I only read part of page 1 (forgive me), but let me pipe in here.

I''ve been playing alot of NES, SNES and Genesis games on emulators recently - games that I played on the consoles when I was younger. Obviously since I took the time to play them again, it was because I enjoyed the original versions and wanted to rekindle that feeling.

As many know, emulators have "save state" features, which is one step above savegames. It physically stores the contents of the emulated RAM and CPU state into a file on your computer and you can load it anytime. You can save absolutely anywhere you want, whenever you want. Obviously this goes against the way most games were meant to be played. I found myself using this feature frequently.

The games weren''t the same when using this feature. Most were pathetically easy. It takes away most of the challenge, the risk, the thrill. I''ll say emulators are worse off for having this feature, because it hi-jacks the game design that went into the console games that they emulate. Then again, this feature came about when emulators were just coming out and some wanted to set their emulator apart from the rest. Can''t blame em really for innovating, but it hurts the gameplay.

Sure, you can say "just don''t use the feature." But, that''s easier said than done. I tried to play the games "legit" but all that happened was when a setback occured and I had to re-do stuff, I would say to myself "this wouldn''t have happened if I had saved state," so I ended up succumbing and using savestates. The temptation was there and I took it for almost all the games I played (I went on a downloading binge )

I would have to agree with the original poster, too much opportunity to save can ruin a game, because it takes away from the challenge.


p.s. does this post remind anyone else of their mother saying "dinner''s ready" then you saying "i just need to save, it''ll take 2 minutes" ?

The save/load game feature is that mechanism by which a player learns the language of the game, providing the opportunity to compare the success of various syntagmatic (sentence-like) structures within the context of the game's paradigm (wordset-like). Iterating the load/save teaches the player to think using the game's language, essentializing their goals on optimal game states.

Tetris, mentioned earlier, is a perfect example of a game that inherently revolves around this feedback loop.

To suggest that save/load is bad game design is to suggest that, for a game which you designed, the need to know how to communicate using the game's language would be irrelevant. True, albeit limited. The question "why does God allow evil in the world?" represents a similar train of thought.

PS. Would you prevent people from editing posts too?

[edited by - deClavier on May 21, 2002 4:27:10 AM]

A suggested alternative to the strictly linear save/load feedback loop is to randomize the occurence/decision procedure of enemies within a game after each load. This maintains suspense, whilst allowing the learning to continue.

Would it be possible to randomize the starting location/stats of the pc after each load to the same effect?

Remind me not to post friday at 5PM when I won't be able to revisit the thread for at least another 48hrs.


The argument is all about failure, as Wavinator pointed out. Too many times, failure is really easy, REALLY final, and really illogical. "Lives" in Mario are no more or less than minor saves, to allow for a few mess-ups. In role-playing games of long duration, it's painful, because using a similar system would leave you having to restart after several days of gameplay only because you happen to have a bad roll of the dice. As I suggested earlier, it depends a lot on what the intention of the game is. If you are not meant to repeat parts of the game many times (read: if it's no fun), it's going to be really annoying if you have to do so anyway.

Perhaps it has to do with challenge spikes. Many CRPGS (keep coming back to those, because they seem to be the most invulnerable to the no-save-game-anywhere strategy) consist of a bunch of talking-in-the-town-with-no-risk-at-all followed by a stream of encounters that are highly deadly. Having to repeat that talking the same way over, and over, and over, just to hear a bunch of things you already knew, is absolutely worthless as gameplay. You don't want to subject your players to that kind of abuse, they are going to be using the game CDs as frisbees, aiming for YOUR head. There's a massive challenge spike that is completely unproportional to the exploration part of the game play.
Really, many CRPGs consist of two enmeshed pieces of software: an exploration engine, fueled by a back story and a huge lively world, and a bunch of combat scenes to provide "challenge". Many players are into the game for only one of the two things: the exploration (aren't you, Kylotan and Wavinator ), and the fact that combat turns back your exploration clock if you lose is a royal pain in the part you sit on.
Or, paraphrased from how Kylotan put it eloquently:
Combat is a stick, not a carrot.(in CRPGs, for the exploration player specifically). It's only made worse by how combat rarely affects the exploration part of the game at all. All that the exploring player is doing is trying to get past the combat to the next part of the story, it is not actually part of the story.

