Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Discussions Rules and Guidelines
Report this post
I mean, do you see any reason why the mosquito enemies in Greenpath should be in Greenpath?
The first room of City of Tears goes like this : Belfly, two Crossroads enemies (a jumper and a "bully"), another Belfly, a guard, another crossroad enemy. Is there a reason why the first two Crossroads enemies are there?
What's the point of bushes having currency and/or other stuff in Zelda games?
Though, Greenpath especially does introduce one other mechanic which can be used in a clever way for an enemy that you might have picked up on during the False Knight battle of hitting projectiles to deflect them and possibly damage the enemy if you're gonna use the nail alone when facing them as they hide and can't be damaged when you get close. There is also a section where you can cut down platforms to harm enemies, though I don't see that being reused anywhere later, ergo another reason it seems to me like very basic and obvious stuff being done here and not much beyond that.
Zelda games having currency in bushes? I don't know much about these games since I haven't played any and only seen brief amounts of footage, so it would entirely depend on the context of a given game as many of them are very much different from each other in terms of gameplay from what I've seen. My blind guess would be to incentivise exploration to the player for secrets, maybe give the player some easy boosts if they are thorough and observant enough, or to let the player have some easy stuff if they end up losing some stuff if that's a mechanic in any of the games, maybe?
If I seem blind on any of this, do point it out though.
There is another reason i see : They train the player for something specific, something in Greenpath.
One could also justify why the jumper enemies and the Husk Sentries (the ones with a club) in Forgotten Crossroads are in Forgotten Crossroads with the same reason.
I see. I'll ask about the Hollow Knight equivalent of the purpose i was thinking of, then :
What's the purpose of breakable background objects, apart from some being usable for skips/adding to the scenery?
Note: The reason for the first two Crossroads enemies (that i mentioned in my previous post) in the first room of City of Tear is the first Belfly :
The first time a player encounters a Belfly, they'll probably get hit, so they'll take two points of damage. Somewhat unfairly, too, since players are probably supposed to take damage, as a lesson intended to teach that Belflies deal double damage.
The two Crossroads enemies are there to counter that damage/unfairness, by allowing players to get enough soul to completely heal the damage caused by their first Belfly encounter.
Out of curiosity, have you made, produced and published a game to the standard that you are looking for? If so, is it on steam or similar?
Edit: Just to be clear, this is not meant to be antagonistic, you have mentioned that game design comes naturally to you and that the game developers for Hollow Knight have provided something that is the basic minimum standard to be expected. I am presuming that there is a gold standard that you have already achieved that we could use as reference?
Having doubts due to no direct reference is understandable, it is much more difficult to convey with words than it is to show a working example, especially since a working example might present issues you may not have considered before, in particular during testing. Since I am actively working on stuff here and there, it is possible someday you'll see some of it released in the future. Given the growing improvement of game design in games over time, I expect in a few years time, what Hollow Knight has done will probably be looked at as the base line, not the peak.
However I have referenced Darkwood earlier in this thread, while not exactly the same genre, you can look at video reviews of the game as well as try it for yourself to hopefully see exactly what I mean. It's not just the horror aspect that it nails above most horror games out there, but everything else surrounding it as well, for the most part.
What about rooms that are not explicitly lore dependant? Because most rooms aren't, especially considering the fact that NPCs travel from place to place where you could not before, leaving you wondering how they got there. It's a gameplay and progression reason that the lore would not explain, at least not to my knowledge.
My point was that lore dependant rooms would still exist in their respective regions, if absolutely required, like for example the room with the White Lady would always be found in Queen's Gardens region. That does not mean that many of the traversal and puzzle rooms cannot be procedurally generated to still give adequate challenge and have conditions to attempt to teach the player something specific at certain junctures. Procedural is not pure random, it's random with a whole bunch of conditions set in place, whether it's so you don't get softlocked or otherwise.
>Really unfortunate this isn't a thing
I couldn't be more glad that it isn't. Seems like every bloody indie game is rolling out some procedural stuff and I think it's because it's easier to program something to make areas rather than make them by hand. Hollow Knight is great because every area feels rich and full of life since they were all done by hand.
