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.

Editorial Rampant Coyote on Dice Rolls and Luck

Infinitron

I post news
Patron
Staff Member
Joined
Jan 28, 2011
Messages
99,596
Codex Year of the Donut Serpent in the Staglands Dead State Divinity: Original Sin Project: Eternity Torment: Tides of Numenera Wasteland 2 Shadorwun: Hong Kong Divinity: Original Sin 2 A Beautifully Desolate Campaign Pillars of Eternity 2: Deadfire Pathfinder: Kingmaker Pathfinder: Wrath I'm very into cock and ball torture I helped put crap in Monomyth
Tags: Jay Barnson; Rampant Games

Jay Barnson (that's the developer of Frayed Knights in case you've forgotten) has an insightful new blog post. After a particularly brutal experience with the dice in a board game called Talisman, Jay had some thoughts about the ideal degree of randomness in a CRPG. Here's an excerpt:

You’d think that with a few stories like this one (and I have a lot of ‘em), I’d be in Craig Stern’s camp, all “randomness in RPGs, BOO!” But no, I’m actually a fan, seeing it much more like Daniel Cook in his article, “Understanding Randomness in Terms of Mastery.” Particularly with modern RPGs, I really tend to go with systems that allow enough strategy to either minimize the role of luck of I’m at an advantage (which in normal RPGs tends to be the norm), or to actually maximize its role when I’m at a disadvantage. I try to see what I can do to get the numbers on my side (my side often meaning, “my team’s side” when I’m playing with others.) That skill – and being able to discern see through the random ‘noise’ as Daniel Cook explains it – are key items that make me feel like I’m in control of the game.​

So if I’m so focused (usually) on marginalizing randomness, why do I not favor getting rid of it altogether?​

In three words: Because it’s fun.​

...​

It just comes down to making sure that a little bit of bad luck doesn’t ruin the game. Major failure (the kind that requires a reload) shouldn’t feel arbitrary. This can be resolved by at least three ways:​

#1 – The game system is forgiving enough with random chance that deviations are limited and have limited impact on the game. Most JRPGs (at least that I’ve played lately) are like this – misses and crits are rare, and damage is in an extremely predictable range. But the occasional misses and crits that give it some “spice.”​

#2 – The game system grants the player the ability to ‘equalize’ bad luck (or setbacks) – like Frayed Knights‘ drama stars, or the ‘overdrive’ meter in (non-random!) fighting games which can fill up from (among other things) taking hits.​

#3 – The game AI pulls the sort of dramatic intervention a human might in a dice-and-paper game, and tweaks AI skill or decision making to give an unlucky player an opening to recover from setbacks.​

Chance can be a very fun element in games, but it can also suck the fun right out of a game if handled poorly, as I felt during the Talisman game. It really comes down to whether or not the players can feel like they are either the masters of the odds, or masters in spite of the odds. But they should not feel like they are at the mercy of the random number generator.​

In short, randomness is good, but give the player enough breathing space and options with which to manage it and avoid arbitrary failure scenarios. I wonder if Jay has been reading Josh Sawyer lately - their approaches seem to be quite similar.
 

Infinitron

I post news
Patron
Staff Member
Joined
Jan 28, 2011
Messages
99,596
Codex Year of the Donut Serpent in the Staglands Dead State Divinity: Original Sin Project: Eternity Torment: Tides of Numenera Wasteland 2 Shadorwun: Hong Kong Divinity: Original Sin 2 A Beautifully Desolate Campaign Pillars of Eternity 2: Deadfire Pathfinder: Kingmaker Pathfinder: Wrath I'm very into cock and ball torture I helped put crap in Monomyth
Also, how does this have anything do with what Josh Sawyer is saying?

Pro-gamist, anti-savescumming approach. Against binary success/failure dice rolls. Click the link, read the entire post:

But what about those days when the dice are cold, bad luck reigns, and things just turn against you even when you have every advantage in your corner?

In a good gaming system, IMO, you still have options. The option may be “run away to live and fight another day.” It’s why I’m not a fan of “critical failures” all that much – or older editions of D&D which had frequent “save or die” rolls (although many editions of the game made that merely an expensive, not necessarily game-ending, problem). This is one reason I’m still okay with the “hit point” mechanic, in spite of its lack of realism: It allows non-critical attrition to occur so you have a reasonable chance of executing a “plan B.” Or “C” or “D.”

In human-moderated RPGs, even a poor system can be compensated for by a good game master. Even without fudging the rules, they can do things like deliberately forcing mistakes on the part of the bad guys to allow the players an opening to turn things around or escape. Unfortunately, computer games generally don’t do that. But unless the games do the “permadeath” thing, there’s always the handy reloading the saved game option to compensate. But that’s a poor substitute for a good game system or a well-moderated one. Having to reload a saved game because of some randomly bad things happening is not fun. It feels arbitrary.