Hrm, how would CRPGs improve if we only had combat when it was necessary for the story? And how would they be if you couldn't "lose", but rather you could really win or not win very much? What am I babbling about? Well, take the following situation: you're at an inn, and there's a group of ruffians hell-bent on getting into a fight with you. You can get into this fight, and there's hitpoints and wounds and the whole common RPG shebang. However, should you not kick royal behind, then the town guards will bail you out right before the lights go out. The effect is that your reputation diminishes somewhat, but you don't have to replay talking to every gardner in town to find out where you might find a white flower of gandilea to trade for the sword of mighty shamrocks at the wizard's place.

What have I done above? I've tried to take into account several key points of the game design.
1. It's a role-playing game with a reasonably linear storyline, that I do not wish the player to repeat constantly until they finally finish it.
2. I have removed "final failure", because it only violates the principle outlined in 1, and WILL lead to frustrated players since they then have to replay parts of the game.
3. I have somewhat removed superfluous combat by making it mandatory to have a "bailing-out" condition for every combat that occurs, in case you lose.
4. I might have still annoyed some players because without the save/reload feature, they cannot go back and "win" the fight (even though I've made abundantly clear through design that there IS no losing this game). If you are one of those players at this point, all I can ask you is "why are you playing games? Rent the movie, and watch that instead." Or I could add a "simulation" mode, where you can choose the outcome of each challenge as you see fit, making it a "toy" rather than a game.

This is a treatment that really only works well for some types of CRPG, and not at all for many other types of game. I think FPS games with a storyline need an entirely different focus, because the combat there is the central issue.

People might not remember what you said, or what you did, but they will always remember how you made them feel.

[edited by - MadKeithV on May 21, 2002 5:00:27 AM]

quote: Original post by Grellin
Just out of curiosity, how would the designer have any idea when a situation will arise (need to leave, going to bed, phone rings, aliens invade) to be able to "decide" when a player can save?

For the last #!@(#$(@@#$ time there is a difference between

SAVE AND QUIT
vs.
SAVE AND RELOAD

Please don''t keep responding if you aren''t willing to read, this has been mentioned numerous times. We are discussing save & reload. AFAIK nobody has a problem with save and quit.

quote: Original post by deClavier
The save/load game feature is that mechanism by which a player learns the language of the game, providing the opportunity to compare the success of various syntagmatic (sentence-like) structures within the context of the game''s paradigm (wordset-like). Iterating the load/save teaches the player to think using the game''s language, essentializing their goals on optimal game states.

Under this oddly worded logic, simply starting all over from the very beginning also accomplishes the same thing. In fact, starting over from the beginning may very well accomplish it better.

You seem to be drawing the line between play through once vs. save and reload. But if anything without saves the line is play through from the start a bunch vs. play each tiny section a few times without context.

Let me ask you this: Would you consider the following an example of learning?

1: Begin by choosing any word from the language.
2: If the word cannot be used to start a grammatical sentence, start over.
3: Choose the next word.
4: If that word, following the fist can not be used to start any grammatical sentence, start over.
5: Repeat ad infinitum.

Congratulations! You are now the master of grammar! Your method of "learning" assures that yes you will produce a proper sentence.

That''s silly though huh? You don''t write by randomly choosing words and backtracking when they don''t work out. You learn to construct entire sentences from start to finish.

quote: Original post by MadKeithV

Having to repeat that talking the same way over, and over, and over, just to hear a bunch of things you already knew, is absolutely worthless as gameplay.

I agree. This is why any designer who decides to restrict save/reload should allow the player to save his progress in a way that doesn't require replaying conversations or things of that nature. In CRPGs with save spots, a save spot in every town or meeting place would accomplish this.

quote: you're at an inn, and there's a group of ruffians hell-bent on getting into a fight with you. You can get into this fight, and there's hitpoints and wounds and the whole common RPG shebang. However, should you not kick royal behind, then the town guards will bail you out right before the lights go out. The effect is that your reputation diminishes somewhat, but you don't have to replay talking to every gardner in town to find out where you might find a white flower of gandilea to trade for the sword of mighty shamrocks at the wizard's place.