No thanks to any procedural stuff in my HK. Leave that to the hundreds of other games that go that route.
And seriously, are you guys really entertaining OP as an actual developer? I've never seen a guy use so many words to say absolutely nothing. He's supposed to be a pro yet is advocating that Team Cherry use a crappier, cheaper and lazier method to make their levels. Spare the rest of us the gullibility, please.
Every bit of this is completely wrong and makes assumptions that go against what I've stated and clarified. If you don't know how the process of procedural generation works in this regard, which clearly you cannot tell me you do, since every single thing you've stated here does not function the way you've described. Procedural generation often incorporated hand-crafted sections with logic to account for you not getting soft-locked or having things placed in areas they shouldn't be. It's never really a one or the other scenario. If it does happen that way, you didn't do it quite right. That's really what procedural means, to have rules in place, otherwise full randomisation would suffer from problems that you're probably most concerned about.
I don't claim to be a pro on the entire subject of game development, but I do have much better understanding and capability of working with this sort of stuff than most.
And hand-crafted content is definitely nowhere near as rich in the game as you claim either. There are several areas that just have lots of empty space for no real gameplay reason.
Nope, but gladly, explain exactly to us how this would have zero gain, despite what I have clarified in this entire thread.
That is partially correct, yes there are 3D objects and the scene is layered in such a way so as to give you the illusion of depth between foreground and background elements (well in this case the depth actually exists, which is cheaper and easier to do). The assets could be integrated into a procedural room generation system given that there are several assets that are reused for various areas and they don't look off either.
Most notably, many of the City of Tears rooms have asset parts reused, same really goes for most areas of each location. It's not like the entire section of a level is drawn in full from start to finish (with the possible exception of some backgrounds, which can also be reused at times), stuff gets copied and pasted right next to one another without significant variation because otherwise it would take too long to create any assets for the game's levels, and if a level layout were to change during development, you would have to redraw parts of an entire thing instead of just shuffling parts around to better fit the design of your level/room layout.
That might be a fair trade to some, but proc gen doesn't even add replayability for a majority of players. Hollow Knight has a lively speedrunning community and plenty of people I know have beaten the game more than twice. Hollow Knight finds replayability in redundancy, crossing paths, keeping in bugs and skips, and shortcuts with secrets for knowledgeable players. It doesn't need procedural generation.
Procedural generation is also loosely approximated with a mod that makes level transitions randomized. The consensus on this experience overall is that it is far too confusing and drawn out even with a map, and isn't fun to play unless your a a highly proficient player who is aware of everything in the game.
An important thing to note is the developers approach to game design. It happens to be the exact opposite of yours. They treat HK as a living world and, in a sense, allow it to "design itself". They have made it clear that they do not place benches with the intention of smoothing out gameplay, they do it because the world would have a bench there. There is not a bench before every boss fight, nor is there the same amount of benches in every biome. By creating a world like this, hand-made and self-reliant, they make an organic experience where the player can't guess what happens next. Ironically, in a procedurally generated world, the player can quite easily predict the next obstacle or resting place because of the nature of computers. (if you didnt program in a failsafe, then there is a chance the total allowance of benches are placed next to eachother. When you do add a fail-safe, the player is now aware that benches must be roughly this far apart, and can predict the next bench location.)