This could easily have been written by Josh.
 

felipepepe

Codex's Heretic
Patron
Joined
Feb 2, 2007
Messages
17,310
Location
Terra da Garoa
Relevant to the subject, reading the wiki I discovered that the new XCOM actually lies about to hit % on lower difficulties, raising it by 120% on the actual roll. So your to hit chance on easy is not the displayed 80%, but rather 96%... must solve a lot of casual players issue with %.

Back to the article, I'm on the pro-luck side, especially after playing Craig's Telepath Tactics demo. The lack of RNG turns the game into mere execution, and while that may be great for some, IMHO it loses that "will it land?" feeling, that inside cheering for your strategy to work, that glimpse of hope that even if you did a tactical mistake the enemy might miss and you'll be saved... I like that, I think it's part of the game, that it makes your playthrough more personal.

Even on silly things, like WoW pets I think that's cool. A bro of mine got a über-rare sea turtle mount (0,03% drop rate) by casually fishing for the lulz, while thoudands of players farm it everyday and never got it. Is his story, his experience, and it's unique. If everyone got the mount after 500 fishing attempts, it would be just a matter of spare time and execution, and everyone's story would be "I fished 500 times".

IMHO, only people really butthurted about bad luck bitch about things like that... "urrr, I did Molten Core 200 times and never got the rare as fuck legendary blade, this game is broken, the GM should give me one to make it fair"... sadly, we all know there are tons of those douches out there.
 

Curious_Tongue

Larpfest
Patron
Joined
Mar 2, 2012
Messages
11,905
Location
Australia
Codex 2012 Codex 2013 Serpent in the Staglands Codex USB, 2014
I'm pretty sure that developers assume that a lot players are save-scummers. This makes randomness more interesting and less frustrating.
 

FeelTheRads

Arcane
Joined
Apr 18, 2008
Messages
13,716
This is one reason I’m still okay with the “hit point” mechanic, in spite of its lack of realism:

HURR DURR REALISM. Because we all know mechanics are usually good or bad based on their realism.

But what about those days when the dice are cold, bad luck reigns, and things just turn against you even when you have every advantage in your corner?

What about it? Oh my god, DO NOT WANT TO FAIL!!!11 plz remove RPGs from RPGs they suck!!!!!

This could easily have been written by Josh.

As far as I can tell it could easily have been written by Bethesda or insert-random-gaming-journalist.
 
Self-Ejected

Excidium

P. banal
Joined
Aug 14, 2009
Messages
13,696
Location
Third World
Captain Obvious said:
Chance can be a very fun element in games, but it can also suck the fun right out of a game if handled poorly, as I felt during the Talisman game. It really comes down to whether or not the players can feel like they are either the masters of the odds, or masters in spite of the odds. But they should not feel like they are at the mercy of the random number generator.
Ok.
 
Joined
Aug 28, 2012
Messages
997
Location
Dreams, where I'm a viking.
Project: Eternity Torment: Tides of Numenera
This is one reason I’m still okay with the “hit point” mechanic, in spite of its lack of realism:

HURR DURR REALISM. Because we all know mechanics are usually good or bad based on their realism.

But what about those days when the dice are cold, bad luck reigns, and things just turn against you even when you have every advantage in your corner?

What about it? Oh my god, DO NOT WANT TO FAIL!!!11 plz remove RPGs from RPGs they suck!!!!!

The full statement from that post is actually:

But what about those days when the dice are cold, bad luck reigns, and things just turn against you even when you have every advantage in your corner?
Sometimes that can be kinda fun, too.

I didn't read him as saying randomness should not cause problems for the player, rather as criticizing the kind of all-or-nothing randomness that allows for catastrophic failure with no opportunity for the player to exert control over the situation. So he likes HP because it allows bad things to happen to the player, but in a way that still allows the player to exert some control over the negative effects (heal, escape before you lose them all). He dislikes save vs. death because it doesn't provide that.

I think the implicit comparison is between a game that, for example, gives you enough hit points to survive 20 enemy attacks and a game in which each enemy attack requires a save against death, where the probability of failure is such that you will survive an average of 20 hits before dying (meaning you could die on the first hit or the thousanth, but it will average out to 20).

