#Lancer Tactics
1 messages · Page 6 of 1
Full release was May '23
But d1 was modded in early access to iirc
I think it's more a difficulty in replicating vanilla art style/quality than inherent to 3d/2d necessarily
like minecraft is a 3d game but very easy to replicate
like I'd agree that replicating the sprite style is easier than the model style shown but replicating the current 3d map assets is easier than both imo
dd2 isn't just 3d models, it's much higher quality assets relatively speaking in general
Its much easier to slap together some pixel sprites than it is to do entire 3d models alongside their rigging, animation, etc
I say this as a person who has done a fuckload of pixel art
Dd1 still has a very particular art style most people cant replicate without skill, yet it has a fuckload of mods that introduce new classes with completely new art
In comparison dd2 has like, no new class mods, only reworks of the existing ones
I think i saw one mod that adds a new class, but it used existing game assets to do so (an enemy model with its animations)
yeah but my point is that's because the "entire 3d model" is a much higher quality asset
analagous to like a full live2d model
not a simple sprite
Okay but we dont have fucking live2d sprites on the docket for this game
and that there are 3d model workflows that are just as simple
3d models alongside their animations require more upfront work
Whereas 2d pixel sprite animations require you to make a few pictures in sequence
And let me repeat that the quality of the models being shown here isnt minecraft
You get more leeway in terms of what you can make since this is a mech game and its easier to make passable machine objects compared to humans
.
Im not talking about map assets, the entire conversation is focused on the mech models
Also i absolutely disagree with that statement that replicating even the simple environment assets is easier than making a sprite
Because a layperson can just open up mspaint and eventually doodle something passable, even just editing the existing sprites
I think that's unfair to the skill of the person who made the mech sprites, frankly
Whereas creating a 3d model requires downloading an entire 3d modeling program and learning how that works as an entire first step
Buddy i absolutely know how hard making proper sprites for things are. Ive done pixel art for more than a decade. This still doesnt change the fact that its easier to pick up pixel art and make something at least passable than it is to do 3d models especially to the quality of the proposed mech models
Like, thats one of pixel arts greatest strengths
yeah I'm disengaging
You can easily kitbash pixel art with like, no prior knowledge in art. Such cannot be said about 3D model, and I work on those, I know. Took me years to start delivering aceptable results. (And it took a few month to be familarise with the software and workflow and the 3D concept in the first place)
Something being easier does not inherently devalue it as art
This discussion has gotten more combative than I'm happy with.
The point about relative difficulty between styles has been made. Some stuff is easier, some stuff is harder.
my bad for my part in it
Carpenter and I have decided that LT is going to stick with 2D sprites for mechs in order to err on the side of being conservative about further style scope creep.
Perfectly valid decision, I remain enthused for the final product 🎉
whats with the snail react
haha yeah I meant to represent creeping with the snail
someone else did microscope and pinching to say small, I think
Just a teeny bit of (micro)scope creep

Lancer Tactics can have a little scope creep as a treat
it's already got one whole extra dimension!
Idea : Metavault that adds one or two dimensions to space ( this is a joke, don't take it seriously )
nah. removes a dimension.
But not the one you might expect
LT becomes a sidescroller
... ok but I'd play a Lancer sidescroller.
Blackbeard with Worms-style grappling
one of the pf2e aps does that for a fight
Bastions on a ledge with Assaults behind them that keep shooting you down when you jump up
smashing open crates to refuel my Jump Jets
hotswapping between Jump Jets and Type 2 Shield when the snipers show up
Hey the 3rd party license is open, anyone could do it
think it's this one (from a map remake project)
pretty sure it wraps around at the sides too, because you're on the side of a tower?
The dimension it adds is a narrative dimension, and the story transforms from a traditional narrative to a David Lynch project
I did attempt a 4d map at one point for my game
Does darkest dungeon count as 1D, gameplay-wise?
Ruin has come to our union.
there's a 2d map of the dungeon, but outside of that i think pretty much everything is 1D
What is a 2D map besides a series of 1D corridors
One time I ran a D&D encounter using a side-on map. It was very silly
When we think of it, what is everything but a net of OD points ?
Well for one you would need an infinite number of them, and that seems much harder to program.
Secondly, I'm not actually sure infinity is enough.
With a finite plane, you can get by using a finite number of corridors
It won't be truly continuous, but not gamespace is
But you're right that mathematically speaking, you'd need card(R) lines to form a 2D plane
Aka an infinity, but uncountable ( think of it as infinity, but bigger than some other infinities, one of them being the number of natural numbers )
I mean, I literally did a 1-D map for a game like this before. It was closer to chess than a full lancer implementation, but still; the game map was a single one dimensional array. A given position was at index y * map width + x, and a given index would represent y: (int)(index/width), x: index%width.
It makes adjacency a bit harder, but you don't need that for chess-like rules and in assembly this was easier than multidimentional arrays or any other more fully-featured data structure.
Oh yeah I'm doing that kind of mapping all over the place in LT. My current data task is actually figuring out how I want to store voxel data for the map blocks — a 3d grid mapped to a big 1D array works but it's like 21000 entries long
So I might split it up by layers or block type or something
What works best with the larger terrain pieces like the hangars?
