Hacker News Reader

It's worth noting that at least in recent versions, the in game formulas have been getting simplified. For instance armor used to give you both AC and GDR (guaranteed damage resistance), whereas now GDR is a direct function of AC regardless of its source. Similarly DEX now more directly influences Evasion, whereas before it was artificially capped to give very diminishing returns past ~25 DEX, and now allows you to get very high Evasion.

That's good to hear. I hope they keep going down that path. To refer to another game, in any of the Soulsborne games (Dark Souls, Bloodborne, ...) by From Software, as you level up you get to contribute points towards stats which affects your ability to wield weapons and how much damage they do. You don't even have to equip them, the game gives you a diff of what your new damage will be along various dimensions (physical, magic, etc) as well as how your "equip load" is affected just by inspecting it. Although this doesn't necessarily mean you can understand the difference between a really heavy, highly damaging weapon vs a fast weapon, by reading a wiki or by playing through once or twice you'll have a very good base understanding of what different weapons bring to the table and how your character drives those strengths/weaknesses.

With DCSS? Well... how much damage you do is going to depend on whether or not you're wearing a shield (what size?), what kind of armor you're wearing if any, how much you've trained for that weapon class, how much STR you have, what bonuses it has, and what bonuses you have from gear. Then RNG will play a significant role on top of that. Maybe RNG applies twice, once for your swing, once for their defense? It's been too long, I can't say confidently.

I realize it's challenging for the interface design because everything is already tight but I'm sure there's a solution. If I equip this armor/weapon/jewelry, how fast will I be attacking compared to now? What will my damage distribution look like before enemy resistances? Same with allocating stats, please give feedback in the form of derived data that is closer to the thing I care about (surviving).

I don't think the problem is complexity as much as it is opaqueness. I have played other games that are also very complex but yet still very enjoyable because the complexity is open to be examined.

The opacity is a remnant from the game as it was designed twenty years ago, which took cues from roguelikes of the day such as NetHack (which is famous for its opacity). The developers of Stone Soup have spent a few decades improving the interface with the goal that you shouldn't need any external wikis or resources to win: all info should be in the game itself. For example, a relatively recent-ish change is that you can inspect enemies and see a list of all possible spells that they can cast, along with estimates of the damage that each spell will do. A much older feature is the in-game help menu, which is impressively thorough and will give you a high-level overview of every monster, spell, stat, ability, item, god, level, status effect, and so on (although it's not necessarily super-detailed when it comes to obscure things like calculating damage; simplifying these calculations is the only way to go).

That actually sounds nice. Compared to most other "competitive” games, where there is an optimal meta way of playing it, and doing anything else is basically trolling and instant loss

Isn't this what this is though? There are certain builds where if you follow them to a T, then the game is "easy" and you win almost always. If you just experiment then yeah you can win by some crazy stroke of luck, but most likely you will just have a bad time.

I had the same experience in Caves of Qud(which is great btw) - there are some builds that really really really work, and others which are unwinnable.

There’s a great YouTuber named Ultraviolent4 who plays lots of “unwinnable” or maybe just really odd combinations, and watching his games really improved my own.

No not really. Very few builds will let you win the game easily. Situational awareness is required. Relatively few enemies in the game are just melee brutes. Most of them occupy some niche that requires you to think. They can confuse your character, or lock all the doors, or summon, or irresistibly cut your health in half, etc. An exceptionally fortunate character, say, a kung fu octopode transmutated into statue form and wearing 8 magical rings of power can probably karate chop their way through 99.9% of the game without much of a worry, but may still lose in a 1:1 slug fest with certain lords of hell.

There are easier builds and there are harder builds, but if you aren't good, you will lose. Case in point, the bug described here effectively doubled damage output and win rates were still pretty low.

