutes between checkpoints at most, and a lot of the time half of that. If you can't stand the idea of replaying through that little a few times then you're probably playing a bad game. Did you think the original Far Cry was a bad game because you weren't allowed to save everywhere? What about console shooters, like Bulletstorm? And why aren't people clamoring for save-anywhere in other genres, especially traditionally console-exclusive ones like 3D action games? It's not a developer's responsibility to keep providing a convention that can hurt their game just because many people have grown used to it.

Also, as mentioned, manual saving allows players to revisit sections of a game that they necessarily don't want to play all the way through the game for. For example, before each major branching speech interaction in Deus Ex: Human Revolution, I perform a manual save. That way I can see both sides of the coin without being forced to play through a game which has no relevance on that decision.

I agree that autosaves are harder to implement in a game like Deus Ex or an open-world title with a more freeform structure. I still know that I'd prefer them if they were at all possible to implement, though. And something like your example could easily be done with autosaves/checkpoints, too. Just keep track of the player's saves as usual and let him copy some to a separate place in case he wants to revisit certain parts. You could even put checkpoints before the branches if you really wanted.

Yeah exactly. Just think of excessive quicksaving as a cheat.

Hard Reset actually has quicksaves as cheat codes lol. That might be another solution.

[Painting]

Did you think the original Far Cry was a bad game because you weren't allowed to save everywhere? What about console shooters, like Bulletstorm?

Sometimes a good game has bad bits, man. Nobody is saying that a game is suddenly bad because it has a non-optimal save system, it just isn't as good as it could be. Are you arguing against games being the best that they can be?

And why aren't people clamoring for save-anywhere in other genres, especially traditionally console-exclusive ones like 3D action games?

I want a save-anywhere in everything, up to and including interactive fiction and train simulators. What could possibly be the downside?

I agree that autosaves are harder to implement in a game like Deus Ex or an open-world title with a more freeform structure. I still know that I'd prefer them if they were at all possible to implement, though. And something like your example could easily be done with autosaves/checkpoints, too. Just keep track of the player's saves as usual and let him copy some to a separate place in case he wants to revisit certain parts. You could even put checkpoints before the branches if you really wanted.

But why would you want to go out of your way to do that? There is literally not a single rational reason to not have anytime saving.

[hanako]

Instead of speaking in broad platitudes "When the game should lend the user more control. No matter what, the system should suit the game,", give me explicit examples of games that are improved by save anytime rather than a good checkpointing system.  

A game in which speedrunning was a desired activity?

But a sufficiently flexible definition of 'good checkpointing system' can always win out. A game that rewards frame-by-frame mastery might well save every step automatically anyway. A narrative/mystery game might have reasons why the player would want to be able to reload an old save just to play a scene again and refresh their memory for clues, but if you anticipate that and mark each scene with an autosave and don't overwrite them, the player would have no need to make eir own saves.

In a game with a branching narrative many players will want to make their own saves midway through the game that they can jump back to after reaching their first ending, so that they can explore possibilities without having to start all the way over from the beginning again. While sufficient stored checkpoints would still make this possible, players are likely to get more of a sense of control and choice from leaping back to their own bookmarks. (And if there are lots and lots of points marked out with generically-supplied name, it would be harder to work out which one to jump to tan if you had your own personalised saves.)

There are all sorts of reasons in different games why a player might want to go back and revisit a particular part; you can make this possible by keeping tons of autosaves but that's an awful lot of wasted space, isn't it?

[Dragonmaw]

I'll make a longer response later, David, but I'll just refute that Far Cry doesn't have manual saving. it does. On the PC, at least.

[DavidCaruso]

WTF? Could have sworn it didn't. Just Googled it to check and people have made mods for inserting it in and everything, and console workarounds. Maybe that was in a later release or something. Either that or I played it the best way possible without knowing it.

EDIT: oh, it was inserted in a later patch after the first release. okay, glad to see my memory isn't going defunct.

[Painting]

(except for the aforementioned pause/restart-save).

Why isn't this reason enough?

[mirosurabu]

Quicksaving ruins the pace of the game. Checkpoints set the pace of the game.

If you want noobs to play your game, make a noob mode.

If you want people to skip some parts of your game, make a skip button of some sort and give it a bad consequence/price.

