|
TooMuchAbstraction posted:Sure, and I recognize that my opinions are not always shared by the average gamer. But the average gamer has horrible taste, so this doesn't bug me much. I love all this, for real. TooMuchAbstraction posted:But the idea is that items would be a reward for accomplishing specific major things, not things that are randomly doled out by a Skinner box. Typically that's the case for big milestone improvements, but along the way having little boxes of loot that may or may not be neat is part of the entertainment. Consider this: if you want to go where the game rewards like items are provided sparsely, like once every half an hour or so, what else do you got that keeps the player going for all that time where there's no chance of even a micro reward? Stick to the genre-- clearly people play games for high scores, best times, etc.
|
# ? Jun 5, 2015 01:22 |
|
|
# ? Apr 28, 2024 04:05 |
|
madjackmcmad posted:Typically that's the case for big milestone improvements, but along the way having little boxes of loot that may or may not be neat is part of the entertainment. Consider this: if you want to go where the game rewards like items are provided sparsely, like once every half an hour or so, what else do you got that keeps the player going for all that time where there's no chance of even a micro reward? Stick to the genre-- clearly people play games for high scores, best times, etc. Items provide a feeling of progression, as your character goes from being weak to being strong. There are other ways to provide progression without significantly impacting the player's power level, though. You can increase the complexity of the environments they explore (as well as simply changing the environments, e.g. surface -> caves -> volcano, so at least the player isn't moving through the same kind of area over and over again). You can change the monsters they face, fiddle with their AI, etc. Necrodancer is a fantastic example of this as a Zone 4 monster isn't more powerful on paper than a Zone 1 monster, but they have more complicated behaviors. You can raise the stakes involved; say that failure in the first region "sets you back" in the metagame less than failure in the fourth region does. Of course you'll have more time invested in the latter case anyway, so the stakes are automatically raised. Less friendly to roguelikes, you can also reward the player with bits of plot, nice scenery, cutscenes, new music, special animations, etc. All of this is expensive to produce and/or only "useful" once for player motivation, hence why I said it's not so roguelike-friendly. Finally, to the extent that most roguelikes are basically large tactical puzzles, being given a new puzzle to solve is a reward in itself.
|
# ? Jun 5, 2015 01:35 |
|
Sergeant_Crunch posted:How good is sleep gas generation in Qud? I'm thinking of tinkering with my main build a bit so I'm not using paralyzing stinger, since apparently a back slot is extremely useful to have later in the game. Sleep gas seems like it would be a decent alternative, even if it means giving up corrosive gas. I'd be curious too. I assume it doesn't work on robots and that there are loads of them. Are flamers really the end all weapon in qud? Can oil be crafted?
|
# ? Jun 5, 2015 01:35 |
|
The only ranged weapon with close to the same penetration is the phase cannon, which eats 3/4 of a chem cell every shot. Melee needs a 2handed Metametal Maul or a vibro weapon, and I don't think metametal actually drops anywhere. Oil can't be crafted but there's tons of it in the Asphalt Mines. A fun trick: If you have Strapping Shoulders and the ability to tinker flamers you can use flamers instead of waterskins to halve the weight of your water and other liquids.
|
# ? Jun 5, 2015 01:46 |
|
Is there any Roguelike that has the same "Fast Paced Action" feel of DoomRL? I doubt it, but I'd love to play it if so. Also I keep hearing good things about Sil, so I should try that.
|
# ? Jun 5, 2015 01:51 |
|
AliensRL was by the same guy and has a similar style. I actually prefer it to DoomRL myself. Ammo is a lot less plentiful so you'll find yourself having to conserve and pick your battles.
|
# ? Jun 5, 2015 02:06 |
|
In Caves of Qud, the Golgotha dungeon is basically the game shouting "SURPRISE, rear end in a top hat" and dropping you into DoomRL for a while while you scream and flail your arms in confused terror. It's otherwise much slower-paced though, rather like basically every roguelike that isn't DoomRL. I've heard that BerserkRL, also by the DoomRL guy, is pretty good but I've never played it myself.
|
# ? Jun 5, 2015 02:13 |
|
I would design a roguelike similar to the ripping mechanics in Last Stand. Every major monster you kill you can rip a part off and slam it on to a part of your body. You have several slots and what you rip and what skills you get depend on what part of your body it goes onto. By the end of the game you'd be a monstrous freak covered in monster bits, using them to obliterate your enemies.
|
# ? Jun 5, 2015 02:16 |
|
Angry Diplomat posted:BerserkRL It's pretty cool, not something you can play for hours on end just do to it's nature but it's a fun time.
