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.

Design Topic #2: RNG

Should we adjust the RNG to represent THC probability more accurately?


  • Total voters
    35

Goral

Arcane
Patron
The Real Fanboy
Joined
May 4, 2008
Messages
3,552
Location
Poland
http://www.irontowerstudio.com/forum/index.php/topic,7561.0.html
Vault Dweller said:
The solution seems to be simple: rig the RNG to deliver what the player expects (or at least avoid what everyone hates – missing 3 times in a row despite seemingly high THC), so I have two questions for you:

1) Should we rig the RNG to meet players’ expectations?
2) If yes, how? Meaning what should we aim it? What outcomes should never ever happen when your THC is 70-80%?
 
Last edited:

agentorange

Arcane
Patron
Joined
Aug 14, 2012
Messages
5,256
Location
rpghq (cant read codex pms cuz of fag 2fa)
Codex 2012
http://www.irontowerstudio.com/forum/index.php/topic,7561.0.html
Vault Dweller said:
The solution seems to be simple: rig the RNG to deliver what the player expects (or at least avoid what everyone hates – missing 3 times in a row despite seemingly high THC), so I have two questions for you:

1) Should we rig the RNG to meet players’ expectations?
2) If yes, how? Meaning what should we aim it? What outcomes should never ever happen when your THC is 70-80%?
Keep it random as is. Any rigging and it would feel like I only won a battle because the AI decided to throw the game, which does not make for a satisfying win. As for players that get aggressively frustrated by this, they can get over it.
 

Goral

Arcane
Patron
The Real Fanboy
Joined
May 4, 2008
Messages
3,552
Location
Poland
My first reaction was "WTF? Decline!" but then I thought that it is strange that a trained soldier (not an elite but a trained one) would miss 3 times in a row a target that is rather close. It's especially ridiculous for melee when you're in point-blank range (there is a reason why knife attacks are so dangerous, for big hammers it's might be a different thing but not for lighter weapons, hence shit-skins knife attacks are so deadly in Western Europe even when their skills are amateurish and their IQ is lower than monkey's). Not impossible but very improbable unless you consider it as a pure event which I don't think is true because there are more factors in there which aren't considered by the formula. Although for guns it's much more probable than for melee so...

My proposition: change nothing internally but obfuscate numbers a bit, instead of giving percentages go with low chances, medium chances, high chances and only after you would click on a "low chance" label you would see the percentages but labelled as "estimated percentage". Or you could give an interval and instead of saying 70% you would err on the side of caution and say 50-70% so even if one would miss 3 times in a row it wouldn't feel as random.


 
Last edited:

agentorange

Arcane
Patron
Joined
Aug 14, 2012
Messages
5,256
Location
rpghq (cant read codex pms cuz of fag 2fa)
Codex 2012
I've always just taken it as video game abstraction. Say you have a 95% chance to hit. For me that means your character has a 95% chance to hit based on predictable factors such as training, hand stability, calmness, distance, etc. But not including outside, unpredictable factors, all those minor details that can't be adequately taken into account or shown in a video game. So say your highly trained marksman is lining up a shot and it's an almost guaranteed hit, but then the target suddenly bends their head in a way that wasn't expected, or some sweat falls into the marksman's eyes, or they hesitate for just a split second, or any of those million things that could take place during a heated battle.

Edit: Here's a film analogy as an example. The Jackal is basically a max level marksman character with a sniper rifle, 99% chance to hit, but his target just happens to crane his head at the worst moment possible, causing a miss.

 
Last edited:
Joined
Jan 18, 2018
Messages
1,301
Grab the Codex by the pussy
I've always just taken it as video game abstraction. Say you have a 95% chance to hit. For me that means your character has a 95% chance to hit based on predictable factors such as training, hand stability, calmness, distance, etc. But not including outside, unpredictable factors, all those minor details that can't be adequately taken into account or shown in a video game. So say your highly trained marksman is lining up a shot and it's an almost guaranteed hit, but then the target suddenly bends their head in a way that wasn't expected, or some sweat falls into the marksman's eyes, or they hesitate for just a split second, or any of those million things that could take place during a heated battle.
It happens in real life all the time. See this scene:



The shooter from BOPE, braziliam special ops, is close to the kidnapper, but misses the shot. He shoted the guy in the fucking face with a giant rifle and missed. BOPE kills more than malaria in Rio's favelas. They are trained to kill. But the guy missed the easiest shot possible, in close range.
 
Last edited:
Joined
Jan 18, 2018
Messages
1,301
Grab the Codex by the pussy
My first reaction was "WTF? Decline!" but then I thought that it is strange that a trained soldier (not an elite but a trained one) would miss 3 times in a row a target that is rather close.
It is decline. One of the problems with Age of Decadence is that battles become too predictable after you achieve a certain level. One of the few things that make them interesting is a series of bad rolls. You remove this and you are making the game even more predictable and boring. Your enemies can't compensate all your experience acquired with multiple quests and your tactical knowledge.
 