If your game is not replayable (e.g. adventure game) and you have content that cannot be all seen in one playthrough (e.g. different endings/paths), make a map that allows players to jump to specific points instantly. Make this map available only when the player finishes the game.

If maps aren't an option, make your game replayable.

If you want to allow players to stop their game at any time and continue later, use an autosave.

If you don't like saving in GTA, don't ask for quicksaves, ask for autosaves. This way, safe house ceases to be a save point and becomes a mere respawn point.

Also, grinding is different to quicksaving because grinding has a major disadvantage which is tedium, whereas quicksaving easily becomes dominant strategy.

[Ichigo Jam]

I want a save-anywhere in everything, up to and including interactive fiction and train simulators. What could possibly be the downside?

Even in Tetris, or Pac-Man?

Also, I think most fans of Demon's Souls would disagree. The whole tension and atmosphere of the game would be spoilt by being able to reload when things went poorly.

I agree that nearly all games should have suspend/resume type saving - if you have a pause function it's basically the same as that anyway. It's just unfortunate that it's significantly more dev work if the rest of the game is based on save points or levels.

If you want noobs to play your game, make a noob mode.

Yes, maybe you could only allow players on 'Easy' to save at any point.

There are already some games with perma-death if you play on 'ironman' mode, which is a similar concept.

[Fallsburg]

If someone is having fun by quicksaving and reloading to get the optimal solution, why deny it to them?

Well, my argument is that it isn't fun.  Do you think it's fun? Do you know anyone who thinks its fun?
My argument is that people are going to play to win, which doesn't necessarily mean playing for fun.  As a game developer, I feel it is my duty to make winning fun.

Also I am not sure you understand grinding. Quicksave abuse is savescumming, which isn't really grinding so much as rigging the slots in your favor.

Well, before we start some definitional arguments, I'll lay out my definition for grinding and see if you disagree.
Grinding -- A repetitive act that allows the player to reduce risk in other parts of the game.
It's a broad definition, but I feel it fits.  

And as to why I think that quicksave abuse, savescumming, is grinding is that it fits my definition.  
e.g. In X-COM, I can take the time to save after every single step.  This will reduce the risk that my party gets wiped due to some fluke, but at the cost of wasting a lot of time and fun for me the player.  It isn't fun for me to do this, but it also isn't fun for my team to get blown up on the first turn.

I know that savescumming isn't grinding in the traditional "I'll kill 1 million boars to level up,"  but I feel that it is just a different in the mechanic being abused.  One abuses a bad leveling system and the other abuses a bad save system, but both have the same causes (poorly balanced game) and the same symptoms (player taking a repetitive action to make the game easier).  

When you start saying that users shouldn't be allowed to save anywhere, you are outright dictating the pace of the user's experience as well as what session times will be.  Users hate this.

I strongly disagree.  By creating the game, you are dictating the everything about the user's experience.  If I want to dictate their pace, that's my prerogative.  Now, whether I do that well or not falls on me, but throwing up my hands in the air and saying "Wow, it's hard to pace this properly.  Well, I'll just let the user do it." smacks of defeatism/laziness.  If the developer does their job correctly, the dictation of pace/session time shouldn't be a problem.

Also, as mentioned, manual saving allows players to revisit sections of a game that they necessarily don't want to play all the way through the game for. For example, before each major branching speech interaction in Deus Ex: Human Revolution, I perform a manual save. That way I can see both sides of the coin without being forced to play through a game which has no relevance on that decision.

This is probably the closest I would come to agreeing with a manual save solution, but the game knows when a big decision is about to made.  Wouldn't it be better for the game to seamlessly create a save (or have some sort of chaptering system) such that when you hit a point like this you wouldn't have to do it yourself?  What advantage is to be gained from forcing the user to do it?  I only see downside.

Why isn't this reason enough?

Maybe I wasn't clear.  I strongly agree that there should be the pause/restart save (or whatever the save should be called that allows a player to say, "Welp, I'm done for the night" and then can restart at that exact spot, but it is wiped when the player restarts).  I understand that no checkpointing system will be perfect and that it is supremely annoying to force a player to reach the next checkpoint to save (particularly for mobile games when play session length is not directly in your control).  
The reason I think that sort of system is good is that it isn't abusable (except for extreme cases of players subverting the intentions of the game by messing with the file system), it allows them useful capabilities, and it doesn't mess with the intended pacing of the game.