His examples were a bit popamolish, but the idea itself isn't bad.
 

mediocrepoet

Philosoraptor in Residence
Patron
Joined
Sep 30, 2009
Messages
13,516
Location
Combatfag: Gold box / Pathfinder
Codex 2012 Codex+ Now Streaming! MCA Project: Eternity Divinity: Original Sin 2
Back to the article, I'm on the pro-luck side, especially after playing Craig's Telepath Tactics demo. The lack of RNG turns the game into mere execution, and while that may be great for some, IMHO it loses that "will it land?" feeling, that inside cheering for your strategy to work, that glimpse of hope that even if you did a tactical mistake the enemy might miss and you'll be saved... I like that, I think it's part of the game, that it makes your playthrough more personal.

This is pretty much exactly my perspective on it. It's why in the tabletop realm, systems like Amber never held much appeal to me (since resolution is about looking to see who's stat is higher and they win, no dice involved) and why although a game like chess is interesting, it's not fun in the same way that a randomized RPG system can be.

Besides, even though it can make you reload, having the possibility of a lucky crit that instantly gibs one of your party members like I've had happen a few times during my IWD2 replay, gives you that bit of tension to help keep you interested in what's going on and approaching certain enemies with caution and weighing out what spells and resources you have at your disposal to ensure it won't happen. If you can just look at the monsters like game pieces that do fixed damage and such, there's not the same uncertainty and you can generally figure out whether or not you need to waste buffs/consumables on the fight or not just by glancing at the overall layout. It's an entirely different situation and feel for a game.
 

RK47

collides like two planets pulled by gravity
Patron
Joined
Feb 23, 2006
Messages
28,396
Location
Not Here
Dead State Divinity: Original Sin
I believe Bloodbowl offered a balanced dice roll vs dice manipulation mechanics.
We have the standard 6-sided to determine most success/failures and a roll modifier skills to reflect how an experienced player is better at doing things than a complete rookie ever will.
Over 100 hours played and it's still the most transparent (AND fun) dice-based system I've every played I think.
 

Esterhaze

Augur
Joined
Jan 15, 2012
Messages
123
For most "isolated" skill checks, I'm an advocate of a random factor being calculated at the start of the game and then set in stone (hidden from the player). This takes the "edge" off of skill checks (that is especially amplified if the player knows the cutoff requirement, i.e. I know that I need a 7 INT in fallout to receive certain dialogue options) and at the same time greatly discourages reloading. Making skill checks non binary as well is also ideal.

As for combat, real time dicerolls should obviously remain, as a) removing them would make for unenjoyable, not to mention unreal-feeling combat and b) if you are compulsive enough to reload every turn to optimize performance on something that is generally as common in rpgs as combat you deserve the pain. If there was some hypothetical rpg where there were like 3 total combat encounters, then I would reconsider the above stance, and say that calculating the random factor at the start of the game and setting that calculation into a deterministic combat framework would be ideal. But wake me when somebody makes a game like that.
 
Self-Ejected

Excidium

P. banal
Joined
Aug 14, 2009
Messages
13,696
Location
Third World
For most "isolated" skill checks, I'm an advocate of a random factor being calculated at the start of the game and then set in stone (hidden from the player). This takes the "edge" off of skill checks (that is especially amplified if the player knows the cutoff requirement, i.e. I know that I need a 7 INT in fallout to receive certain dialogue options) and at the same time greatly discourages reloading.
What a terrible idea.
 
Self-Ejected

Excidium

P. banal
Joined
Aug 14, 2009
Messages
13,696
Location
Third World
Checks should have a target number that is fitting for the difficulty of the task. Using a random number either fucks that up or the variation is too small to make any difference.

Doesn't AoD have that, by the way? People still hoarded skill points.
 

Esterhaze

Augur
Joined
Jan 15, 2012
Messages
123
Obviously you would adjust the probabilities to fit the demands of varying degrees of difficulty. I.e. instead of requiring minimum of INT 7 for a dialogue check, make it the mean of something like a binomial distribution, so, greatest probability of requiring INT 7, lower but still significant probabilities of INT 6 or 8, etc. I'm not convinced that variation calculated thus would be too small. I don't see how this isn't better than a totally static skill check or a dice roll in realtime ( i.e. INT 7 means 78% chance of success or whatever). edit: with the former, prior knowledge of game affects how you would approach the encounter, and can essentially guarantee success, which is bad. With the latter, because of reload, failure means nothing unless the chance of success is very low or you restrict saving in some bullshit way. This is also bad. My approach avoids both of these pitfalls.

edit again: you'd have to communcate to the player that the mechanics work in this way, but I don't see how that's an issue
 

J1M

Arcane
Joined
May 14, 2008
Messages
14,739
No. Your suggestion encourages reloading. Try using the skill, see if you pass. If not, reload and try something else.
 

