|
Fangz posted:Missing an attack is certainly frustrating. That said maybe there's such a thing as "good frustration". This is already somewhat what I've planned in the project I'm currently sticking to, although accuracy concepts is something I'm still working on. My current calculations lean towards more guaranteed hit chances in general, as al-azad posted:Recent designs in RPG have made me lean towards “an attack should always do something” even if it’s as basic as missing an attack increasing the probability the next attack will hit. this is how I've been feeling over the last few years and is something I've also been thinking about. Just thinking of a reply to this topic helped consolidate some stuff floating around my head, so thanks for that. "Good frustration" can exist, particularly in the form of an unexpected turn of events that forces a bit of fun scrambling to recover your strategy. The problem is that it's such a small range to hit, and what's acceptable is different for every player. Although, I suppose it's also somewhat influenced by what kind of standard you set in the mechanics. If hit chances are high across the board, the hits that do miss feel worse. If they're low across the board, you notice it less, but it can make the flow of combat rather unrewarding or drag on too long. (There's also just a lot of perfectionists seeking optimal gameplay drawn to the strategy genre, but only they can make themselves happy.) The Rabbids XCOM games do recover from the latter somewhat with missed shots still contributing a little, whereas a single missed swing makes negative progress in an FE game as you eat unnecessary (and potentially fatal) counterattacks and lose weapon durability. The few bits of weapon skill gained are comparatively irrelevant. I may hate D&D's roughly 25% hit chance, but at least the momentum isn't so directly negative on a miss- it just drags fights on too long for me personally. Yet, the adversity is where a lot of the inherent fun in gambling is to begin with, and what is an accuracy system but gambling? Fae Tactics had very few enemies with evasion chances and the ones who did usually had massive numbers if not 100% guaranteed dodge. But it's also in a game where back attacks are a thing and guaranteed hits, as well as a buff you could take that guaranteed your hits as long as it lasted (and also ignored invisibility). But tactics games are frequently by design games with near guaranteed hits anyways. At the end, I guess it's a balancing act you just have to do based on the target audience.
|
#
?
Nov 9, 2022 16:41
|
|
|
|
| # ? Jun 3, 2024 17:19 |
|
|
Your Computer posted:I appreciate the input y'all but I'm just trying to make a simple little traditional roguelike work here, not re-invent turn-based combat drat because I just had a good idea but I don't know of a turn based game that uses it. I'm going to say it anyways and you can just ignore me. SoulsBorneRing games do this kind of thing by making bigger weapons slower and therefore naturally harder to hit with because you have to predict your enemy's movements. Obviously, that doesn't completely translate to a turn based game, but I think it would make sense if each weapon had a natural speed that gets modified by your character's stats and then that gives you some output of how long an attack will take to commit to. Many turn based games have a notion of time where turn order is based on some cooldown after doing an action. If your enemy's current cooldown is lower than the amount of time it will take you to attack, the enemy could attempt to dodge or block or even move out of the square, if they have room to. You could even base their degree of success on how much extra time they have. If they don't have enough time to recover from whatever they just did, you're pretty much just relying on your character's ability to consistently hit something of that size. I could see a system like that giving a lot of depth to combat. You could even get crazy with it and allow the player to do something like "the enemy has enough time to try to sidestep here and space to do it, so I'm actually going to attack the square I think they plan on moving to." But it would probably require a lot of tuning more than anything else, and that's the hardest part I think.
|
|
#
?
Nov 9, 2022 16:44
|
|
|
I like the cute old trick of using fake percentages and straight lying to players' faces. I can't find any of the old articles, but I'm sure most people have heard of it. A lot of old SNES games would roll a 1-100, then use that number to grab the actual chance from a table of predefined numbers. So if you have a 90% chance to hit, it's actually a 95% chance, and a 10% chance would actually be a 30% chance. In essence, it just bumps small number up a lot and large numbers up a little bit, which makes the RNG feel more accurate even though it's not really. On top of it feeling better, I also like how it punishes players for focusing too much on the math instead of just playing the game.
|
|
#
?
