What is your opinion on challenge in gameplay? What makes a game challenging?

[somerandomdude]

I'm kind of exited to play Wo Long, because it looks like another game that's going to force me to adapt and change the way I approach it sort of in the same way that Bloodborne did. I welcome this type of challenge. I don't want to go into another one of these games and just totally fucking crush it based on previous game knowledge/experience in the genre. I was able to pretty much crush Elden Ring, no boss took more than 3 or 4 attempts, and most dropped on the first attempt.

[Tweed]

I was half-joking. KotC 2 doesn't get that part right at all, but it does reload quickly. Although it does make me ask how often should you be saving and reloading before the challenge in a game is flawed. In KotC 2 you're pretty much required to give all of your characters improved initiative and KotC 2 isn't the only RPG with the "go first or die" or "one round everyone down" problem.

It's not really about the quantity of reloading so much as the reason for it. If a fight is simply hard, then it's fine if you need to retry a lot. On the other hand, if it's "hard" because RNG can totally throw you under the bus on a single roll, that's not good at all. It's fine if there's a remote chance that freak RNG can screw you over the course of multiple rolls, but if a single roll routinely makes-or-breaks a fight, that's bad design.

There's still something about the gameplay that makes KotC 2 enticing where I would have dumped other games. It felt good to overcome the difficulty. The reworked Augury of Chaos module is a bit better and affords people more options for meeting challenges though the module itself is basically a mess. Hopefully Pierre's later stuff will be more concise.

[Cedric]

KotC 2 has some idiotic, never-tested encounters and the game would benefit highly from solving the 'initiative problem.' See, while the game is very hard indeed, the main problem are initiative rolls. To even stand a chance at all it is imperative that your casters go first. It is almost as if the encounters, if they have been tested at all - which I doubt - were designed to be played with take 20 initiative.

There's no 'rolling with the punches' in KotC 2 because the enemy mages have such high DCs on their spells, and the melees such high ABs that you are not going to survive if they all or mostly all go first - not ever, at least not without turning the difficulty down. But when I played there was only one difficulty.

I am somewhat undecided whether this is a problem with KotC 2 specifically or the initiative system in general. I very often think that if you end up just reloading until you roll well on it, then it might as well fix your initiatives at 20 rolls. In pen and paper, init. is exciting, but in a game designed to test and punish you, it is a needless RNG factor.

To any "well, just build your characters to have high initiative then, dingus" midwits: in KotC 2, not only did I in fact have all my casters have the maximum Dex score, the initiative familiar and the Improved Initiative feat and gear, but also Pierre, in his wisdom, gave the enemy wizards something like 20 Dex, +10 Initiative boots and +6 Dex gear later on, in addition to the initiative familiar & Improved Initiative feats. You are not going to casually 'build for initiative' in that game. As a matter of fact, unless you minmax your own initiative hard, you will probably never go first. Which lategame just means your party is vaporized instantly by mages.

---

What is good difficulty? In RPGs, I suppose that is hard to say, due to the RNG of dice rolls. I suppose a good difficulty in an RPG would be "challenging, but account for bad luck in some way." I am not sure how you would do that, which is why I want to talk about action games instead. The very best action games - and in fact, maybe even best games in general - all have what I would describe as 'perfect' difficulty. What is that?

Perfect difficulty is when, in theory, given some IQ 300 human with super-human (or at least, peak human) reaction time, it would be possible to beat the game without getting hit on your very first try. As in, not a single instance of RNG screwing you in ways you cannot recover from, as well as 100% fair design. Such games exist. They are, in fact, a staple of the cream of the crop of Japanese action games. Example games like this are almost any Bullet Hell game (all by CAVE, but Touhou also qualifies, as does most of the Namco stuff) as well as the early Clover/Platinum/Kamiya 3D action games (that would be God Hand, Devil May Cry 3, Bayonetta, etc.)