You used Darkwood as an example? Darkwood is an entirely different game, and your bringing it up works against your point. Darkwood is a world that is natural. Things can be scattered around at random and fit the feel of the game. When things like towns or other large structures exist they are large set pieces that were designed by hand; no proc gen. There are just plenty of reasons for Darkwood to be procedural that it seems futile to list them. Its top down, there aren't platforms to get caught on and it doesn't need to generate tiles or ledges or pools, it just needs to place set pieces a specified distance from other set pieces. The story involves an entity-like forest that forms into random and confusing configurations. Its atmosphere is built to be unsettling and becoming lost and confused is part of the design. The map is built around landmarks and only has a few large "rooms" that are entirely open area with set pieces. Compare this all to HK. Most of HK's world is bug-made, meaning in Darkwood's terms it should be handcrafted because randomly generating human structures is unfeasible. HK is side-view meaning all objects need to be made with a specific sprite-size, jump height, and jump-length in mind. HK has lots of enemies in every room meaning some form of a cohesive generator would need to be made to spawn them so rooms dont look like an enemy randomizer. (Darkwood has singular enemies spread far apart, or enemies are built into set-pieces.) Hollow Knight is a metroidvania, items are necessary to obtain for progression, how would you make a stable and satisfying generator to string together an intended path of progression? Hollow Knight has shortcuts, secret paths, unintended jumps and enemy interactions, could those be implemented into a proc gen engine? Darkwood pulled off procedural generation because its entire setting and gameplay loop benefited from it, and Darkwood is still criminally unpopular. Perhaps they would have benefited from handcrafting the world...
I have an "intuitive grasp" on game design. I've "had the pleasure of working on several mods." I'm "smarter than the average person about these things." You sound every bit as obnoxious as I have in this post. Let your accomplishments and the structure of the argument speak for your ethos, not your self-proclaimed mastery of design to the point that Hollow Knight is child's work to you. Do you know why its design comes so naturally to you? Because the developers constructed a world so beautifully that even the most basic player can come to grips with its formula and be comfortable playing it through familiarity. Look up the design of Disco Elysium, that game is a masterwork and a piece of art and it accomplishes this through having no pride and making the story and characters digestible to the average reader.
I'm open to further debate.
I have a sneaking suspicion that when it comes to good game design, people do not understand, and maybe even some developers themselves that, from a developer standpoint we all end up having to follow many of the general rule sets for making better games one way or another, just some of us acknowledge, understand or admit them more and better than others. I do acknowledge my intuition in design and try to make sense of it, some others simply do not (and most of the general public certainly does not). But following intuition for game design but not paying attention to the specifics of it and happening to do it well doesn't necessarily mean you've gone off the road of doing it in what would be scientifically sound, all you're really doing is ignoring that it is the fact of what you've been following and then possibly excusing lack of quality as "art" and "not following the norm" where you may have not done so well. The results then tend to be often very close or the same either way you look at it but spiting that will just leave gaps in quality. Experimentation itself is part of scientific conduct anyway however so it's not like that thought creates monotony inherently, experimentation is why I much prefer indie games over whatever large studios make since they just rehash formulas that are shallow and boring because they were new and worked well enough the first time. But experimentation is in and of itself how you find what works, why, and how to do better. Assessing that can come after to help develop it further. That's exactly how you apply science to an art form.
That entirely depends on how deep the procedural generation goes, and how much hand-crafted content is part of that generation to account for cohesion, pacing etc. I'll talk more about this when we get to the section about Darkwood.
Exclamations like this don't really mean anything, it's fluff. Why did you write this? And it appears to be the primary type of excuse from almost every single reply here. I mentioned the option of keeping the hand-crafted world as an option, maybe even the default setting, or requiring the player to choose between one or the other before starting.
Procedural generation option would then follow a similar pattern as the original world craft, but with things changed enough to give better replayability and a more continuous world exploration dynamic and of course, making maps and the compass still useful after your 5th playthrough.
No, it is pretty much in the same ballpark.
Which is exactly what I would do, keeping in mind scope and not ruining gameplay for potential logistical sense (you can't, or at least are not supposed to super-dash upward without one specific mod that does that, for example).
I don't think that's as true as you think it is. They wouldn't for example have a bench in a position to permanently soft-lock you "just because it makes sense for one to be there" unless it's also a result of them overlooking a condition, under which it would be imperative for them to find a suitable solution to the logistical problem that is still coherent with the established world and gameplay (I have recently found a way one can soft-lock themselves in the Junk Pit if you are in Steel Soul mode). The bench near the bottom of the Deepnest for example, or even the one in the centre of the Royal Waterways, to me seems very gameplay intentional. They aren't there because "it makes sense", the developer had to come up with a creative way TO make sense of it being there to give you a fair leeway.
