How do you stop a randomized game from randomly being boring sometimes?
arstechnica.com
arstechnica.com
From that perspective, the definition of a game and what constitutes gaming changes dramatically. It undercuts the question here. Preventing a game from being randomly boring is an odd question. A better question might be, how can we keep the player's decision-making interesting, regardless of what is generated? The article says it directly, but misses the golden quote which I think is essential. That quote offers up some really unique ways to define a game too. Very cool.
Another option for a similar effect is to keep a counter and compare it to sequences of prime numbers: when the count is a multiple of 23 then X happens, if the count is a multiple of 7 and 11 then Y happens, and so forth.
Balanced with enough thought this will ensure that enough happens without it seeming too regular to the player.
Various games (L4D is the first prominent example, there may be others before it) include "AI directors" that try to keep the player involved through balancing exiting events and periods of tension. This makes it a lot harder to have a "boring" game, or one where you get "unfairly" hammered by the randomness.
In fiction I'd trace this to Asimov's Foundation series, which is premised on "psychohistory" being a sort of fully deterministic fusion of economics and psychology enabling the development of future society to be predicted. Until a few books into the series where the "mule" is introduced, who is an individual powerful and unpredictable enough to throw off the determinism of historical inevitability.
Full disclosure: I work as a Software Engineer for Ghost Games and was involved in the development of Need For Speed (2015). I am not an official spokesperson for Ghost or EA and I am not authorized to speak on their behalf.
In Need For Speed, AI racers will slow down when you're behind and speed up when you're ahead. The idea was to always make the player feel like they were having a fun challenge, regardless of skill level.
In the end, the AI rubber banding was such a strong effect that, when the game first came out, it was one of the main complaints. We ended up fixing it in the very first patch.
IGN reporting on it: http://www.ign.com/articles/2015/11/23/need-for-speeds-most-... Our announcement which first mentioned it: http://www.needforspeed.com/en_AU/news/future-plans
I actually tried a bit to look back and see it happen beacuse, I was wondering, do they just teleport or actually drive at like 800km/h.
Many games do, actually, and it's an interesting topic in of itself. In case someone wants some further reading:
- http://www.redblobgames.com/articles/probability/damage-roll...
- http://sinisterdesign.net/unpredictability-and-control-in-tu...
- http://www.gamasutra.com/blogs/RogerHgensen/20101123/88511/C...
- http://techland.time.com/2012/10/12/why-xcoms-random-results...
- http://forum.zdoom.org/viewtopic.php?t=50643
Another funny anecdote is that Wesnoth had to include a FAQ entry about the game being unfair because it relies too much on "luck": http://wiki.wesnoth.org/FAQ#There.27s_too_much_luck_in_this_...
XCOM 2 had a few interesting implementation details. You can actually see it all in the source code provided in their modding tools, which is a really awesome thing IMO, to be able to see a huge chunk of XCOM 2's actual code.
On the two lower difficulties (equivalent to "easy" and "normal") there is a relatively minor, invisible bonus to your next shot's chance to hit if you have a chain of misses. On the "hard" difficulty, this correction factor actually increases to compensate for several other changes that make the game much less forgiving of tactical errors. Then when you get to the "legendary" difficulty, the bonus goes away entirely and you're fighting a completely fair RNG along with all of its streakiness. I think this puts an interesting perspective on how "RNG fairness" works.
The story behind TFTD's difficultly (you may have heard this one) is that people complained that XCOM was too easy, even on the hardest difficulty. But this was due to a bug: save/load did not persist the difficulty level properly, so most people played one or two missions on ultra-hard and the rest on easy, without ever realising. So when the developers turned up the difficulty and fixed the bug, suddenly the sequel was incredibly hard.
That’s amazing—a logical fallacy as a feature!
In games of chance on the other hand, people understand that you can work at making better bets but you can't make yourself "luckier" by force of will. So it's easier for people to blame the RNG gods than berate one another for lack of skill.
Edit: remember in game design, especially human vs. AI, it's not about beating the player but about maximising their enjoyment of the process. You have to make the defeats fun as well.
If you're rolling a fair dice, if you cap at 12 (i.e. the 12th roll will land a 6 no matter what) then your expected time to roll a 6 goes down from 6 to 5.4.
So by applying a "correction" on strings of misses without applying a similar correction to a string of hits it skews the mean down which actually changes what your average hit-rate is. It would be interesting however if the tested hit-rate was adjusted lower to account for this effect, that would make it "fair" again by the hit-rate stat reporting the average hit-rate, but would make the randomness generate "appear more random" to humans.