Parabalus

Arcane
Joined
Mar 23, 2015
Messages
17,442
http://www.irontowerstudio.com/forum/index.php/topic,7561.0.html
Vault Dweller said:
The solution seems to be simple: rig the RNG to deliver what the player expects (or at least avoid what everyone hates – missing 3 times in a row despite seemingly high THC), so I have two questions for you:

1) Should we rig the RNG to meet players’ expectations?
2) If yes, how? Meaning what should we aim it? What outcomes should never ever happen when your THC is 70-80%?

I think must multiplayer games use some form of pseudo-RNG, like Xcom on easy but unbiased, stuff like this:

$


Not sure how much it'd add to a single player game, some RNG can feel great and since it's against an AI you can always reload when you get screwed over, but continue on highs.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
This is casual gay shit.
Pretty much. First, they decide to implement storymode for retards who hate combat.
We did?

Now this.
Now what? First, we're discussing RNG and getting feedback. No decision has been made yet. Second, we're talking about not changing your actual THC but tying the number of consecutive misses to your THC. If your THC is 70%, you'll still hit 7 out of 10 and miss 3. The only question here is should you be allowed to miss all 3 in a row?

"While you’re thinking, here is how our RNG works. It draws numbers like cards from a deck, meaning you can’t draw the same card twice until the deck is out of cards and reshuffled. We round up, so if you draw 17, for example, you cannot get numbers 11 to 20 until the deck is reshuffled. This approach ensures that if your THC is 70%, you’ll miss 3 times and hit 7 times. If it's 63% though, you're not guaranteed to get 63 out of 100. Each 10 rolls you'll get 6 guaranteed hits, 3 guaranteed misses and 1 can go either way."
^
This is casual gay shit. Now as semi powergamer I am thinking into the past instead of the future to dictate my tactics... counting how often I hit someone (is dumb busywork) to optimize my next move... Blackjack, not vidya.
And the ingame display is probably gonna lie too. Like, 70 cth, you miss, your cth just went up but the game will probably show 70 on next attack. And when your next attack is a guaranteed miss you are still looking at a 70% kek, but not if you kept count and know.
But whatevs IDC
It all depends on what 70% THC means.

It can mean that out of 100 attacks you hit 70, so it's ok to miss even 20 times in a row. It's all good, man.
It can mean that out of 10 attacks you hit 7, guaranteed, so if you miss 3 times in a row, you KNOW that you'll hit the next 7 even though your THC still says 70% not 100% (which is how it works in AoD).
Etc.
 
Joined
Jan 18, 2018
Messages
1,301
Grab the Codex by the pussy

From the "Taking care of business" thread:

I wonder whether something like "Story Mode" is the way to package it, rather than as a difficulty level. It's silly that this kind of marketing/wrapping is so important, but it does seem to me that really there are just two classes of players at issue here, one that wants to play AOD for its narrative/C&C/exploration (but might still enjoy combats as filler material) and one that wants to play excellent and challenging tactical combat. I doubt the second set really needs a difficulty slider, nor does the first set, really, they just want to turn combat from content to filler.

Precisely. Realms of Arkania games let you autoresolve combat which didn't make these games any less hardcore.

I'm saying that offering options to customize the player's experience is not a bad thing. For example, some people didn't want to see the check tags like [persuasion], we gave them an option to disable them. Was it really a big deal? When it comes to skill checks, the top request is to show the numeric values of the checks. If we can make it optional, it would be great. It won't take anything away from people who don't want to see the tags or the values, but make the game more enjoyable for people who want this info displayed.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024

From the "Taking care of business" thread:

I wonder whether something like "Story Mode" is the way to package it, rather than as a difficulty level. It's silly that this kind of marketing/wrapping is so important, but it does seem to me that really there are just two classes of players at issue here, one that wants to play AOD for its narrative/C&C/exploration (but might still enjoy combats as filler material) and one that wants to play excellent and challenging tactical combat. I doubt the second set really needs a difficulty slider, nor does the first set, really, they just want to turn combat from content to filler.

Precisely. Realms of Arkania games let you autoresolve combat which didn't make these games any less hardcore.
Agreeing with MRY in general and stating that Realms of Arkania's autoresolve feature didn't make the game any less hardcore is not an indication of intent. I really don't see how you jumped from that post to "they decided to implement storymode". IF we ever consider going with 3 difficulty modes - and so far we haven't even discussed the difficulty modes internally - I will post open another design topic, present our thoughts on the subject and see what everyone has to say about it. Right now I have other things to worry about.

Skipping or autoresolving combat is not something we'd consider, ever.