These games feel next to impossible due to their intense difficulty level - and yet they are never once "unfair." Every single time you get hit it is 100% on you, and never on RNG, or bad bullet patterns. It is always on you.

That sort of challenge makes me the happiest.


That said there's nothing wrong with the autistic reload challenge of KotC 2. Despite that RNG idiocy in that game, it still feels satisfying. And I believe that's because, to even get to the point where you're experiencing the reloadfest, you already need to have pretty much perfect mastery of the systems and mechanics of the game, which Tweed has.

[WhiteShark]

What is good difficulty? In RPGs, I suppose that is hard to say, due to the RNG of dice rolls. I suppose a good difficulty in an RPG would be "challenging, but account for bad luck in some way." I am not sure how you would do that, which is why I want to talk about action games instead.

There are a few things you can do. You can make the RNG follow a bell curve distribution so as to lessen the swinginess of the system. You can design the system such that there are many rolls each with a small effect instead of one big roll that determines the whole thing. You can give the player a lot of tools to try and stack the deck in advance, such as by allowing the player to draw out enemies one by one, choose where to fight, pre-buff, etc.

Freak bad luck will always be possible in a non-deterministic system but you can mitigate it to a very high degree. The most refined roguelikes, for example, have a huge amount of RNG, and yet are designed such that the top players almost never run into a truly unwinnable situation.

That said there's nothing wrong with the autistic reload challenge of KotC 2. Despite that RNG idiocy in that game, it still feels satisfying. And I believe that's because, to even get to the point where you're experiencing the reloadfest, you already need to have pretty much perfect mastery of the systems and mechanics of the game, which Tweed has.

It just sounds like the game is good enough that you enjoy it despite that, not that there's nothing wrong with it.

[rusty_shackleford]

RPGs are(should be) similar to poker in that good players can mitigate bad luck. If you feel like you're just playing a dice-rolling simulator at the whim of where the RNG falls, there's a chance you're playing an RPG that failed at this.
Deterministic "RPGs"/tacticool games are just puzzle games. You can see this in a game like e.g., Druidstone: The Secret of the Menhir Forest, which claims to be an RPG. Each scene is just a puzzle, you aren't reacting to anything, and there's an optimal way to solve it.

[Ol' Willy]

Most of the games (of any genre) create difficulty by giving unfair advantage to AI, while the AI itself is braindead.

I really want the games to have good and capable AI, and thus, difficulty would be managed by how good AI acts

[Vic]

Most of the games (of any genre) create difficulty by giving unfair advantage to AI, while the AI itself is braindead.

AI should be smart enough but not smarter than me, an average AI does the job well enough. For tactical games, map design is an important factor in how the level plays out.

[wndrbr]

Most of the games (of any genre) create difficulty by giving unfair advantage to AI, while the AI itself is braindead.

AI should be smart enough but not smarter than me, an average AI does the job well enough. For tactical games, map design is an important factor in how the level plays out.

level and encounter design play a huge, if not the main part.

People rave about how advanced AI was in FEAR, but also shit on FEAR expansions despite them utilizing the same engine and same AI. What was the difference? The levels in original game were built with their AI in mind, meanwhile expansions were handled by an inexperienced team, so the arenas aren't as Ai-friendly and don't allow it to shine. STALKER has a good AI, but a typical stalker bot won't perform all that well if you put it into an open field, or an overly linear CoD-like shooting range.

[Lutte]

Challenge is when you die several hundred times and boss battles take 6 hours to win. Anything less is for pussies.

Who even makes games that difficult? I would have to go back to my childhood when I sucked at games to recall getting an ass kicking that hard. Outside of super boss in a game that requires you really optimize your build/setup in order to win, I can't think of any situations where a boss would be that difficult. IMO, If you lose more than about a dozen times to a boss, then you need to take a step back and re-analyze your entire approach to the game, and make some big changes. I like it when a game makes me rethink my strategy and approach, and any sufficiently challenging game should make a player do that once or twice, IMO.