|
# ? Jun 5, 2015 02:20 |
|
TooMuchAbstraction posted:Items provide a feeling of progression, as your character goes from being weak to being strong. There are other ways to provide progression without significantly impacting the player's power level, though. You can increase the complexity of the environments they explore (as well as simply changing the environments, e.g. surface -> caves -> volcano, so at least the player isn't moving through the same kind of area over and over again). You can change the monsters they face, fiddle with their AI, etc. Necrodancer is a fantastic example of this as a Zone 4 monster isn't more powerful on paper than a Zone 1 monster, but they have more complicated behaviors. You can raise the stakes involved; say that failure in the first region "sets you back" in the metagame less than failure in the fourth region does. Of course you'll have more time invested in the latter case anyway, so the stakes are automatically raised. Are you of the opinion that the one item you get every thirty minutes should be randomly generated or predetermined?
|
# ? Jun 5, 2015 02:24 |
|
madjackmcmad posted:1 item per 30 minutes of gameplay is unfathomable. Is that really 30 minutes where every chest is empty, barrel barren, and the monsters drop nothing at all? That's Diablo 3 functionally at high gear levels, if that. I don't even click anything that's not orange or green and I probably get a useful one once an hour or less.
|
# ? Jun 5, 2015 02:29 |
|
I think I'm gonna stick with sleep gas generation. It seems a bit less reliable than the paralyzing stinger, but it can usually proc if you kite an enemy through 3 tiles of it, plus it can put multiple enemies to sleep, so it makes a decent panic button if you get swarmed. I swapped out corrosive gas for phasing to take over the role of going through walls. I think I actually prefer phasing. It's not as versatile as corrosive gas, but it is excellent for getting out of a rough fight and can get through walls without destroying them or causing any collateral damage.
|
# ? Jun 5, 2015 02:30 |
|
Sergeant_Crunch posted:How good is sleep gas generation in Qud? I'm thinking of tinkering with my main build a bit so I'm not using paralyzing stinger, since apparently a back slot is extremely useful to have later in the game. Sleep gas seems like it would be a decent alternative, even if it means giving up corrosive gas. It's a solid escape mutation at its new cost of 2 points. It doesn't have the offensive capabilities of the paralyzing stinger though, since taking damage will wake up a sleeping mob.
|
# ? Jun 5, 2015 02:31 |
|
A good example of a game that gives you one item per 30 minutes is every Zelda game. Obviously there are rupees and keys and whatever, but you get exactly one "real" item in each dungeon. That probably doesn't translate perfectly to roguelikes due to randomness, but I can definitely imagine a roguelike in which you clear several floors -- which are devoid of floor junk -- knowing that you're guaranteed a sweet upgrade for your current gun (or whatever) at the end. megane fucked around with this message at 02:37 on Jun 5, 2015 |
# ? Jun 5, 2015 02:34 |
|
megane posted:A good example of a game that gives you one item per 30 minutes is every Zelda game. I was going to make the point that metroid (+ castlevania? I honestly dont know) and zelda games pretty much operate this way, but I dont know how far outside the realm of RPG they're willing to travel. It's pretty much the best comparison though in my opinion.
|
# ? Jun 5, 2015 02:41 |
|
Lilli posted:Are you of the opinion that the one item you get every thirty minutes should be randomly generated or predetermined? You can do either and in any number of different ways. The Isaac / Risk of Rain style of having a ton of predetermined items of which you get one would work well, but so could a procedural item generator. Really procedural items are just bodging together combinations of hand-designed abilities (even if those "abilities" are just numeric improvements), so as far as I'm concerned this is just a matter of where on the spectrum you want to be -- anywhere from "there are N hand-designed items and you will see them all, in order, each game" to "there are 1000 different hand-coded abilities and we're going to mix and match them and generate N items from that process". The main thing is that, since items happen so rarely, they should each be distinct. You don't want to play for an hour only to get two items that do basically the same thing. So in a game where items are handed out rarely, any procedural generation ought to consider the complete set of items to be generated and "lock out" attributes once they've been assigned to an item, so they don't show up on another one (or at least, don't show up again until a sufficiently long time has passed). In that sense, then, items would be predetermined because the complete set of items is decided when you start the game
|
# ? Jun 5, 2015 02:44 |
|
Symphony of the Night is a great example of suffering through a lack of loot for good stuff. I mean you do end up with a pile of items but it's not too bad.