Technically you would see the effect of your shot (i.e. too much to the left) and adjust your aim (more to the right).
the mistaken belief that, if something happens more frequently than normal during some period, it will happen less frequently in the future, or that, if something happens less frequently than normal during some period, it will happen more frequently in the future
Along with "the further behind you are, the more likely to get a blue shell".
Though I think that the reason this feels necessary, isn't that people are irrational and believe in streaks - it's that they used a poor random model in the first place. Maybe a spearman should simply never beat a battleship.
Do you have a source for this? I'm not doubting you, but I've played copious amounts of XCOM2 (until it broke on NixOS..) and didn't notice. But now that I think about it, the hit chances are less predictable than in the prior games.
--- note: seems i remembered this wrong, and i can't find the source for this anymore.
iirc, they fixed this by creating a "random queue", i.e. a fixed size queue (say, twice the average) with the amount of needed drops randomly distributed. this way the players bad luck streak would have an upper bound.
--- note: what they actually did was increase the drop chance progressively, but it might be that this was only for single item drops?
Linear Regression of Progressive Percentage in World of Warcraft
This is because human brain is Terrible at statistics. Many games curate random events into Normal distribution, otherwise you get complaints about all those missed 99% accuracy shots.
I have experienced more incredible stories unfolding in both MoO2 and MtG than I could possibly remember. Both are games where careful strategy and tactics combined with randomness can lead to some unbelievably epic moments. Here's the thing, though: in MoO2, despite a lot of random events, I can't remember a single time I played a session which wasn't exciting or didn't feel fair. In Magic, this happens routinely. Just Google "mana screw" or "mana flood" for more results than you can read, complaining about randomness leading to boring matches where you, as someone put it, "don't get to play Magic"[1].
I still play Magic on a daily basis, and I still love the game, but I've spent a lot of time damning the frequency of these boring matches and wondering if there was a simple strategy or rule change which would fix this, while still leaving the soul of the game intact.
[1]: https://www.reddit.com/r/magicTCG/comments/4c2f65/what_funda...
The basic rules resemble Hearthstone, but rather than building a deck beforehand, you just pick one of seven 10-card starter decks, and 3 of 20 "codices", and then at the end of each turn you take two cards from your codices and add them into your deck. So there's still luck of the draw, but not to the extent of having to shuffle a 30 or 60 card deck beforehand. (Also you go through your cards Dominion-style rather than one-card-per-turn Magic style, but I'm honestly not sure as to whether that counts as an increase or decrease in luck!)
I've heard this compared to Mage Wars, but looking that up, it looks like that just lets you cast spells straight out of your spellbook, no deck at all. Is that right? So I guess Codex is another intermediate step there.
(As regards "mana", it doesn't follow Hearthstone, but rather takes the fairly-common approach where any card from your hand can also be turned into a mana producer (this permanently removes them from your deck, so you can remove cards as well as adding them).)
Sounds very similar to Dominion et al then - just slightly hybridized. I take against Sirlin based on his writings and his emphasis on reading your opponent which isn't a mechanic I find fun; I enjoy deckbuilders but there are plenty of them out there.
> I've heard this compared to Mage Wars, but looking that up, it looks like that just lets you cast spells straight out of your spellbook, no deck at all. Is that right?
Yeah - each turn you choose two cards from your spellbook and maybe play them.
> I take against Sirlin based on his writings and his emphasis on reading your opponent which isn't a mechanic I find fun
Huh, I'm surprised. There's a lot of dislike of Sirlin out there obviously but I haven't seen many say it's his writings they don't like. I think they're some of the best writing on game design I've seen! To the point that I'd basically count myself as a "Sirlin fan" even though I've only played two of his games of which one is pretty mediocre. :P (Namely, Flash Duel. Kongai, though, was great, back when people actually played it...) I don't think there's anything too advanced in most of what he writes -- a lot of emphasis on such notions as even playing field, cutting out extra steps that get in the way of playing the real game, the focus on contested skill rather than unconstested skill, that yes strong competitive balance is achievable -- but these are basics that all too often go neglected, IMO, so it's good to see someone explicitly talking about them. There seems to be this philosophy out there of "Well, we don't really need an even playfield..." or "C'mon, nobody can actually make a balanced competitive game with that many characters" and I'm glad to see somebody countering that. Whether he's good at it or not, Sirlin is, to my mind, somebody who gets what game design should be. Would you mind elaborating on what about his writing you don't like?