I don't know if it took me a full 6 hours, but the final boss of Rabi-Ribi's main story forced me to go on multiple walks to cool off and recharge before I could beat her. If we're restricting ourselves to RPGs, I recall the optional fight with the soldiers in Voidspire Tactics took many attempts and multiple hours to beat as well, though certainly less than 6.

Ah, a man of culture! Rabi Ribi is easily the best metroidvania to have come out in the past decades if you can overlook the animu shit. If you can't, then I'd nominate Hollow Knight.
Games that don't offer any form of challenges are not games to begin with. They are visual novels. Movies. Whatever. For something to be a game it needs to challenge something. Anything. Your physical abilities (real time action) or your brain (chess) but it needs to challenge something.

This is why I laugh when people debate whether Disco faggoysium is a RPG. Because it's debatable whether it's a game in the first place. RPG = Role Playing GAME. Dunno man, but I would have more fun in a real world golf course than going through this filth.

[Vic]

DE is a game tho because you have quests, choices and consequences, itemization and exploration, basically everything an RPG has bar combat

[Vic]

level and encounter design play a huge, if not the main part.

yes, I'm currently playing an older tacticool game* and they have a bunch of maps, there is no difficulty setting for the AI but the difficulty comes from interesting map design. Sure, ultimately it might boil down to giving the AI an unfair advantage through positioning, amount of troops, etc. but it's still more interesting than having a superhuman AI.

*the game is Advance Wars Dual Strike for the Nintendo DS, playing it when I have some downtime or taking a dump on my old 3DS, it's a surprisingly well made tactical game with very cringe teenage writing. Tons of custom maps to play besides the campaign too. The second DS game in the series seems more mature in tone so I'm definitely excited to play that after I'm done with the first one.

[Gastrick]

Challenge is when you die several hundred times and boss battles take 6 hours to win. Anything less is for pussies.

Who even makes games that difficult? I would have to go back to my childhood when I sucked at games to recall getting an ass kicking that hard. Outside of super boss in a game that requires you really optimize your build/setup in order to win, I can't think of any situations where a boss would be that difficult. IMO, If you lose more than about a dozen times to a boss, then you need to take a step back and re-analyze your entire approach to the game, and make some big changes. I like it when a game makes me rethink my strategy and approach, and any sufficiently challenging game should make a player do that once or twice, IMO.

I talk about it here.
One of the bosses took 6 hours on this game with these settings turned on. That there were a few nights of playing a couple hours and still not winning. Since the game is turn-based, with a team, and with unskippable animations, the attempts take longer than they do in most games, so that is why it took so long. That there are several bosses stronger than him later on has me not wanting to play any longer or turning all those difficulty settings off.

[Gastrick]

The hardest challenge for me is where there's something that you are stuck on, with zero hints on what to do, and you just have to find out what convoluted nonsense the game needs you to do to progress. Not with clear puzzles that you can figure out with some intelligence, but "puzzles" and "mysteries" where no guidance is given at for what the puzzle is and whether you have the right tools to solve it. This has been most 80s adventure games that I have tried playing; the text adventure game Zork is the only one I encountered that doing random shit can at least let you progress.

[rusty_shackleford]

The hardest challenge for me is where there's something that you are stuck on, with zero hints on what to do, and you just have to find out what convoluted nonsense the game needs you to do to progress. Not with clear puzzles that you can figure out with some intelligence, but "puzzles" and "mysteries" where no guidance is given at for what the puzzle is and whether you have the right tools to solve it. This has been most 80s adventure games that I have tried playing; the text adventure game Zork is the only one I encountered that doing random shit can at least let you progress.

A lot of those games were designed this way on purpose for predatory reasons to get kids to call the 1-900 paid tip-line. Something I think was largely forgotten when discussing the 'difficulty' of those games.