|
# ? Jun 5, 2015 02:45 |
|
FTL seems like an example of itemization done well. Even if the player can't/won't use whatever weapon their ship picked up, the player receives items infrequently enough that just selling found items results in a very desirable scrap income that doesn't require ridiculous inventory tetris or other typical item shenanigans.
|
# ? Jun 5, 2015 02:47 |
|
The issue with FTL is that if you don't get certain items your run is absolutely hosed.
|
# ? Jun 5, 2015 02:49 |
|
Million Ghosts posted:The issue with FTL is that if you don't get certain items your run is absolutely hosed. LazyMaybe fucked around with this message at 03:05 on Jun 5, 2015 |
# ? Jun 5, 2015 02:59 |
|
Million Ghosts posted:The issue with FTL is that if you don't get certain items your run is absolutely hosed.
|
# ? Jun 5, 2015 03:22 |
|
Million Ghosts posted:The issue with FTL is that if you don't get certain items your run is absolutely hosed. People are reliably able to beat FTL on Normal, which is the mode it was balanced around. Or at least as reliably as people can beat any "real" roguelike (i.e. at least 90% of the time if they know what they're doing). I don't consider FTL to be unbalanced, just rage-inducing.
|
# ? Jun 5, 2015 03:24 |
|
I haven't played FTL in a long rear end time so I'm probably wrong. I also played on Hard 90% of the time so bias and all that.
|
# ? Jun 5, 2015 03:28 |
|
Lilli posted:Are you of the opinion that the one item you get every thirty minutes should be randomly generated or predetermined? I'm not him, but this is a spectrum, not a binary choice. It doesn't have to be 100% random over the entire specturm of loot. Predetermined is one way to go, and a lot of games do it well, but you could obviously make it have some random properties, and some properties guaranteed to be relevant to you or an improvement in some way. Don't bother dropping items that your current gear is a pareto improvement over is a decent baseline I'd say, though you could do better than that still.
|
# ? Jun 5, 2015 03:34 |
|
I'd also like to see more roguelikes with low damage. like paper Mario low damage
|
# ? Jun 5, 2015 03:38 |
|
Are there any one item per half hour games with randomized drops? Zelda/Metroidvania games are a great example, but those items are static as hell and more often than not game-changing. Hell, that's pretty much the whole point of Metroidvanias; slowly amassing completely new powers that radically change how you interact with the map. I'm curious if it would be possible to introduce randomness to the formula without making a big mess of progression or an unfun RNG-fest.
|
# ? Jun 5, 2015 03:41 |
|
play binding of isaac really slowly
|
# ? Jun 5, 2015 03:46 |
|
Big Mad Drongo posted:Are there any one item per half hour games with randomized drops? Zelda/Metroidvania games are a great example, but those items are static as hell and more often than not game-changing. Hell, that's pretty much the whole point of Metroidvanias; slowly amassing completely new powers that radically change how you interact with the map. I'm curious if it would be possible to introduce randomness to the formula without making a big mess of progression or an unfun RNG-fest. My dream game is a procedurally-generated Metroidvania, but it turns out to be really hard to procedurally generate 2D platforming maps that are fun to interact with. And even then, the items would all be hardcoded; it'd just be the map layout, item order, and which subset of the overall available items you get that changes from game to game. Conceptually you could have a top-down Metroidvania with items that similarly change how you interact with the map, which would simplify map generation considerably. But generating the items procedurally while still making them all be game-changers that give you interesting new traversal powers would be a major undertaking. Most roguelikes are happy to just give you new interesting ways to kill things without making those ways also be keys that unlock new parts of the map.
|
# ? Jun 5, 2015 03:59 |
|
Generate a small map and the first item. Once that is generated do a second pass that adds areas you need that item to get to and through with. Then repeat? That's probably the way to do it but I am not a programmer. Probably would need a lot of sanity checks to make sure the first item didn't just let you everywhere.
|
# ? Jun 5, 2015 04:19 |
|
Combine that with your small damage numbers idea and I'd play the poo poo outta it.