Nov 9, 2022 16:59
|
|
|
KillHour posted:But it would probably require a lot of tuning more than anything else, and that's the hardest part I think. it's kind of funny and i should probably have been more explicit about it but the very reason why i'm working on this project in the first place is to allow myself to just follow some tried and true designs and hopefully to learn something from them in the process. constantly i find myself falling into the trap of refusing to just take something that worked for hundreds of other games and instead wanting to do something "better" without actually having the design chops to follow it up. it's a very "game dev idealist" way of thinking and i feel like the thread goes there quite often, as in this discussion. i had another experience like this on discord where i wanted advice on how to implement stealth in the project (which in a traditional roguelike essentially means how many tiles away an enemy spots you) and instead i got a discussion on why games shouldn't shoehorn in stealth and how a game has to be built around it and various opinions on what stealth means to a game and alternatives to stealth systems. it's frustrating when all you want to do is make a little thing that works. i've butted my head up against this enough times now to recognize it's a weakness - my ambitions outspeed my abilities. i'd love to design things that are "just better" than conventional designs but i am simply not a good enough designer to make that happen, and i'm hoping that if i actually spend some time learning how and why those conventional designs work instead of just dismissing them, i might actually get better at designing.
|
|
#
?
Nov 9, 2022 19:06
|
|
|
Your Computer posted:yeah that's really the the thing with all of these ideas - i understand the desire to discuss ways of inventing new combat systems but all of this requires a tremendous amount of design consideration, tuning and thoughts to how it affects everything else in the game Just not using things that worked totally fine for other games always brings to mind Sting's Riviera and Yggdra Union. They really just went about trying to reinvent standards, and it worked in a weird, janky way. But yeah, your approach is great and it's absolutely a way to go forward. When you start to get knees deep in an idea, and the assumptions of what you think you know fade in the face of reality, learning is the only way forward. Following tried and true designs is initially what I did, copying Link's Awakening is what I do to learn any new engine, but I guess my ambition just doesn't want to hold back anymore. At this point, I'll just be happy if I finish something. A lot of folk like to play armchair game designer, and I'm calling myself out with that as well. I've not finished a project in the now 7 or so years I've futzed around with gamedev, the only "game" I can say I made was a rough, kinda wonky tabletop system. I guess growing up I tried to come up with games at times, but they were probably nonsense as I was like in elementary school. I just like to post what ideas I can think of when someone asks for advice, as it gets me thinking and maybe it gets others thinking. I do have to say, you make really fast progress with this stuff, you put together this roguelike engine drat fast, and the same goes for the projects before it I've seen.
|
|
#
?
Nov 9, 2022 19:34
|
|
|
Game theory is a lot easier subject to broach than game design because design is 50% implementation and 50% iteration. I can give you my thoughts on why I think the D&D style of to-hit roll vs. flat dodge isn't the most exciting idea in 2022 but even just messing with the formula is grounds for a deeper discussion on why it does/does not work.Your Computer posted:looking for advice on how to do hit chance calculation Approaching this from a blank slate because you don't like To-Hit vs. Static Armor, you have to decide how granular you want the calculations to be. If the player has to first hit and the enemy then evades it puts a higher focus on evasion which outweighs any bonus from to-hit and severely punishes inaccurate weapons. If evasion is factored in to-hit it still becomes a question of "damage per second" where a weapon that is 100% accurate on average outperforms a stronger weapon that is 50% accurate. This is partly why D&D evolved to more factors in weapon decision. You have slash/bash/pierce because some enemies are resistant. In earlier editions a dagger guaranteed the first strike while a two-handed axe would attack last. As it stands if the only factors are damage and accuracy, a smart player will always choose accuracy. e: I actually think small = accurate and heavy = inaccurate is the opposite approach. A dagger should be inaccurate but attack multiple times while an axe is highly accurate but only hits once. This lets you add cool modifiers that are weapon dependent and have more value to the player. Like which is more interesting Dagger: +5 to hit, 1d6 damage or Dagger: 2x attacks, 1d3 damage The second example even has a higher median damage which looks better on paper to the savvy player al-azad fucked around with this message at 21:17 on Nov 9, 2022 |
|
#
?
Nov 9, 2022 20:51
|
|
|
I don't know what I'm doing. But I'm still pleased I got basic interactions and dialogue running. 
|
|
#
?