Writing... I find him pretentious/self-important/self-impressed, and he doesn't seem to acknowledge the validity of other points of view.
Another interesting idea was a format which someone suggested[1] in which any non-land card can be played as if it were a land producing colorless mana and any land has cycling 1 (for 1 mana you can discard it and draw a new card). I found this a fascinating idea, but I feel like this would completely wreck the balance of the game, especially when played with decks specifically built to abuse these new rules.
[1]: http://www.mtgsalvation.com/forums/magic-fundamentals/magic-...
Top-decking is still very real in both games, sometimes you just need that 1 card to win or you'll lose next turn.
In their search for 'dramatic' cards, they've started introducing "triggers on draw" effects, "get a random creature from your deck" effects (maybe it'll cost 2 mana, maybe 10!), and my favorite - "play this card, then shuffle it back, then play it again, then all your creatures become different ones".
The card designers have consistently chosen exciting events over stable gameplay, to the point where it's far more random than MTG.
Also, if you're still seeing top decking in Hearthstone, you haven't progressed enough in the ranks to see the long-term players. There is rarely any top decking going on past even rank 15.
I don't think the core, high-ladder gameplay is broken, but I think a lot of their design efforts are badly counterproductive. MTG's long-term survival can largely be credited to fighting power creep and strong-but-random effects, and I think Hearthstone's design looks more like a shorter-term cash in on those effects.
There are still RNGesus decks in the lower rungs of the ladder - I play one. But above 15 or so and most decks are using predictable cards with few random effects.
Arena was definitely my preferred playstyle - it had a lot of variance, and I never felt punished for not having all the right legendaries. On the main ladder I felt like I was hitting a ceiling where I played lots of netdecks I couldn't "afford", and just lost interest. Maybe it's worth another look!
The cards he's talking about trigger when they're drawn, not when another card is drawn while they're in play.
For example, Flame Leviathan[1] which deals 2 damage to all minions when drawn. Or Iron Juggernaut [2], which shuffles a 'mine' into the opponents deck, dealing 10 damage when drawn (personally I think this one's ok, as at least you can plan to mitigate it most of the time)
[1] http://www.hearthpwn.com/cards/12290-flame-leviathan
[2] http://www.hearthpwn.com/cards/12295-iron-juggernautFlame Leviathan annoys me a lot, though - it's a complete board rebalance that neither player can anticipate. It's not even a good card (I wouldn't expect to see it high on the ladder) but it's aggravating to have such high variance on a zero-mana outcome.
The solution is to tune the game and add additional mid-game content (the part of the game with the most room for dynamism), which the author points out is on its way.
One of the more interesting comments said that the game would be well served if its developers collected play through data and started prioritizing the game seeds that tended to cause the most interesting games, arguing that there would be enough seeds to hide the lack of randomness.
Where did he say that? People seem to keep mistaking reviews for impossibly objective analyses of games.
[0] http://www.ign.com/articles/2016/05/09/stellaris-review [1] http://www.twitlonger.com/show/n_1solvon
Neither page you link to contains the word "objective". The only mention of the words "bad" is in the phrase "bad luck", in a sentence specifically talking about how other people might have a better experience and Rowan might have been just unlucky. There is a score, but to interpret that you'd look at the review guidelines for IGN. To quote: Under no circumstances are review scores influenced by anything other than _our own opinions_ on the quality of the product in question. (Emphasis mine).
What would help is, instead of prioritizing seeds, looking at optimal distributions of random events that lead to fun gameplay. This might mean for instance making your next random event dependent on previous ones (e.g. to compensate previous bad events with more fortunate ones, and vice versa).
An example of this is that perhaps numbers that are even produce more interesting games than those that are odd, so they could then write something like the following:
int new_seed(void)
{
int seed = random();
if ((seed % 2) != 0)
seed++;
return seed;
}
It does seem to suffer from a lack of gameplay options. I think a useful comparison is Civilization V at launch versus after its 'Gods and Kings' and 'Brave New World' expansions, both of which added many new gameplay avenues.
The game seems to be more organic than strategic, lacking crossroads where your decision is based on your priorities. The major exception is the research system, where the 'choose one of three' feature means selecting one option closes off the other two possibilities for an indeterminate period. But besides that, in my experience the game is about expanding like a slime mold and gradually wearing down your opposition (as conquest is the only viable victory type).