Infinitron

I post news
Patron
Staff Member
Joined
Jan 28, 2011
Messages
99,596
Codex Year of the Donut Serpent in the Staglands Dead State Divinity: Original Sin Project: Eternity Torment: Tides of Numenera Wasteland 2 Shadorwun: Hong Kong Divinity: Original Sin 2 A Beautifully Desolate Campaign Pillars of Eternity 2: Deadfire Pathfinder: Kingmaker Pathfinder: Wrath I'm very into cock and ball torture I helped put crap in Monomyth
No. Your suggestion encourages reloading. Try using the skill, see if you pass. If not, reload and try something else.

Precisely. "Pre-baking" randomness is just a way of telling the player "OK, this is the thing that you must never do. Now reload and try something else." whenever he fails at a specific action.

For added ridiculousness, it may work AFTER he has tried something else because the pre-baked random seed has now shifted in his favor.
 

Esterhaze

Augur
Joined
Jan 15, 2012
Messages
123
Just in case this isn't clear, the "pre baked" randomness I'm proposing about would be set at the start of the game. It would not be reset with each reload. Only at the creation of a new character. Thus, for all intents and purposes it would be set in stone, no matter how man times you reloaded, just as AoD's skill checks are, if I recall correctly. The only functional difference between this system and a system like AoD's (or something like Fallout New Vegas, without the visibility) is that even if the player was familiar with the game before hand, he would not be able to know precisely what would be necessary to complete various tasks skill wise, as is the case ideally in your first playthrough of a game like AoD or Prelude to Darkness, that does not use dicerolls in real time.

I suppose you could start another character from scratch upon failing to pass a skill check, but if you are that obsessive compulsive there are better hobbies for you than RPGs.

"Pre-baking" randomness is just a way of telling the player "OK, this is the thing that you must never do. Now reload and try something else." whenever he fails at a specific action.
How is this different than if the skill thresholds are hardcoded? I mean intrinsically binary skill checks [pass/fail] in a game that is at all reloadable are going to encourage gamey behavior, especially if they are presented without suffiencient frequency, diversity, and context. The player, I suppose, could reload and try another path, though they can do that with any check system. They could leave, improve their metrics if the character system allows it, and come back and try again if the situation allows for it, but again this is no different than any other skill check system.


it may work AFTER he has tried something else because the pre-baked random seed has now shifted in his favor.
no, see above
 

Haba

Harbinger of Decline
Patron
Joined
Dec 24, 2008
Messages
1,872,096
Location
Land of Rape & Honey ❤️
Codex 2012 MCA Divinity: Original Sin Project: Eternity Torment: Tides of Numenera Wasteland 2
Hah, glad to see that esteemed Staff members retain the long cherished 'dex tradition of failing at reading comprehension.

Esterhaze is onto something good here. Some minor fixed randomized adjustments to all fixed skill checks would definitely be a great addition. It certainly wouldn't completely prevent min-maxing for the checks, but it'd be a good start.

Possibly one would have to balance the adjustments overall so that you can't end up with +2 in every check for the rest of the game if you get unlucky, but that is rather trivial to do.

But going back to the topic at hand, the actual mechanics are rather trivial in the end. The important thing is that combat encounter feels like you just survived several tough near-death experiences through your own skill and occasional luck. I wouldn't mind the computer occasionally bending the rules behind the scenes to ensure that - as long as the illusion isn't broken.
 

Infinitron

I post news
Patron
Staff Member
Joined
Jan 28, 2011
Messages
99,596
Codex Year of the Donut Serpent in the Staglands Dead State Divinity: Original Sin Project: Eternity Torment: Tides of Numenera Wasteland 2 Shadorwun: Hong Kong Divinity: Original Sin 2 A Beautifully Desolate Campaign Pillars of Eternity 2: Deadfire Pathfinder: Kingmaker Pathfinder: Wrath I'm very into cock and ball torture I helped put crap in Monomyth
Hah, glad to see that esteemed Staff members retain the long cherished 'dex tradition of failing at reading comprehension.

Esterhaze is onto something good here. Some minor fixed randomized adjustments to all fixed skill checks would definitely be a great addition. It certainly wouldn't completely prevent min-maxing for the checks, but it'd be a good start.

Possibly one would have to balance the adjustments overall so that you can't end up with +2 in every check for the rest of the game if you get unlucky, but that is rather trivial to do.

Where's the reading comprehension failure here? You want a game's stat check thresholds to be randomized on character creation, in an otherwise static, deterministic world? I don't see the point - it seems extraneous.

"Oh wow, I had to roll a 4 to open this door in my last game, but now I need to roll a 5! SO DYNAMIC!"
 

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