I'm saying that offering options to customize the player's experience is not a bad thing. For example, some people didn't want to see the check tags like [persuasion], we gave them an option to disable them. Was it really a big deal? When it comes to skill checks, the top request is to show the numeric values of the checks. If we can make it optional, it would be great. It won't take anything away from people who don't want to see the tags or the values, but make the game more enjoyable for people who want this info displayed.
How does giving an option to hide the skill tags and show skill-check values become a story mode?
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
It all depends on what 70% THC means.
Well, it always means that your chance to hit is 70%. Everything else is lying.
Not that simple. 70% means that if you play long enough eventually you'll hit 70% of the time. The question is how long. 100 attacks? 1000? If it takes a 1000 attacks to achieve that 70% probability, then missing even 100 times in a row is perfectly fine. Bad rolls, what are you gonna do? Needless to say, if a player misses even 20 times in a row with a 70% THC, he will rage quit and will never ever believe that the system worked as advertised (i.e. 70% THC).

We don't fudge rolls (and it's not news either, I explained how our RNG works many times before), we merely reduce the size of the sample to achieve a more or less even distribution of hits and misses. And still way too many people, including those posting on this prestigious site, bitch about missing 3 times in a row, holding that as the ultimate proof that the system is broken.

Vault Dweller having rolls being a shuffled deck does sound odd and encourages counting.

What exactly are you looking for when you ask for solutions to your problem? Are you willing to go away from shuffled deck approach or it is staying and you can at most iterate?
We're willing to consider any alternative that works and doesn't make things worse. Our main goal isn't to help or hinder the player but represent the to-hit chance probability in a consistent and reasonable way. If my THC is 50% I want to hit half of the time, preferably not long term over the course of the entire game but over the course of a fight.

You didnt answer to pathofexile approach of dynamically changing odds based on past failures/successes.
It's a very different direction, more suitable to a diablo-like than a slow-paced TB RPG.

While we are on topic of deck shuffling: is it global? When is it reshuffled?
Before a fight and when you run out of 'cards'.
 

Grampy_Bone

Arcane
Joined
Jan 25, 2016
Messages
3,667
Location
Wandering the world randomly in search of maps
It all depends on what 70% THC means.

What THC means to players is, "what is the chance my turn will have a productive outcome?" When the odds are over a certain threshold (51%+), the player feels it is worth a gamble. When the player gambles three times and loses, it feels like they've been "cheated" because their turn was wasted. They would have spent the turn more productively.

The solution is to make failed shots still productive in some way. Examples:

-Expeditions Vikings adds the "harried" status to targets who've been shot at. I don't even know what it does, but it feels like something.

-I can't remember what game specifically but I know I've seen systems where missed shots make the next shot more accurate ala "zeroing in"

-Battletech removes Evasion pips from targets for each shot.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
Vault Dweller What does "more accurately" mean, anyway?
Means meeting reasonable expectations in regard to your THC. If my THC is 70% I expect to hit more often than not, say 6-7 out of 10. I don't expect to miss 5 times in a row even though it's more than possible.

i guess important question is... how many actions do you expect to be made by each combatant over course of 1 battle?
Too early to say as it's a question of balance. We don't have any particular target in mind at the moment.
 

MRY

Wormwood Studios
Developer
Joined
Aug 15, 2012
Messages
5,716
Location
California
As the Wormtongue of decline, I would normally endorse something along these lines, but too many of my favorite game experiences have been the result of strings of good or bad luck. (The first, and most formative, being when a single archer defeated a full squad of archers when defending a fort in The Ancient Art of War. It was the first "emergent narrative" I encountered as a wee lad of 6, and it stayed with me. Obviously, the point-blank spray fire miss on a Chryssalid in X-Com, which then obliterated my squad through a series of mishaps, is another classic.)

Anyway, it is definitely true that people get sad when a series of bad rolls loses them a match. But I think you can mitigate this in various ways. The simplest would be to have bad-luck achievements, and have the highest of these achievements set relatively high thresholds ("Missed 100 shots with >80% likelihood of hitting." "Lost 20 battles in which you missed at least 3 shots with >80% likelihood of hitting."). Because there is significant overlap between "person who gets mad at missing and posts a negative review on Steam" and "achievement harvester," I suspect that if you calibrate these right, you will be able to eliminate a lot of the hard feelings. I realize this kind of stuff enrages the Blakemoreland Hybrid Boss folks of the world, but I really think it's the best approach. (Just as I still believe that allowing players to autoresolve combat, a la Master of Orion 2 or Realms of Arkania, is a reasonable accommodation.)

Another, gamier solution would be to have outrageous misses accumulate some kind of "golden die" option (a la Frayed Knights) where you can get a critical hit if you've missed X number of times. This feels possible mood breaking, but I like it more than your proposed solution because rather than just smoothing things out, it makes the eccentricities of random chance fun in both directions.