|
# ? Jun 5, 2015 04:23 |
|
100 HOGS AGREE posted:Generate a small map and the first item. Once that is generated do a second pass that adds areas you need that item to get to and through with. Then repeat? That is the very broad gist of it, yes. It gets considerably more complicated when you consider things like * you want the map to be reasonably compact and to favor horizontal travel over vertical travel (assuming you're doing a platforming game) * for a traditional Metroidvania, you want to reward revisits by having areas locked behind "high-level" locks that have various minor powerups * there should be shortcuts back to lower-level areas to speed revisits and improve interconnectivity * you don't want to have lots of long corridor segments going not much of anywhere * you have to fill in the actual map with interesting rooms (presumably by throwing down a bunch of templates and then populating them with monsters, but those templates have to be carefully designed) Last time I worked on this I got this far: Which isn't awful, but it's also still years away from something that's actually playable -- there's no underlying structure to the map, it assumes vast flexibility in ability to fill in the map contents, all of the "lock" rooms (in red) and "key" rooms (in cyan) are 1x1, etc. Every constraint you violate in the name of making the maps more interesting also makes them harder to generate. TooMuchAbstraction fucked around with this message at 04:29 on Jun 5, 2015 |
# ? Jun 5, 2015 04:25 |
|
Ugrok posted:Hey ! As you have gathered, most people here really dislike the Pit. Hi guys. It's so lonely here. I really enjoy it. If you haven't already, I'd grab the expansions through Juggernaut so you get the ammo conversion recipes (which should have been in the base game, but I digress). I found this guide useful for the beginning: http://steamcommunity.com/sharedfiles/filedetails/?id=163785559 The wiki (essential) is here: http://sword-of-the-stars-the-pit.wikia.com/wiki/Sword_of_the_Stars:_The_Pit_Wiki The Pit forums at Kerberos are here: http://www.kerberos-productions.com/forums/viewforum.php?f=54 I haven't found a good lets play to recommend for the Pit so your guess is as good as mine. Full disclosure, I have not beaten the Pit once (or Stone Soup, or Sproggiwood, or Dungeonmans, or...) I am bad at roguelikes. However I have some advice. -Start with the Ranger. Three actions per turn is invaluable. -On normal explore and tinker with all objects on levels 1-5 for skill points. From levels 5+ you have to consider how much ammo, health and durability it is going to cost you to clear a room. Are the skill points you will acquire/objects in the room you can use worth the trade off? -Get life sense in Psionics immediately. -Get Foraging to 80 or so -Skills = items - Computer allows you to hack Armor terminals. - Mechanical opens ammo boxes, and repairs Cookers and Labs which are invaluable as you go further down. - Lockpick you will increase by opening every door normally and you will find/make boosts like lockpicks. Worth investing a few points in at your own judgement. Lockpick directly translates into opening weapon crates which will give you worthwhile powerful weapons if your foraging is high enough. - Decipher is actually really important as it allows you to access tesseract wells and ammo dispensers. From depth level 7 on, I recommend increasing this skill.
|
# ? Jun 5, 2015 04:27 |
|
100 HOGS AGREE posted:I'd also like to see more roguelikes with low damage. like paper Mario low damage In Crypt of the Necrodancer your starting weapon does 1 damage, and the highest-tier weapon material does 4. You can also improve your weapons by getting ones with better attack patterns though. A spear can hit enemies two squares away, while a broadsword hits all three squares in front of you. The game is definitely more about positioning than damage, though. The majority of enemy types in the final zone only have one or two hitpoints, with their difficulty coming from their patterns and special abilities. RPATDO_LAMD fucked around with this message at 04:37 on Jun 5, 2015 |
# ? Jun 5, 2015 04:33 |
|
Some Qud weapon info: Flamethower: 9 shots in a 10 degree spread, 1d2 damage, 15 penetration, also raises temperature 4d12 degrees per hit Phase Cannon: no spread, 4d12 damage, 12 penetration Eigenrifle: line AOE, 2 degree spread, 1d12 damage, 8 penetration Masterwork Carbine: 3 shots, 6 degree spread, 1d10 damage, 5 penetration 2 handed Metametal Maul: 4d5 damage, up to 12 penetration, probably doesn't drop Crysteel Hammer: 3d3 damage, up to 9 penetration, drops in the Asphalt Mines Vibro blade: 1d10 damage, always penetrates once or twice. Explosives seem to involve some wonky physics calculations to do damage so can't really compare them. They do tend to be good against things like the 15 ac Cragmensch in Bethesda Susa.