Dealing with failure this way is just fine, but if the penalty for loosing a battle is always as small as in this example then may I suggest that your game is more like an adventure game (like Lucasarts' SCUMM games) than a CRPG. I quite like adventure games, and non-fatal failure is a perfectly valid way to design these kinds of games, but for other types of games bigger penalties for failure might make more sense (at least on occasion).

[edited by - chronos on May 21, 2002 6:50:07 AM]

MadKeith I don''t really agree with what you are saying.

First, good games will allow you to save before a part where you are likely to lose. That is why save points come before bosses. I also pointed out earlier that most RPG''s allow you to save anytime in the overhead map. So you have talking/over head map exploration where you can save anywhere. The places you can''t save anywhere are "dungeons."

What you have done is totally changed the nature of the game, which is ok, but it isn''t an improvement as much as a difference. You can make games like the one you described, but plenty of other people will still like the old style of lose and you''re dead.

--------------

Some people like challenge. Some people play games for challenge. Challenge with benefits but also risks. What people constantly gloss over is that save anwhere is less challenge.

I think a fundamental sticking point is people claim that if you have save anywhere but you want challenge, you can just not take advantage of it. That sounds happy and wonderful but I don''t think that it is true.

If someone recently had a heart attack, would you take them to a Steakhouse for dinner? Probably not, even though sure they *could* order fish. People do things they KNOW are bad for them. This is a given. It CANNOT be debated.

I drink. I assume none of the save/reload people drink. Or smoke. Or drive too fast. Or fail to get regular cardiovascular excersize? Is there a person in the world who drinks heavily, knows it will kill them, and still does it? I would guess yes...am I wrong. Is that really atypical behavior? Surely if people will literally *kill* themselves through drug use or poor diet or what-have-you they might also be willing to reload a game a bunch even if in the end it isn''t that enjoyable? No?


It is not true of *every* person, but in general it is very true. People choose short-term gain over long-term gain, even when the long-term gain is much greater. That''s a fact.

Designers should not ignore real life, real patterns of behavior and real psychology. If you include cheats people will use them. If you package the game with a hint book people will read it. If you allow people to save anywhere, most of them will. To ignore that is stupid. We don''t design games for the mythical iron-willed person, we design games for people. We design games for typical people, not how we imagine people *should* behave but really don''t.

Yes, it''s a choice. But people will KNOWINGLY make the wrong choice. Once again this is not up for debate. That''s human nature.

If you include save/load anywhere, most people will use it regardless if they realize it decreases their fun or not. The person who posted about Emulators was making this point, one I had thought of bringing up earlier. Playing a game on the emulator is a fundamentally different experience. It becomes a toy rather than a game. I like the save/reload for games I have already played where I just want to get to a certain level, hear a certain music or whatever - to appreciate individual moments of the game, but not actually play through the game as a game. The game becomes a series of loosely connected segments. When I try to play emulated games I find them much less involving, much less challenging, the flow is terrible, the excitement is diminished, the thrill is gone - yet I *will* save and reload. And so will most people. I''ve also never played all the way through a game on an Emulator, even games that I later *have* played through on the original system.

So, you can put your hands over your ears and ignore reality, or you can deal with. Games are products people use, not a PhD dissertation. Pragmatism is important.

I can imagine the following conversation:

Person: "Well that game was too easy, I just kept saving and reloading."
Designer: "Well why did you do that then? Didn''t you see it was ruining your fun?"
Person: "Yeah, I did...but I kept saving and reloading, and it was too easy."
Designer: (Frustrated) "Well you shouldn''t have saved and reloaded so much then huh?"
Person: "Yah whatever I guess so...I''m going to return this and pick up CounterStrike."

So who is to blame if the person doesn''t buy the designer''s next product?

I made the dialog sound a bit silly, but the situation is very real. As I mentioned, just go read some game reviews sometime on say gamefaqs.com. You see all sorts of posts like:

"Once you figure out this technique the game becomes really easy. Challenge - D+"

Once you get the super-strong character in Final Fantasy Tactics the game is too easy. Once you get the SnapDragon weapon in Ogre Battle, or a Princess and a Liche together, the game(s) is too easy. Once you learn to save your spin-sword in Ninja-Gaiden the game is too easy.