[Gastrick]

The hardest challenge for me is where there's something that you are stuck on, with zero hints on what to do, and you just have to find out what convoluted nonsense the game needs you to do to progress. Not with clear puzzles that you can figure out with some intelligence, but "puzzles" and "mysteries" where no guidance is given at for what the puzzle is and whether you have the right tools to solve it. This has been most 80s adventure games that I have tried playing; the text adventure game Zork is the only one I encountered that doing random shit can at least let you progress.

A lot of those games were designed this way on purpose for predatory reasons to get kids to call the 1-900 paid tip-line. Something I think was largely forgotten when discussing the 'difficulty' of those games.

Yes, I've heard this theory before. If it's true, then that makes them even harder, that they were designed to be impossible without cheating and looking up the answer.

[Vic]

mh, more likethey needed a way to make their games seem longer to justify the high prices at the time

[Vic]

@Gastrick seeing how you mention Dragon Quest, the game I was thinking of was Dragon Warrior 1, which I have beaten about 2 years ago. The total assets and size of the game are ridiculously small, but I have played that game for probably 20+ hours, grinding all the way to max level (30 iirc).

Early adventure and platforming games relied on insane difficulty to prolong playtime, JRPGs added an insane grind.

Then I see people who grew up with these games say how modern games are easy, when in fact modern games are barely limited by disk space, so they just play the first hour or so, which is nowadays usually just the tutorial, and complain how the new Mario games are too easy.

Which goes both ways for me, because I know when I start a new game, that I have to invest some time playing the easy/tutorial missions before it gets interesting. Currently playing the Tactics Ogre remake and they go full handholding in the first two missions that all but one characters are controlled by the CPU

[Gastrick]

@Gastrick seeing how you mention Dragon Quest, the game I was thinking of was Dragon Warrior 1, which I have beaten about 2 years ago. The total assets and size of the game are ridiculously small, but I have played that game for probably 20+ hours, grinding all the way to max level (30 iirc).

Yeah, the grinding is quite annoying in that game, just when you think you've grinded well-enough levels to coast through the next area, it reminds you that there's still 3-5 hours of grinding left to do. Thankfully by the SNES generation, you could beat these games without grinding if you play correctly.

Then I see people who grew up with these games say how modern games are easy, when in fact modern games are barely limited by disk space, so they just play the first hour or so, which is nowadays usually just the tutorial, and complain how the new Mario games are too easy.

I 100%'d the first New Super Mario Bros on DS a couple years back, and it really is much easier than any of the classic 2D marios. Only on the last world will you die once or twice per stage, which is a big drop down from the classic 2D Platformers. Many other people who have played Mario Odyssey all the way through have said the same thing about it being too easy.

[Ranselknulf]

I'd be interested to see how AI could be used to augment gameplay difficulty in a less rigid manner than it's currently applied.

I think in the next 5 to 10 years AI will be at the point where it can reasonably outwit people in games that are more abstract than things like chess or go

RPG's that might not have linear quests where you have to gather "10 wolf asses" and take them to "NPC X", but instead the challenge is dynamically updated based on your gameplay style and weaknesses.

I'm not saying the AI should always make a perfect defense against what ever player class and equipment you are using, but if you are raiding a npc encounter repeatedly for coin, potions, exp, etc, then maybe the AI should change its tactics in the encounter to accommodate the min/max nature of the player.

Possibly make it an option in games too, so players who like being able to repeatedly do the same encounter can turn it off.

[Vic]

RPG's that might not have linear quests where you have to gather "10 wolf asses" and take them to "NPC X", but instead the challenge is dynamically updated based on your gameplay style and weaknesses.

isn't that more the case in MMORPGs? Quests in traditional RPGs are basically just an excuse to do combat, "go there and kill this necromancer" and then the difficulty comes from the combat system and encounter design.

[Ranselknulf]

RPG's that might not have linear quests where you have to gather "10 wolf asses" and take them to "NPC X", but instead the challenge is dynamically updated based on your gameplay style and weaknesses.

isn't that more the case in MMORPGs? Quests in traditional RPGs are basically just an excuse to do combat, "go there and kill this necromancer" and then the difficulty comes from the combat system and encounter design.