Although it's worth noting that there are different degrees of "winning the game". Only three runes are required to win, but maximizing your score and prestige will require you to get all fifteen runes. If you're just going for a three-rune victory, then there are more-or-less "safe" builds. For example, a Minotaur of any melee class, or a Berserker of any species that isn't completely inept at melee can reliably win a three-rune (not a guaranteed victory obviously, you still need to know what you're doing, and most importantly when to run away), although these characters are going to have a very hard time in the extended endgame without some sort of pivot.

In some sense, yes, but in another, not so much. Crawl is quite difficult and you're always a few steps away from death if you don't react to different situations appropriately. It can take players months to years to get their first win. Some people are never able to beat it before giving up on it. There's still an optimal way to play in most situations, but many couldn't tell you what that optimum is beyond some rules of thumb which aren't always right.

Don't get me wrong, there's a lot of gold in those hills and I had a lot of fun getting better and getting my first win. But after that I was very frustrated with how hard it was to just stop at a fork in the road and try to answer, which choice is objectively better? I like difficult games that are fair and require skill growth to succeed in, but I want to be able to understand them so I can make informed choices as I get better as well. Too much: because that's what the black box determined. sorry, you died. should have been more cautious.

Sounds like the game isn't made for you then.

This would be like attempting to play Go, and complaining that it's nothing like Chess because the optimal move generally isn't obvious even after 10,000 hours of play. And then loosely tying it into "part of the joy of being human" instead of simply stating that "this is my preference".

Yes, there's an interesting distinction here. Although I'd say it's more like Chess than Go, in the sense the mechanics are really simple in Go, while Chess has specific rules for each piece and you should memorize openings and quirks to get an edge (Go is much more about generalization).

In reality, both Chess and Go and all board games are fairly simple, you can learn all the rules in a few sittings and then just focus on strategy.

In roguelikes, often learning the system is the point (or a point) of the game. The fun, at least to me, is in dealing with the unknown and discovering how it works and what works.

Nethack (a classic roguelike) takes it to the extreme -- a bit too much in fact, because an essential game mechanic is an obscure passage you'd have to deduct (Elbereth). But the fun is there: you grab a potion and have no idea what happens if you drink it -- maybe you will be killed instantly, maybe it will make you into a god -- it's not just "which of those effects am I going to get?". You have to rely on previous knowledge, as well as expectations from similar games and what you know about the game.

But isn't that quite common IRL? Like, okay figuring out what you should be eating has a science to back it (albeit quite extensive): you read the most effective diets, what risks it reduces, you health, your weight and need to make a decision. Already not very simple strategic decision. And then you need to take into account satisfying your tastes, how often diet fads reverse (is eating butter healthy? go figure), how to do it when sharing a meal with other people, and so on. The unknowns are as significant as skillful strategic planning. The same could be said to running a company: you need to manage uncertainties in a vast decision space, with opportunities like calling for partnerships, merging, buying other companies, spending on uncertain research, and much more. You slowly acquire domain-specific knowledge (with help from the vast literature, plus a few insights of yours) and use that to in turn reduce and manage uncertainties and plan ahead (here your model of other players and capability to find the 'best move' -- least risky, greatest payoff, more robust -- will shine).

Re your first paragraph, sure, and I'm fine with that. But the following analogy is quite off.

Go read their manifesto: https://github.com/crawl/crawl/blob/master/crawl-ref/docs/cr...

Specifically the section titled "Clarity".

> Things ought to work in an intuitive way.

Spoiler alert, they most certainly do not. There's nothing intuitive about being completely unable of knowing the practical effects of something as simple as equipping a piece of armor or a shield or allocating a point to STR vs DEX. I gave an example in another comment of how other games do provide information when comparing between options so you can intuit their effects. I have faith they'll get there, though. I believe they'll gain many more players if they can, too. :) I'm sure I'll be back in a year or two after they improve the gameplay more.

The term common in MMOs is theorycrafting, and it gained particular momentum in an era where some common games allowed for addons that could export telemetry data. Contrived or common scenarios could be mined (Monte Carlo simulations if you will) to determine how patterns lined up with theories.