Nov 9, 2022 22:31
|
|
|
https://twitter.com/byobattleship/status/1590463933173071873 Any ideas?
|
|
#
?
Nov 9, 2022 23:01
|
|
|
A framed portrait of admiral fisher
|
|
#
?
Nov 9, 2022 23:02
|
|
|
A scale model kit with several tubes of really dank glue and a brown paper bag. Seriously though, maybe a RC boat? Or one of those flag books. Sea flag language is hilarious.
|
|
#
?
Nov 9, 2022 23:16
|
|
|
anatomi posted:A scale model kit with several tubes of really dank glue and a brown paper bag. Ideally I want things that have broad appeal while still being somehow thematically related to the game. A cool scale boat would work great, which is why the first thing I thought of that I'm actually happy with is Space Battleship Yamato. It has the ship angle, but also the anime and sci-fi vibes help it be interesting to people who aren't bog-standard vanilla WWII nerds. So I'm looking for more stuff along those lines.
|
|
#
?
Nov 9, 2022 23:33
|
|
|
is anyone making a model of the dai gurren 
|
|
#
?
Nov 9, 2022 23:37
|
|
|
TooMuchAbstraction posted:Ideally I want things that have broad appeal while still being somehow thematically related to the game. A cool scale boat would work great, which is why the first thing I thought of that I'm actually happy with is Space Battleship Yamato. It has the ship angle, but also the anime and sci-fi vibes help it be interesting to people who aren't bog-standard vanilla WWII nerds. So I'm looking for more stuff along those lines. Along similar lines, I think it'd be pretty cool if you got a ridiculous ship or enemy from your game 3d printed and gave that away. Bonus points if you print it with a really cool filament or paint it up to look fancy. Obviously the bigger the better.
|
|
#
?
Nov 9, 2022 23:47
|
|
|
Tunicate posted:is anyone making a model of the dai gurren That thing always puts me in mind of, like, clown pants crossed with the pants that fishers use to wade out into the water. It's kinda dumb and not in a cool way. TIP posted:Along similar lines, I think it'd be pretty cool if you got a ridiculous ship or enemy from your game 3d printed and gave that away. Bonus points if you print it with a really cool filament or paint it up to look fancy. Obviously the bigger the better. That would be cool if I could get the quality good enough. I could easily imagine spending weeks going back and forth on the model and print quality though. It's a lot easier and safer if I can make use of existing products. I would like to do a super high-quality rendition of one or two of the game's bosses at some point though.
|
|
#
?
Nov 10, 2022 00:11
|
|
|
Do you have a ship in a bottle? It's a pretty easy gaffe, just need to scale up a bottle to ship size
|
|
#
?
Nov 10, 2022 02:26
|
|
|
leper khan posted:Do you have a ship in a bottle? It's a pretty easy gaffe, just need to scale up a bottle to ship size Are people going to look at a giveaway, see that one of the prizes is a ship in a bottle, and be more likely to participate? That's one of my problems: I don't really have insight into what kind of prizes will motivate people to get interested. Related: https://twitter.com/byobattleship/status/1590501128042852352
|
|
#
?
Nov 10, 2022 02:29
|
|
|
Me: I hate making menus and UI Also me: What about another UI feature??? https://twitter.com/Shoehead_art/status/1590695110001700864?s=20&t=NF7PUQjO-A2d4ThEEhIkBw
|
|
#
?
Nov 10, 2022 14:22
|
|
|
I like those sounds, they make me happy. Where will y'all go if Twitter implodes?
|
|
#
?
Nov 10, 2022 14:59
|
|
|
anatomi posted:I like those sounds, they make me happy. Where will y'all go if Twitter implodes? I've set up cohost and mastodon (mastodon.gamedev.place, specifically) accounts. It sucks to lose the followers I spent the last 3.5 years accumulating, but diversifying does make sense.
|
|
#
?
Nov 10, 2022 15:51
|
|
|
itch has some features like blogs. I like it a lot as a developer, but I have no idea if players actually care or view anything there. I hadn't looked at the analytics before now, I only have two simple created games there. The one thing I can say is people care more about Sega Genesis homebrew games than Playdate games.
|
|
#
?