As for the bossfights, I heard about this specific thought, and when playing with a bench a little way away from a bossfight, there was a dynamic gameplay reason to have that there too that I ackknowledged, though I realised the reason the first time I heard about it, not when experiencing it for myself. The more you navigate from a bench to a boss, the more it matters if you die or even get damaged as the better you are at getting back, the more opportunity it gives you to be better equipped to fight a boss. Getting to a boss with 1 health from a bench really isn't going to go well for you. This dynamic encourages you navigate the world while making less mistakes in taking damage, in effect also getting you better at doing so. I would therefore argue this is a form of skill progression seen in its prime.
Similar reasoning can be assimilated for the amount of benches and their positioning in a given region, the regions are very different from one another in size and form and while the distance between them can be highly variable at times, it's still calculated to a fair degree and their placement can, at least sometimes change the setting and dynamic of a given area, as mentioned with the previous examples.
This entirely depends on the complexity and thought you put into the procedural generation itself. It is done poorly in a lot of games (or so I hear, not that I play enough games at random as I don't get a game if I don't sense it having a fairly good quality already or something unique to it that would be to my liking, which is why my library is pretty small compared to that of many others), so if that's all you generally know, that's understandable, but deeming that as an absolute I would have to say is very much false. I have actually seen the opposite in a game, where the lack of randomisation actually makes everything dull because you can predict the outcome way too much.
Everything about this seems more or less just grasping-at-straws nonsense from what I can tell. Any sense of "human-made" and "bug-made" create no reasonable argument of any kind here. And "randomly generating human structures is unfeasible", what are you talking about? If you are talking about creating structures that are typically made by hand, that's also false and would tell me that you don't know the field enough to make a claim like that (which if you did, it would be false and I doubt you would have made it... that is if I'm understanding what you're saying here clearly, which I'm not, you need to be a bit more clear about what you're saying here, it's hard to tell).
It appears to me that "Darkwood has singular enemies spread far apart, or enemies are built into set-pieces." is the exact reply to you saying the former sentence of "HK has lots of enemies in every room meaning some form of a cohesive generator would need to be made to spawn them so rooms dont look like an enemy randomizer." Yes. That would be the case, and Darkwood here gives you a fairly good approach to do that. Do you have a point beyond this that goes against the idea of having a decent form of procedural generation here?
Of course not every room would have to be generated object by object, that might increase generation time of the world enough to be too long of a wait, and a lot of the time you probably won't even want that sort of approach. Some could have that, other rooms be hand-made and always there and in specific regions, others hand made and sometimes incorporated yet more that would be hand-made but with some variation in the region they are positioned. The region layout itself on the map could also have variation.
There is a lot of logic you could apply to the procedural generation algorithm that isn't just a uniform solution to the entire generation, because yeah, a uniform solution could very well potentially retract and be a waste of time. Suggesting procedural generation however doesn't necessarily indicate that. No Man's Sky has a very uniform solution and while impressive, it does get very dull. It is, of course, a much bigger world you are given to work with, but Hollow Knight doesn't even have 2 dozen regions to account for. Not like the hand-crafted aspect is all the more better for some as is, The Hive is arguably done in a very sloppy way, it actually looks as though it was uniformly procedurally generated. And I'm not one to argue that such a uniform and dull approach would make a great game, there is a lot to consider, and Darkwood for me has so far been the starting example of how it can be done well.
Hollow Knight actually has a semi-branching path of progression, which works really well.
Yes, yes, yes and yes. The fact you're even questioning this tells me you know very little about the subject. But if there is more to what you're asking here, I'm open for you to elaborate.
Is your argument here that Darkwood pulled it off, therefore it's good that it has it, but Hollow Knight doesn't have it and I'm not sure how that would work, so it probably couldn't benefit from it, therefore it shouldn't even be attempted? Hard to say. But then you backpedal with...
Far as I can tell, that's making a very far-fetched supposed causation correlation. A possibility in the grand scheme of things, sure, but doesn't seem very founded in anything since no further reasoning has been given for it.
...