The obvious question is: "If too easy is a bad thing, why didn''t you just NOT USE THAT ??!?!?!"

Obvious question. The answer isn''t even that important though. The fact is, people will make the sorts of complaints mentioned above. The important thing is that they complaints are made, despite the fact that the person could have "chosen" to have more fun. Can anyone respond to this point directly?

Note that if you like challege you can (in theory) make ANY game arbitrarily difficult by limiting yourself. If I think Contra is too easy I can try to beat it only using 10 shots a level. That''s my "choice." If Super-Mario is too easy I can never pick up mushrooms and never grow big. That''s my "choice." But, in Contra I don''t have limited ammo - the ammo is there and I use it. If I don''t have limited saves, maybe I use them too. Sure, I could "choose" not to, just like I could choose to beat Street Fighter 2 using only jab and short.

Under that logic, you should make your game as easy as possible, and allow people to invent ways to make it harder if they so choose. Maybe in ResidentEvil make it so it takes 40 hits to kill you - hey if that''s too easy every time you are hit just get hit another 9 for good measure - right? Just choose not to take advantage of game mechanics, it really is that easy!

Make everything in Starcraft cost max 10 minerals and gas. If that''s too easy, just don''t collect a lot of gas and minerals! The power is yours! Why should we force people to play a certain way, after all? If they want Resident Evil to play like "Resident Tourist" let them! Let''s just stand here and look at that pretty texture for a while...Zombie, oh those take 30 minutes to kill you, don''t worry about it. Besides there are unlimited green herbs in the next room...(that you could choose not to take of course!)

Why should we even bother having difficulty settings in games? Just make one called "absurdly easy!" Hey kids, think this game is too easy? Try playing upside down, or with your feet! Challenge ensues!

I hope it is obvious what all these crazy examples are supposed to illustrate. People rightfully expect to take full use of the options you give them. If you give them a Green Herb they use it, if you give them unlimited ammo they don''t limit themselves, if they suspect a mushroom is in a brick they''ll stop and hit the brick. Why then should we expect them to arbitrarily limit their saves if they can save anywhere at any time? If they can jump anywhere at any time do we expect them to maybe limit that too for some reason? Maybe because jumping makes the game too easy?

You can try to make the distinction that the save game is a meta-feature or something, but in the end it''s a feature, it''s an option, it''s a part of the game. It isn''t separate from the game, any more that the placement of powerup mushrooms or green herbs or the amount of ammo is. Saves are a resource like anything else, and their timing and number of limitations effect the game as much as anything else. In fact, when you think about it, save have a much larger effect in some cases than green herbs, or mushrooms, or ammo. Save/reload might avoid damage, cure poison and prevent you from falling in a hole all at the same time.

As a final experiment, imagine this:

In Mario you have an item called a "safety net." When you use the safety net you cease doing whatever you were doing and drop down from the top of the screen. You have an unlimited number of safety nets. How does this effect gamplay?

The answer should be obvious. Anytime you are about to die you simply use your safety net. The game becomes about how quickly you can hit the safety net input. Missed that jump? Quick press up+B!

Would including this item in Mario games be a good idea? Remember, players can choose not to use it! (Just don''t be surprised when reviewers call it the single dumbest idea ever in a game)

-----

I still haven''t heard why Tekken needs save/reload anywhere. As far as I can tell I''ve easily "won" this debate. I''ve stated now 5-10 times that many games are very appropriate for save/reload. So to me the argument looks like:

Save/reload can be bad
vs.
No, save/reload is ALWAYS good

Of course there are hundreds of trivial counter-examples to the second argument. Tell my why the Tekken arcade machine would be better for having save/reload anywhere. If you can''t I win, or else you have to clarify your argument. Is that unfair? Tekken is a computer/video game, designed by designers, right?

Should we extend it outside computer/video games? Should baseball have infinite "redos?" Should you be able to take back x number of moves in chess, as often as you want?

So, everyone arguing for save/reload, you have 2 and only 2 choices:

#1: Convince me that Tekken really needs save/reload anywhere.