Nov 10, 2022 16:12
|
|
|
Chainclaw posted:The one thing I can say is people care more about Sega Genesis homebrew games than Playdate games. I can't imagine why.
|
|
#
?
Nov 11, 2022 00:40
|
|
|
I redid the HUD on my game to fit the whole Gameboy style. Maybe I'm just a simple man, but this little zoomy effect on the action bar is like my favorite thing. Here's the old and new style for comparison. The old one was kind of half way there before.  
|
|
#
?
Nov 11, 2022 11:05
|
|
|
could you do those lego ripoff boat building sets? something like https://www.buildcobi.com/cobi-buil...p-viib-set-4828
|
|
#
?
Nov 11, 2022 15:00
|
|
|
Tummyache posted:I redid the HUD on my game to fit the whole Gameboy style. Maybe I'm just a simple man, but this little zoomy effect on the action bar is like my favorite thing. Oh yeah, this is good stuff.
|
|
#
?
Nov 11, 2022 15:36
|
|
|
organburner posted:could you do those lego ripoff boat building sets? something like https://www.buildcobi.com/cobi-buil...p-viib-set-4828 I asked my PR team about the various ideas people had come up with. To summarize their response: - Giving away stuff owned by other IPs can be OK so long as the IP is appropriate somehow. - However, models are a fairly niche interest. The goal here is to attract attention, so the prize should be something that gets eyeballs. - Replica weapons have worked well in the past when they are weapons from well-known IPs [here I'm guessing something like "Master Sword" or "gun from Halo"]. They're not sure that historical replicas will have the same punch -- might, might not.
|
|
#
?
Nov 11, 2022 15:37
|
|
|
TooMuchAbstraction posted:I asked my PR team about the various ideas people had come up with. To summarize their response: I figured not-lego would have more widespread appeal because even if you don't care about military model stuff you still get not-lego!
|
|
#
?
Nov 11, 2022 16:21
|
|
|
Hello, I'm still working on a landscape building game which I never post about, but here's a very exiting update since I've started working on the actual graphics this week. Up until now it's looked a bit like this:  Every tile is a big ol' square; they're boring and dumb!! Now:  I have to make one million transition modules between the various biomes instead, and every time I add a new biome the number of modules grows exponentially, which certainly won't bite me in the rear end later. I haven't gotten around to adding trees and stuff yet since I'm not sure how I want to do it, I just know I don't want them to be a part of the ground mesh like before. Maybe some sort of prop-system to decorate the various tiles. Anyway hope you all have a nice weekend! LoboFlex fucked around with this message at 18:03 on Nov 11, 2022 |
|
#
?
Nov 11, 2022 16:37
|
|
|
LoboFlex posted:Hello, I'm still working on a landscape building game which I never post about, but here's a very exiting update since I've started working on the actual graphics this week. Have you considered making a generic mask for the transitions and giving each biome a sort number? I don't think overdraw is going to be a huge concern here.
|
|
#
?
Nov 11, 2022 16:56
|
|
|
KillHour posted:Have you considered making a generic mask for the transitions and giving each biome a sort number? I don't think overdraw is going to be a huge concern here. There's a fallback system in place where if a tile doesn't exist for a given set of biomes it'll look for subsets that might have a tile and use those instead. I've only gotten it working where three biomes meet though, which is why there's a hole every time four biomes meet. I should also probably mention that everything is based on a system proposed by Oskar Stålberg where the tile meshes are on "the dual grid", which means they're all offset by half a unit along both axis, and mainly make up transitions between different types of biomes. I'm woefully incapable of explaining all of it, but here's one of his tweets on the topic: https://twitter.com/OskSta/status/1448248658865049605 His example deals with just land and water though, and I'm not sure if he ever suggests a better solution for dealing with more than two biomes, but I guess I've got something that sorta works. The tiles I've made thus far look like this:  The only actual transition tiles I've made are a few between grass/forest and grass/barren; the rest are handled by the fallback system.
|
|
#
?
Nov 11, 2022 18:00
|
|
|
The hole can be solved by giving each tile a "height" and the tile with the lowest height is a full square, and then the other tiles are drawn in order on top. That will guarantee no holes.
|
|
#
?