Have to say that I love the research system though. It completely turns a lot of standard 4x/strategy style gameplay on its head by preventing 100% forward planning.
It reminds me of Master of Orion 2, which also had non-deterministic research. There, each point in the tech tree was a category containing multiple items and, when research into that category was done, you got a random item from that category. If you wanted one of the others, you'd have to get it by trading with or spying on other races.
After a while of playing those games, I tend to get a little bored as I'm pretty much following the same strategy over and over. You get the feeling that you're just ticking off the boxes on your perfect playthrough and when I have that realisation, I get immensely bored with the game.
By adding some randomness and preventing the player from ever being able to follow a perfect plan, they've actually made a very interesting highly replayable game. And like all Paradox games, it'll only go from strength to strength as they patch it and release new content for years to come.
But Master of Orion 1 did have a somewhat randomized tech research series.
Uncreative races got a random one in the category they chose and Creative races got all items in the category they chose. It was far too simple to play as a creative race, such as Psilon.
Basically, people at large don't grok statistics at all. Instead they think of randomness in terms of permutations (drawing cards from a deck). When you pit them against "well-designed" statistical randomness, you are rewarded with a cacophony of complaints about how the simulation is cheating, rigged and unrealistic. Sid Meier had a great GDC keynote about his multi-decade struggle to convey basic concepts like "Something that has only a 1% chance of happening will still happen sometimes." This gets worse when you realize that a million daily active users means one-in-a-million events are a many-times-daily occurrence. That means every day someone will complain that "I rolled a one out of ten SIX TIMES IN A ROW!!! That should never happen! You are explicitly cheating me! I demand my money back!"
So, yeah. Don't rand() in games. Do the shuffle and everyone will be much happier.
It turns out you can have unwinnable boards if you build them purely randomly.
So I ended up having the app play the board in reverse to make sure it was winnable!
It still managed to hit stalemates, while playing backwards, but they were very rare. I just restarted instead of trying to avoid that situation algorithmicly.
"Why is God punishing us?"
It's grand strategy, about managing empires at a high level, whereas a 4X game tends to be more about resource management, research, and building spaceships.
I think those three are represented as well in Stellaris. As mentioned above research is done very well, and presents often dilemmas when you want to research multiple things and you don't know when a certain option will be available again.
Ships can be designed by the player and offer a lot of customisability. Resources are abstracted behind three basic resources so is probably not as deep as other games, but it is in.
I really like the randomness of Stellaris, not only of the research options but also the randomly generated galaxy and AI empires, which means you never encounter the same systems or competitors (unless, of course, you start in the Sol system or let certain empires spawn, which can be chosen at game start).
The article seems to be of someone who doesn't really like the game. Arguably it isn't as fleshed out as other 4X games, but to me it doesn't seem fair to blame it on the random generated content.
Here is a game reviewer pointing out that one particular game (Sunless Sea) appears to be somewhat improved by having stretches of monotony interrupted by more interesting gameplay.
Someone toyed with this idea by letting the player of the game play against animals: https://www.youtube.com/watch?v=6DXiuzAQztU
Freecell is a good example, some games are trivially easy or too hard (not sure if there are impossible games)
Consider: some initial layouts can be solved without using all four free cells; others can't be solved without five or more cells. If a Freecell game implements a solver that can find the minimum number of cells needed to solve a given layout, then it could easily tune the difficulty - by only giving players layouts that require certain numbers of cells, or conversely by varying the number of free cells depending on the layout.
one thing i've been interested in was seeing how my solvers results (i.e. # of steps, visited boards, time to solve) correlated with perceived difficulty for a human player - i didn't go that far though.
Instead of starting with your random number generator, build up the discrete probability distribution function first. The histogram can have any shape you want.
So, for example, in Tetris, rather than choosing a uniformly random next piece, we give a 10% probability to a straight one, 20% to right-sided L, etc. Now we begin to see patterns that can make gameplay challenging or interesting. If a shape has not appeared in a while purely out of chance, alter the distribution. Keeping track of what cards have been played and what moves have been made forms the basis of a feedback loop that constantly evolves.
http://tetris.wikia.com/wiki/Random_Generator
Officially branded Tetris games are required to use a "random bag" algorithm where they generate random permutations of the 7 tetrominos, so that for every 7 pieces (aligned with the bag start) you get each type once. You can only get two in a row if they appear at the end of one bag and the start of the next, and conversely there's a maximum of 12 pieces between duplicates (start of one bag, end of the next). Three in a row will never happen.