I don't mind a mini-game that's harder or easier based on your character's skill level in the applicable skill(s). I do think the lock-picking mini-game, if any, has been fairly successful in this regard, especially considering it's pretty quick unlike the hacking mini-game that a lot of people don't seem to like bothering with. And I sympathize.

I dislike rolls with a % chance. Yeah, that's very pen-and-paper role-play'ey, but it just promotes save scumming and, frankly, I don't like how big an impact pure luck on the roll plays. A stronger character should do better on strength checks across-the-board, not sometimes do worse than a weaker character because he made a bad, completely arbitrary roll. In a pen-and-paper situation you'd invent some reason for why this happened, but this doesn't exist in a computer RPG where literally everything is pre-defined.

I think a straight skill-check is fine for most cases, and then do the mini-game thing where it makes sense IMHO.

There are ways to discourage savescumming - like, for one example, timed attempts based on the lock level and skill; and limited tries based on the lock level. And if one still wants to go through the bore of trying and reloading for however many times, why not let them if that's what they prefer to do?

Dicerolls work exactly like you described you wanted them to. Stronger characters are better across the board since they have higher chances, but the chance doesn't depict a random blind attempt of all the time. It mostly depicts that the character is doing the job, and that even strong characters make mistakes (but, of course, to lesser extent than weaker ones).

It is only fair gameplay, whereas minigames downgrade the skill to a mere gating mechanism, or - like in Skyrim and Oblivion - downgrade the skill to a nigh meaningless metric. Minigames rely on the players hand-eye-coordination which leads to minigames being generally either too hard or too easy regardless of the characters skill.

If there were no dicerolls, I'd rather they ditched the minigames aswell, and have the skillgating work so that with the threshold met you simply open the lock, and if not met, you don't open it. Without boring me to death with repetitious minigames that usually are so easy that it questions the reason for the gating to be there to begin with.

I don't necessarily think you need to prevent save scumming. People can play the game however they want. But a mechanic that tried not to promote it, resulting in a more flowing experience for everyone, seems preferable to me. As always, the devil's in the detail of that particular mechanic.

The problem with dice rolls is that you can have, to keep the Strength thing going, a strong character fail a Strength check to, say, force something open, because they made a bad roll, but have a weak character succeed because they made a good roll. As I touched on, I don't think this is necessarily a problem in a pen-and-paper situation because you'll make up a logical reason, but in a computer RPG it makes little sense, only being depiced as a straight fail/success. By across-the-board I mean a stronger character shouldn't really ever be outdone in this particular area by someone weaker. That's his character trait. Let him play it, and others be directed toward other solution options pertaining to the traits they have gone for.

FWIW I don't think the latest system of set thresholds to even try (25/50/75...) and then introducing an element of chance (skill-affected mini-game) is that bad. I only think it fails in some instances because the mini-games aren't that great, e.g. how I can basically pass any Oblivion lock regardless of skill because the mini-game is too easy after a while. Mixing player and character skills to pass a check is maybe not ideal, but it does let you at least try if you're of a reasonable skill level (threshold). The balancing of the mini-games seems key here to me.

Can someone explain to me why the lockpick thresholds are all set at regular intervals(25/50/75/100)? I really don't understand why they are not spread out all over the 100 point scale. Is this to suggest there are only 4 types of locks? because that would be rather dumb. I really hoped Obsidian would change this but unfortunately I was left disappointed.

Not sure. I guess it maps nicely to "Very Easy", "Easy", "Medium", "Hard", "Very Hard" which is conceptually intuitive. Maybe one to ask JSawyer (http://www.formspring.me/JESawyer) about.

Because the system created in F3 was horribly arbitrary. Its just a poor design decision/massive simplification.

I guess it meant less work setting up locks with different levels, rather they could just stick a 'hard' lock here and an 'easy'lock there

Because there are only four kinds of locks. The idea (as I understand it), is just that the thresholds indicate (at a glance) complete certainty of success to the player. My guess about the equidistant values ~it's just that. They only have a scale of 100, and are not using it as a percentile ~so 25 to start and an equal amount between all thresholds. (And they could not put the top threshold any less than 100, for there would be no reason to ever increase the skill past whatever top number they picked.)

I do believe that it is specifically to address the problem of players reloading when "they" fail.

I am of the opposite opinion on this. Most RPG's that I've played (that use percentile) do so with a weighted skill system, and the results are not exactly random, and support PC failure due to circumstance. It really does a good job of depicting skill over adversity.
The problem I have with systems that do not do this is that impossible to believe; it makes for perfect characters in situations they should not really control.

For example: Two thieves; one's an expert, one's a 1st day novice. Both try to pick locked doors on a warehouse... the novice opens the lock by fluke chance (and a very, very unlikely thing that just happened too). The expert could not open his lock and did not understand why; and never finds out that his lock is very tough to open with the key.
Maybe he tries later and succeeds; maybe he dislodged some iron filings that were in the lock; who knows... The percentile system can plausibly depict probability, where the alternatives generally don't.

In any case, when using the percentile based skill system, the expert (in any skill) will generally succeed, but can fail; where the novice will generally fail, but can succeed ~but not always. Some systems account for difficulty with penalties. Where an expert (say with 95% to pick locks) encounters a security lock that is extremely tough (-65% tough). That would mean that the lock picking expert would only have a 30% chance of managing to pick it open ~and the novice would have no chance of any kind. In the case of Fallout... even a Luck Stat of 10 couldn't help with that.