[Fallsburg]

Quicksaving ruins the pace of the game. Checkpoints set the pace of the game.

If you want noobs to play your game, make a noob mode.

Exactly. 
It comes down to what you want your game to be. 

Going back to my earlier example X-COM:
There are two games here, one with quicksaving, one without

WITH:
The game isn't hard.  The player doesn't have to worry about losing any squad members.  Their is little tension in the game.

WITHOUT:
The game is hard and cruel.  The player is constantly worried about losing squad members either due to their incompetence or the capricious hand of fate.

Personally, I think the game was intended to be played without.  The music, the atmosphere, everything in the game seems to be there with the goal of ratcheting up tension and scaring the player. But the game undermines itself by allowing the save anywhere.  If the game was intended to be an easy, low-tension game then it should be designed as such from the ground up.

Also, grinding is different to quicksaving because grinding has a major disadvantage which is tedium, whereas quicksaving easily becomes dominant strategy.

I don't follow.  You don't think quicksave abuse introduces tedium, and you don't think that grinding is easily the dominant strategy?  Because, I'm pretty sure I strongly disagree with that.

[mirosurabu]

You don't think quicksave abuse introduces tedium

Nah. I do agree that quicksaving takes more time than playing properly, but then, I think that, unlike grind, quicksaving cannot be or is too hard to be paced.

But then, most don't adjust the pace of the grind, so my point is moot.

you don't think that grinding is easily the dominant strategy

Provided that it takes a lot more time than normal play does, it is not.

[Fallsburg]

You don't think quicksave abuse introduces tedium

Nah. I do agree that quicksaving takes more time than playing properly, but then, I think that, unlike grind, quicksaving cannot be or is too hard to be paced.

But then, most don't adjust the pace of the grind, so my point is moot.

you don't think that grinding is easily the dominant strategy

Provided that it takes a lot more time than normal play does, it is not.

Well, I guess we disagree with about the definition of dominant then, because I don't see how length of time (unless we're talking about a game that has a timed component, e.g. speed chess) enters into the equation. 

Thought experiment time.
I make a game.  In that game you can grind and level up to level 100 and guarantee success, or you can play through regularly and have a 50% chance of success.  The optimal strategy is to level up to level 100 and guarantee success, no matter how long it takes to do that. 

Or
If there existed a way to play a flawless game of chess, the fact that it takes a long time to compute (assuming the game isn't time) doesn't make it not the optimal strategy.  It might be infeasible, or impossible for a human, or take more time to compute than the universe has existed, but it is still the dominant strategy.

[mirosurabu]

I see. I'm confusing dominance with unfairness. Actually, I cannot yet make a clear difference between the two.

Would you then agree that following options are fair?

1. takes less time, requires thought (skillful play)
2. takes a lot of time, doesn't require thought (grind)

[DavidCaruso]

In most grinding-heavy games it doesn't take that much time or effort to become overleveled for an obstacle, especially when compared to the time or effort it would require to stay at said obstacle and find a strategy to get past it. Likewise it doesn't take much avoiding enemies to become underleveled either. In addition to the mechanic being anti-meritocratic (i.e. unfair) it also leads to a situation where most of the time the player's avatar is either a war machine who can steamroll through everything without worrying about getting damaged or a weakling who turns fighting regular enemy fights into tedium, unless you specifically limit yourself and try to find the sweet spot inbetween (which isn't a fun thing to do.) There are exceptions, though, like the Souls series where despite the possibility for grinding being kind of disappointing anyway, it would still take a shitload of it to completely get rid of the need for proper thought and just steamroll through the game.

I think Fallsburg is getting it pretty well here. Giving the player a powerful weapon with infinite ammo that can be used at any point for the entire game and then expecting him to never use it is a pretty dumb design decision. Ideally you'd just get rid of the weapon, and then the player would have more fun and you'd have a better game. This is completely different in degree from making one simple decision at the start of the game on a menu, or opting out of a prompt when you lose three lives in a game that lasts half an hour.