Nov 11, 2022 18:13
|
|
|
Oh poo poo, that's a great idea! Thanks!
|
|
#
?
Nov 11, 2022 18:17
|
|
|
I'm trying out pitchyagame, every few months they do a twitter thing that seems to get some eyeballs on games https://twitter.com/TipVfl/status/1591211695057563651?t=WLFAsIgSxBpIfuci1DiD4w&s=19 still 25 minutes left in this round if you want to quickly throw up a tweet with #pitchyagame in it
|
|
#
?
Nov 12, 2022 00:35
|
|
|
LoboFlex posted:There's a fallback system in place where if a tile doesn't exist for a given set of biomes it'll look for subsets that might have a tile and use those instead. I've only gotten it working where three biomes meet though, which is why there's a hole every time four biomes meet. If you put borders at the edges, you can easily put collision borders on tile borders and have characters walk right up to the edge of where the visible walkable region stops. Just do a tile walkability lookup. If you do middle transitions you have to do something more complex like a hidden 2x2 subgrid of walkability on each tile. Or do what he suggests which is have multiple different grids which are offset from each other, have a 'terrain' grid which autocalculates a half-step-offset display grid (rather than placing tiles directly) which seems like a nightmare. Tunicate fucked around with this message at 01:29 on Nov 12, 2022 |
|
#
?
Nov 12, 2022 01:22
|
|
|
Yeah, things like colliders or buildings or props would still be placed on the "normal" grid. The internal tilemap uses this grid as well; it's just the tile renderer which deals with the offset grid while assembling the visual mesh. If I ever get around to adding rivers then I suppose those would go on the offset grid as well, since it'd make the most sense to have them run along the edges of the tiles.
|
|
#
?
Nov 12, 2022 02:45
|
|
|
Shoehead posted:Me: I hate making menus and UI That’s WHY making menus and UI sucks, there’s always more you can do and while feature creep is easy to recognize with UI it’s “well this is quality of life, the game SHOULD have this it’ll be objectively better for it”
|
|
#
?
Nov 12, 2022 06:14
|
|
|
Making menus and UI sucks because we don't have good tools for it. IDK about Unreal, but I've been making uGUI UI since it releases and while I know all the esoteric tricks to get the expected UX behaviors purely out of experience it still takes me a dickfuckton of time to do so purely because purely because of how many steps basic UX behavior requires. It unironically takes me days to accomplish UI tasks in Unity that in webdev takes me hours. And this isn't even enjoyable dev work, it's just pure tedium. I was incredibly excited when Unity first released their new CSS style UI system... 5 years ago... And it's still not production ready.
|
|
#
?
Nov 13, 2022 05:03
|
|
|
Did we ever make a Trombone Champ thread? https://www.youtube.com/watch?v=aUJuwNxNUWQ Because Dragonforce made a video reaction about the mod for their song in Trombone Champ. Youtube is recommending videos with like 40k views from channels with 400 subscribers. To me that just screams overwhelming success.
|
|
#
?
Nov 13, 2022 05:28
|
|
|
SweetBro posted:Making menus and UI sucks because we don't have good tools for it. IDK about Unreal, but I've been making uGUI UI since it releases and while I know all the esoteric tricks to get the expected UX behaviors purely out of experience it still takes me a dickfuckton of time to do so purely because purely because of how many steps basic UX behavior requires. It unironically takes me days to accomplish UI tasks in Unity that in webdev takes me hours. And this isn't even enjoyable dev work, it's just pure tedium. I was incredibly excited when Unity first released their new CSS style UI system... 5 years ago... And it's still not production ready. i've only barely scratched the surface but i actually enjoyed the ui tools in godot........ and i say this as someone who vehemently hates ui work
|
|
#
?
Nov 13, 2022 05:38
|
|
|
Godot is nice, but I've already been burned hard on using non-battle tested engines in the past. I'm not touching that for production until the engine sees common adoption by at least mid-tier indie studios.
|
|
#
?
Nov 13, 2022 06:09
|
|
|
|
| # ? Jun 3, 2024 17:19 |
|
|
They've just announced a slew of updates for UIToolkit at Unite 2022. The editor flavor, of course, runtime stays behind 
|
|
#
?
Nov 13, 2022 08:44
|
|