Grampy_Bone's examples seem okay as well, but they may not really work. For instance, missing a Harried target when you have 90% chance to hit will just amplify the annoyance. Turning evasion into a shield generator (which is what I gather evasion pips basically would be) would significantly change combat, possibly for the worse.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
How does it work with switching between different CTH attacks? Are those separate 'decks'? Say, Fast attack at 73% misses, did you just improve your Power Attack and by how much?
In AoD we merely draw numbers without making any streak adjustments. We don't reset the deck when you change attacks.

In The New World it all depends on whatever parameters we'll set. Let's say, if your THC is less than 50%, you can miss as much as you want. If it's 50-64%, you can't miss more than 3 times in a row (i.e. after the third miss you automatically draw a hit card, but it's not a freebie as this card is now removed from the deck). If it's 65-80, you can't miss more than twice, etc. These are random ranges for discussion purposes only.
 

Vault Dweller

Commissar, Red Star Studio
Developer
Joined
Jan 7, 2003
Messages
28,024
Now I dont understand anything.
First you say that the rolls in AOD are adjusted ( "which is how it works in AoD" ). Now you say that AOD does not make any streak adjustments.
We didn't adjust the streaks but we used the deck of card design (i.e. can roll the same number twice until the deck is reshuffled), so it's not True Randomness either.

MY MAIN POINT
Dont lie in the interface. If its says 70%, I fucking expect 70%.
And that is precisely what I want to. If my THC is 70% I want to know what to expect. If I have to make a hundred attacks to get that 70% chance, it might as well be 30 or 50%. These numbers would be equally meaningless in a short fight. I want the player to know that if his THC is 70%, he can expect to hit 7 out of 10. The only question is whether to allow long missing streaks, both for the player AND the enemies.
 

deuxhero

Arcane
Joined
Jul 30, 2007
Messages
11,381
Location
Flowery Land
Why not make it a character trait (Focused: You're really good at things you're good at but really bad at things you're bad at. Chances shown at 50% or higher are even more likely to occur for you, but chances 49% or lower are even less likely.) or options tick box? Shouldn't be too hard to have an option "Use two random numbers" with a tooltip that says "When a random number would be generated, roll twice and average the result. Improves likelihood of succeeding at 50% chance or higher, but increases chance of failure at 49% chance or lower .". It's a fairly simple way to "rig" an RNG that a few games have used (the most documented being a few of the Fire Emblem games.). Math wise it means likelyhoods 50% above increase exponentially (55% is still less than 60%, but 75% is actually 87.75%), but those under similarly fail, which rewards high investment and punishes going for low chances which designwise serves an actual purpose compared to just making sure 20 d20s result in different numbers every time.
 

MRY

Wormwood Studios
Developer
Joined
Aug 15, 2012
Messages
5,716
Location
California
As per my prior post, I really think that "I want to know what to expect" is really the wrong route here. The whole reason to have an RNG is permit the unexpected -- there are at least three things it provides (the ability to gamble on a high-risk, high-reward option when it's the only hope you've got; the thrill of an unexpectedly good result [e.g., enemy missing you with a point-blank burst, critically hitting and killing your foe]; the panic of a well-laid plan going awry due to chance). It's true that the frustration of missing your own point-blank burst is extraordinary, but the idea that the frustration stems from an honest belief that the game is cheating seems implausible to me. Instead, I think "CHEATER!" is simply an emotional reaction to the fact that something bad unexpectedly happens to the player for no good reason -- e.g., the player took a conservative, low-risk, low-reward option to finish an opponent off and still missed. As long as there is any random chance, there is a chance of such a thing happening, and the player's reaction will be resultingly foul if the consequences are high enough.

The only ways to address this are: (1) remove bad RNG results for player (yikes); (2) remove RNG entirely (losing the upsides described above); (3) give the player something when he has a bad roll. As my prior posts indicates, I think #3 is the way to go. But what to give? IMO, an achievement is the simplest answer because it's totally extrinsic and can't break anything. Because TNW is a serious setting, I wouldn't recommend in-lore options (e.g., early on you meet some guy from Lady Luck's Jilted Swains who mentions that his elite club's benefactor makes a point of letting in those cursed with a string of unfortunate events, and then in the Habitat you find the club and can get in if you have enough unlucky die rolls), but you could do some kind of extrinsic thing (like the golden die option). But I think having it "outside" the RNG is the key. Trying to adjust the RNG itself just feels weird. It's better to have the game acknowledge you got screwed and offer some offset.
 
Joined
Jan 18, 2018
Messages
1,301
Grab the Codex by the pussy
The only ways to address this
is to acknowledge that this is a pointless discussion because a player that has this frustration doesn’t have a clue what ITS games are about. You either streamline the whole game, or forget about it.
 

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