#2: Specify your argument better to make it look less silly and trivially false. Because right now it is *trivially* false. Do I need to be able to save and reload my Tekken game in the middle of a 10-string? I''m dubious.

quote: Original post by chronos
Dealing with failure this way is just fine, but if the penalty for loosing a battle is always as small as in this example then may I suggest that your game is more like an adventure game (like Lucasarts' SCUMM games) than a CRPG. I quite like adventure games, and non-fatal failure is a perfectly valid way to design these kinds of games, but for other types of games bigger penalties for failure might make more sense (at least on occasion).

Who says that losing reputation is only a small penalty in a role-playing game? You may be laughed at in the next town, for instance, or have to resort to underhand actions to get certain things done. Your character's "role" can be influenced greatly. Example: if you have a really bad reputation, instead of being able to ask the town mayor for the Book of Awful Recipes that you need, you now have to break into the town hall through the sewers, leaving you open to much worse abuse later...

I mean, you agree with me first that replaying certain parts over and over is bad, yet you still want "fatal failure", while simply being able to restart from right before the fatal failure? Why have fatal failure at all? Just don't let the player lose, instead change the experience for the rest of the game.

This does NOT apply to arcade-style games, obviously, where getting toasted is generally a very important part of the gameplay, and replaying sections is not such a bad thing because it contains a lot of the essential elements of the gameplay.

quote: Original post by AnonPoster
First, good games will allow you to save before a part where you are likely to lose. That is why save points come before bosses. I also pointed out earlier that most RPG's allow you to save anytime in the overhead map. So you have talking/over head map exploration where you can save anywhere. The places you can't save anywhere are "dungeons."

I thought the object of the discussion was "why have saves at all", that is why I argued from that point of view. I know that most current games allow that kind of save, but it accomplishes little in most cases. If the penalty for losing is simply getting to try again until you don't lose, I don't really see the point - especially if that part of the game is not really the focus of the gameplay. Just make sure you can't fully lose. The save system points out that RPG-players do not accept losing in a combat (unlike Mario players, who accept that sometimes you run out of lives and start over).

Earlier, there was quite a good point about "losing" and combat - if you can "pull back" from a near loss, then things would not be so bad without saves, without changing the underlying nature of the game by an extreme amount. If at any time you feel like you're about to die, you can pull out of combat and go lick your wounds, and get to try again at a similar strength to the first time, that would put the "save" option in-game, and at the player's discretion.
If you choose not to pull back, that's the same as choosing not to save. You don't get to bitch about it, because you chose to risk your hide by continuing, plus it keeps the timeline of the game continuous instead of the odd time-jumping when you save/reload.
You could leave a single absolute failure condition in certain play-style, being a time limit. That would be the only limitation on how many tries you get. Of course, you now potentially have people saying "I didn't know I spent too much time getting to checkpoint A in the first three weeks of the game, and now I don't have enough left to finish the game so I have to start over", so feedback about the expected time necessary to finish the game might need to be included at this point (or the limit set realistically for even a weak player).

quote: Original post by AnonPoster
What you have done is totally changed the nature of the game, which is ok, but it isn't an improvement as much as a difference. You can make games like the one you described, but plenty of other people will still like the old style of lose and you're dead.

Removing saves is also totally changing the nature of the game, to which the exact same argument applies. It's a careful balancing, and save/reload is a VERY easy tool for the designer to cover up minor to major design and balance flaws.

[edited by - MadKeithV on May 21, 2002 7:14:29 AM]

What comes first?

Save feature or Game design?

If the save feature is such a big discussion topic... at what point does it enter into the design? Do we finish our design first, then, almost as an after-thought, try to insert a save feature that we think will work. Or should we go for the complete opposite? Start out by coming up with a save feature at the very start of the design phase, and tune the design of the game and the design of the save feature to eachother?

Because I feel that in order to create a good save feature, you have to adjust your gameplay and game design. You can''t just tack on the save feature. You have to think about limitations. You have to think about how the save feature can actually improve the gameplay itself. Maybe you can come up with a save feature that actually becomes part of the gameplay. In order for a player to get everything he can out of the game, saving (and loading) becomes part of the game.