I don't think you or I are any bit obnoxious, but I understand what you mean by "sounding obnoxious". I don't know how else to communicate that thought though, because I do think what I have said about my experience matters in this context. The reverse is that would be something along the lines of "no idea", "I've attempted nothing in the field" and "if the general consensus on this is X, then X is probably correct". But that is not the case, it's just false, and it's certainly not a very good idea to not discuss anything to the contrary just to make the general consensus feel good about itself. And Hollow Knight isn't child's work to me, it just can be developed further to make it even better and I believe so far I have argued fairly that a procedural generation is one way that could be done if the developers decided to dedicate time and effort for further development.
I do understand the idea of "show don't tell". And I very much am certain I will be making things to show that demonstrate exactly what I have been trying to argue for, just not yet for a lot of personal reasons. That doesn't mean that I can't make a suggestion and provide reasonable arguments for it. Salty remarks being thrown with lack of understanding about the suggestion being the general response maybe is even less of a thing to throw out there than what I have. On what grounds do you think I have to be the peak of everything on the subject (when I don't even claim to be to that extent, now that would REALLY be stupid and pretentious and more than likely false, though not necessarily), while everyone else can throw whatever they feel like and it's just as reasonable of an argument somehow?
I don't agree that people can generally understand design from a game. I have heard many mistake things when trying to explain why something is so good, or just describe it and then never actually come to analyse it properly. Good understanding of why you or others feel good or bad about something isn't something most people can do, that's just a general fact, and there isn't always necessarily a consensus on what is good or bad about a game to people personally either, people like and don't like different things for different reasons, many of which they are unaware of. But again, there are general trends you can explain and conditions you can account for if you study the subject and maybe somehow get intuition for parts of it, maybe without even trying too hard.
I have heard of Disco Elysium and how well it has done some things. Not sure if it's a game I would be interested in since it appears to be too RPG oriented and I really have not liked the typical RPGs in general personally for a variety of reasons, though that may be something to look at for eventually for sure.
Overall I think the fact I can give examples and reasoning, people clearly haven't even thought of (otherwise it would have been mentioned by others before I even made counter-points) in favor of adding it, if TC decided sure, why not, and my arguments, as far as I can tell being fairly robust to the idea, then the end result being others grasping at straws or getting a little salty simply because I made a counter argument is just strange and a bit amusing. But that's really just unfortunate. I am however of course, completely open to having very solid reasons of why this would never work in a game like this given a solid attempt, given the computing power people typically have available and that's still reasonable for the game regarding that, with everything I have presented taken into account. But a solid argument against it that really does so has really not been formulated from what I can tell.
I do want to preface that I don't want this to get into a "debate" type thing where I strongly hold my opinions, you and maybe some others do theirs and then we "battle each other with random arguments until one gives up or gives in". It shouldn't be a battle, it should be a discussion, thinking about what could work, how and why, and what couldn't and why, that's kind of what forums are for. If you happen to agree on one side more than another, that doesn't mean you have to suddenly establish a black and white scenario on what you stand by and always will no matter what. Be open, as I have been trying to be, even if it doesn't necessarily look as though I have. I'm sure you have, and if so, I'm greatful for that, but I just hope others don't conclude it otherwise. Clarity is important and sadly it comes with elaborate posts with ever potentially bigger oversight on writer's part and people just ending up unwilling to read through and try to understand it because it's too long to read. But I'll try my best anyway, otherwise you just have Twitter.
Your very good at writing and debate, I feel. Those skills should be put to use summarizing studies or creating scientific papers. Before I began typing out my text wall I felt like I might be making a mistake. I don't often argue online and I've always suspected the only way to win a digital discussion would be to avoid participation. This time, though, I thought I might want to intervene because this was a topic I felt passionate about and I was familiar with many of your points. As I read your rebuttal, though, it starting becoming certain that there was no real way to move forward, not without spending another hour or so composing a symphony of baseless thought to toss to the men inside the screen. So why type this? Because I respect your effort and I don't wanna to leave you hanging. I just have some important things to do. I'm currently helping a friend work on randomly generating levels for his game.