|
# ? Jun 5, 2015 04:57 |
|
lost a two and a half hour game in crawl from removing a distortion god gift and not realizing it could cast me into the abyss, at level 16 I survived for around 2000 turns before succumbing to a confusion on my berserked troll bro finishing me off
|
# ? Jun 5, 2015 05:36 |
|
I love this thread Good conversations on design. My two cents since I'm too tired to write a full screed: Inventory is typically a side effect of other systems. Usually two: Equipment and consumables. I generally often dislike the latter for a variety of reasons, the former, well, there's your rpg progression parallel to the usual levels and character customization. If you cut down on (or eliminate!) loot, you're potentially removing decisions from the game (this is ignoring all the psychology of 'oh boy, new items!'), and you're removing a lot of busywork (sort trash, pick up trash, examine trash, sell trash, equip non trash, store possible sidegrades or later upgrades, contemplate navel, or oh god manage weight limits and hauling garbage to town or stashes). The ideal would be keeping the interesting equipment decisions, but removing all the garbage, both in terms of loot and in terms of ui clusterfucks (of which there are many and myriad failures spread across the entire spectrum of games from Z to AAA). It's not an easy ask, but there are a fair number of rpgs (and just games, roguelikes or otherwise) that do a good job of giving you interesting equipment customization without drowning you in either mountains of garbage or forcing you to play the inventory busywork minigame. I'm generally most entertained by a game that has a lot of decisions to be made. The more interesting the decisions, the more engaged I am. Plenty (plenty) of games are bloated with decisions that aren't actually decisions at all, having a very small (sometimes singular) number of correct, optimal, efficient, or otherwise beneficial outcomes. Inventory management ranks even lower on the fun scale than games with busywork decisions (oh boy, a +2 foo to replace my +1 foo!), because it's usually a necessary (?) evil to handle gear and consumables. And again, that's setting aside the psychology of loot games and vertical progression blah blah, ranted about that before. I kinda think it's another of those things that's just part of Being a Good Game, as it's another area that gets polish and attention in a well crafted game, but is often an afterthought in quite a few others for any number of reasons (limited resources, manpower, time, obliviousness). I do very much appreciate games that handle it cleanly though.
|
# ? Jun 5, 2015 08:52 |
|
victrix posted:The ideal would be keeping the interesting equipment decisions, but removing all the garbage, both in terms of loot and in terms of ui clusterfucks (of which there are many and myriad failures spread across the entire spectrum of games from Z to AAA). I'd say modern tabletop design has found the solution to this, but the games I'm thinking of aren't really modern anymore and currently the big name tabletops are in their "40 years ago was perfect!" phase. Basically, if players are given the control to populate the random loot table the games doesn't have to roll on it nearly as often to provide equipment the player is actually interested in. This opens up the possibility for builds that are defined both by skills selected and niche items chosen (even in games with finite loot rolls, like Necrodancer). It also lets players actually experience the "what ifs" that get pondered when the player is looking through the item list. In Necrodancer, what if I combined the Boots of Leaping (move 2 squares), a Ring of Courage (move 1 square when you kill enemies), and a Cat of Nine Tails (attack enemies as you move past)? The answer is probably unplayable, but what if it's not? EDIT: Although it reintroduces the problem(?) of finding a way to force players out of their comfort zones in build choice. It's not a one-size-fits-all solution, but none exist because some people really do enjoy every mechanic ever used. Even the most tedious example of spreadsheet management has a supporter somewhere. TheBlandName fucked around with this message at 16:17 on Jun 5, 2015 |
# ? Jun 5, 2015 16:11 |
|
TooMuchAbstraction posted:That is the very broad gist of it, yes. It gets considerably more complicated when you consider things like That sounds like it could be an amazing game. It also sounds like it would be insanely difficult to get right and take years of effort even with a team of devs behind it. Plus it would probably end up being a seriously niche product since the main audience would be at the intersection of Roguelike and Metroidvania players. That said, I would play it forever and ever and never leave the house again.
|
# ? Jun 5, 2015 16:53 |
|
Big Mad Drongo posted:insanely difficult to get right and take years of effort even with a team of devs behind it You don't say. (I actually liked AVWW, though it's definitely a flawed game.)
|
# ? Jun 5, 2015 20:37 |
|
|
# ? Apr 28, 2024 04:05 |
|
Big Mad Drongo posted:That sounds like it could be an amazing game. It also sounds like it would be insanely difficult to get right and take years of effort even with a team of devs behind it. Plus it would probably end up being a seriously niche product since the main audience would be at the intersection of Roguelike and Metroidvania players. I think it's definitely doable, but your design has to be heavily constrained in certain ways to make certain that you don't run into scope creep because it is very, very easy to overreach in this scenario. Chasm is trying to do something similar, and they recently had a Kickstarter update go out where they basically said "man this is harder than we thought it'd be, and getting all the design issues hashed out pushed out release date out by a year." It does have me wondering about the feasibility of a top-down Metroidvania roguelike, or alternatively a turn-based "platformer". Either way you'd heavily quantize movement, which would simplify large aspects of the design problem. A top-down approach also makes map generation simpler because you don't have to worry about gravity.
|
# ? Jun 5, 2015 21:05 |