From what I've seen in MMORPGs is that quests are hard coded based on your class in some cases, or they have a few different tiers of difficulty if you decide to repeatedly do the same quest.

They don't particularly account for your gameplay style or gear choices though. Like maybe you like to cast fireball a million times instead of lightning bolt a million times for example. Maybe you decide to go with a full on axe armor build instead of using swords or other weapons.

I'm thinking something slightly more dynamic than just hard coding encounters to have the same set up based on your class. Taking into consideration your gameplay style might be important too. I don't think it would be appropriate to aggregate all your gameplay data to fully counter you, but lets say you consistently attack a castle owned by the TroonGuard faction, you'd figure that the next time you attack their castle, Baron DongGirl would learn that you always cast fireballs, so maybe they'd adjust their strategies more than just casting "fire shield" or some other bullshit.

[rusty_shackleford]

"dynamic difficulty tuned to the player" sounds like a special kind of hell

[Vic]

I'm thinking something slightly more dynamic than just hard coding encounters to have the same set up based on your class. Taking into consideration your gameplay style might be important too. I don't think it would be appropriate to aggregate all your gameplay data to fully counter you, but lets say you consistently attack a castle owned by the TroonGuard faction, you'd figure that the next time you attack their castle, Baron DongGirl would learn that you always cast fireballs, so maybe they'd adjust their strategies more than just casting "fire shield" or some other bullshit.

but that would only apply to "grindy" games where you kill the same NPCs over and over, right? So something like Diablo or MMOs. Because good encounter design already does what you are describing to make the player use different spells/strategies/etc. as you progress through the game.

For MMOs and ARPGs that would certainly be an interesting concept, where encounters are not strategically designed by the devs.

[rusty_shackleford]

For MMOs and ARPGs that would certainly be an interesting concept, where there is no real encounter design created by the devs.

MMORPGs which focus on difficult group content have encounter design far above that of any singleplayer RPG.
I really can't think of any singleplayer RPG that would get close to the hardest raiding content in early WoW, or the hardest content in Guild Wars, for example.

This is a natural consequence of a game that is continuously developed for poopsockers that demand harder and harder content vs a game that is developed, released, and maybe gets a DLC.

[Vic]

MMORPGs which focus on difficult group content have encounter design far above that of any singleplayer RPG.

meh, single character RPG maybe but not party based RPGs. From my experience with MMOs it's mostly just timing and reflex skills. Don't stand in the fire, quickly dispell the boss's debuffs, and do all that while coordinating with other people.

"far above" is quite the stretch, in fact it pales compared to the strategic possibilities of a single player turn based game. MMOs are mostly twitch skills.

[Ol' Willy]

STALKER has a good AI, but a typical stalker bot won't perform all that well if you put it into an open field, or an overly linear CoD-like shooting range.

AI in Stalker is not really scripted that hard. It has simple routines of breaking the LOS and not coming from the same corner twice in a row; plus no bumrush. This sounds very simple but this gives them some actual freedom of movement which is quite rare in other shooters

[Vic]

MMORPGs which focus on difficult group content have encounter design far above that of any singleplayer RPG.

Just like to point out that this is your site adming about "higher quality" RPG discussion

[Fedora Master]

Game knowledge and execution.

Leads to Slav autism design a la Wrath

[Fedora Master]

MMORPGs which focus on difficult group content

Haven't existed since EQ and it's greatest autists directly lead to the decline that was WoW.

[rusty_shackleford]

MMORPGs which focus on difficult group content have encounter design far above that of any singleplayer RPG.

Just like to point out that this is your site adming about "higher quality" RPG discussion

And I'm still correct. Exactly which MMORPGs have you done hard content in for comparison?
RPGs are designed to be beaten by masses, the hardest content in MMORPGs is designed to waste hundreds of hours of the top .1% people playing the game.