Double post, please delete this one.

Hmm, I don't know.. It's still gonna svck when you hit a fail on that 95% chance-to-pick lock for no apparent/communicated reason. Seems like a system begging to be abused to me. I mean even trying to rationalize it on my own the way you suggest I honestly think it's a bit far-fetched to postulate that a master locksmith suddenly trips up on a rudimentary type of lock. It makes more sense to me that once your character is familiarized with a subset of them, he/she knows how to pick them (instant pass / threshold+adjusted mini-game), possibly has to take a chance only for ones he's not.

I'm also not sure I see the larger-scale gameplay purpose of maintaining a % chance. If someone's made a character that's good at picking locks, setting explosives, and fighting unarmed, I say let them take these paths through the game. Sure, if they haven't sunk enough skill points into these skills, they can/should not pass such obstacles of any remotely difficult kind, but if that's their focus I think they should be allowed to use those paths. Occasionally blocking that with a bad random dice roll does not seem productive in any way to me.

But are you not saying that your PC never ~ever makes a mistake?
Is one time in twenty too high a margin for error; for mistakes ~for accidents? (How about getting a cramp?)

The point is probability; consider that even a hotel hallway with 10 identical locks might have variance between them ~even variation that's invisible to the eye. Each lock has it's own history of events (we don't have to know what they are; one lock could have been re-keyed and a tumbler left out; another could have been nicked on the in the inside; another is just loose; another gets the rain when the hall window is left open... the exact reason is moot, just as the PC could have a head ache, could have received a static shock while kneeling on the carpet and touching the metal doorknob)... It doesn't really matter. What matters is that ~whatever the circumstance or past history, the percentile result represents the unique difficulty of the current attempt at the task, and the skill of the character allows them to succeed anyway ~or not.

In practice, the lock picker fails, (in this case gets a cramp) and tries again, and succeeds; if they don't and after several attempts they cannot open it, then they can't open it. (They should be guaranteed to open any specific lock of a given type?)

In the case of other skills... Try Speech: The PC can have a silver tongue able to con anyone... and yet talk to a mark that wants to say no, and is annoyed with them; will refuse logical ~even profitable offers just to be rid of them. A professional can also fail for no fault of their own, and percentile rolls can represent that; (and it also usually doesn't happen when they are highly skilled ~but it shouldn't be impossible should it?)

Are we allowed retries? I've been assuming this % thing to work similar to, say, F3 speech checks or the "force lock" option, i.e. catastrophic failure. I think I'd be more amiable to the idea if this is not the case. What do you mean by "then they can't open it," though? It sounds like the stake in what you're proposing is essentially time — the number of retries you can be bothered with; keep at it and you'll eventually make the check? — or was that supposed to imply that the lock will, say, jam after a few tries, F1/F2-style?

I don't think the speech example is really applicable. That character would be written without a speech check in his dialogue tree. Realism aside, I think it's valid to question the gameplay value in random failure. Does it make the game better/more fun? Does it aid or work against role-play? If we play with the idea of a % check for speech F3-style, I really think this works against. Why randomly prevent me from going the Speech route here that I've built my character for? I really think this just promotes save/load'ing, and while that's up to each player, I find it hard to promote a mechanic with this consequence as successful.

Why not? The old expression is if at first you don't succeed, try , try, again.
In a game with open ended rolls, it should be possible to gauge a success or failure by the value of the percentile roll. Roll low enough, and you jam the lock; roll high enough and you don't even scratch it.

I mean they cannot open it (now); and if it has a penalty that drops their skill to... (well some games would be zero, but I think in a fallout game it should be 5%), then they cannot open it until they improve (or get really lucky in Fallout's case).
* When I was six years old, I was out by the pool in our apartment building; I reached up over my head and spun the dial on a combination lock to the pool supply closet ~ then I pulled on it, and it opened. That was a fluke chance. I locked it again, and never succeeded in opening it a second time. This can happen, but it's not likely to.

In RPGs it's useful to have the probability of something be a guideline rather than an absolute rule; in a PnP game, the DM or GM can allow a novice to (for instance) manage to pick open a lock ~because it could happen, and if it serves the story, they can make it happen. Fallout 2 doesn't fail you on the first lock in the Temple of Trials; (or at least, I've never seen this happen). But having a strict rule that dictates success or failure at your current level, imparts the expectation that you can do it again (perhaps when you shouldn't be able to do it so reliably).


No... that would be awful... it robs the PC of the chance to succeed due to consummate skill and or a bit of luck. In that example it should just be very difficult, but not impossible ~perhaps so difficult that most players don't realize that it is possible. IMO it should only be impossible if that's exactly how the designers want it.

Just as I think it's valid to question the gameplay value of assured success. I generally lose respect for any RPG that follows that route.

IMO it facilitates roleplay, and I did not find Fallout's use of thresholds to be fun... less so for the fact that it was a radical change to the established skill system.

Because they are not perfect, and the alternative means that they are; even a professional with decades of experience can make a mistake, or have unexpected misfortune. It may be a very slim chance, but it's not impossible ~and shouldn't be IMO.

It does, but they would reload for anything that didn't go their way ~that's an ego trip where you back up to rewrite history (in the game) to force everything to go their way. Would you think the same same of a casino where you never lose because the player might reload until they win?

(http://www.youtube.com/watch?v=SC05Pb7Qd68. )