You could simplify to just 2d, with a heighmap for initial ground level. Game objects could track their own altitude internally. Two mechs could stand on the same x,y if one is on the ground at altitude = world_height, and the other flying with altitude >= world_height + other_mech_size.
If you want destructible terrain, you can modify the 2d heightmap, and overhangs can just be ground-like objects with the same altitude logic (assuming you want to bother with such)
You don't have to track all voxels minecraft style, just object positions on a 2d grid.
probably not far off from what you've already got, I'd guess.
Ah this is only for like the ground Minecraft voxels. Things like hangars and mechs are handled in different arrays and tile systems
That's what I meant when I said this is still a 2d game under the hood; almost everything else still deals with 2d tiles instead of voxels
I thought about doing a simple heightmap for the blocks but that doesn't allow for having different block types under each other
There's no overhangs
It's not a bad idea if it's just the ground. Once you've got scatter on there it won't matter much
well math wise, a 1d map is infinitely narrow, so to make a finite plane you need an infinite amount.
Yeah, but programming-wise you can just make a grid
You're right that mathematically speaking, you need an uncountable infinity ( same as R's ) to form any plane out of 1D lines, but this isn't a situation where algebra is relevant
weird request but dfid anyone have that big godottutorial collection that got posted in here a while back?
got some friends who are doing a game jam and theyve decided to use that so i thought i might throw some resources at them
getting the 3D camera automatically responding to game events again has done some good things
Looking good!
This brought me back to good old Front Mission days. The graphic style is phenomenal ngl, and the movement is real good and so are the sfx
scoot scoot scoot scoot scoot
This looks so good! Olive, you're killing it with your attention to detail!
ooh that's so fun
Hell yes just played the demo I love this
advance wars music plays in my head
Also I adore the way the travel indicator hugs the terrain down the steps there for a second, when the morning cloak moves? Gorgeous.
this is looking great!
portrait maker is also coming along well
...oh my gosh they're CUTE
Oh look it's them, the HORUS Goblin frame
They look so adorable
I eagerly look forward to playing just for.the portrait creator
I'm so jazzed to see Martina's art in it. We should be getting a standalone public build of the portrait creator in the next few months
Looking really good!
lancer picrew :o
I know there's a TON of work left to do just to get mech gear in, but wow the progress so far has really blown me out of the water
👀 About to see a bunch of players and Gm using this to make a bunch of pc and npc portraits
(I know cus I would love to be one of them XD)
This might be me being overly optimistic but the hope is that investing this long in the underlying structure will make the work of mech content a lot easier... so hopefully that phase won't be too much of a chore.
I've certainly heard worse development strategies.
I always hear about not enough focus on underlying structure messing up a project so like
That seems smart
This is so cool.....
Oh wow Olive! Every section of this is looking spectacular
And it's real satisfying to hear the technical process of creating this and the work going into the software architecture
Spetacular, indeed!
I do kind of wonder, how would the Ultra AI be handled? I wonder if there's a way to customize it a bit. 🤔
How so?
Like if an Ultra has multiple weapons and Volley Module is selected, how does it pick which weapon? And et cetera.
Or if a boss would go after the lowest HP member, nearest, or random?
Like it'd be cool if we can customize a boss' attack pattern to give certain fights a feel.
Interesting. I'm definitely adding that on the dev side because it's necessary for tuning, but I hadn't thought of exposing that in a player-facing way.
That'll be lowish on the priority list most likely but noted.
Would there be speedrun support for LT? Curious about potential LT speedrun scene
Haven't looked into what that entails, but unless there's easy QOL pick-me-ups (e.g. dialogue autoskip?) I don't think it'll be a development focus.
Also, the anthology structure of the campaign modules won't lend themselves well to a linear path, so it'd probably end up looking like "fastest time to complete each one in any order"
Yeah I spoke about it to a narrative designer and taking a stance that focuses on making each campaign (likely one or two missions?) modular will set us up better to fold in player-created campaigns
So prioritizing flexibility over a canonical cinematic sequence
The great thing about the NPC tier system is that it's pretty dang easy to scale NPCs to whatever LL character you enter the campaign with
That makes me wonder, is there gonna be a set party size or is that flexible? Would opfor numbers change based on party size in the latter case?
I'm not certain about it yet, but a good reference for our model is BG3; it assumes a party size of 4, but you can play with less if you want to for some reason. I don't think the world scales down if you choose that.
But the current party size is an available number in the trigger system so there's nothing technically stopping a level maker from scaling like that
I should lay this out in a blog post, but we think we'll have three kinds of character slots:
- fixed — a premade companion lancer that is required to play a mission
- PC — only a player-made lancer
- flex — either a player-made lancer OR a selection from a palette of premade companions
The level designer can choose the available slots for each mission. For the demo, the tutorial would have been a single fixed slot for Granny, and the albatross mission would have been four flex slots
To make it easier on the writers and to make all-PC parties not miss out on story, a companion can talk and react to the mission whether or not you bring them; they radio in from their mech or from base.
you might wanna adjust how much each npc gains from each tier up then from vanilla imo, theres some npcs that have a linear progression in power while others get multiplicative increases
yeah everything is subject to change if we need to from core, but we're not gonna make predictive small tweaks like that. it's only a problem if it proves to be a problem.
I'll be honest, I never really understand why more modern/future set RPGs usually don't do this, it makes so much sense
it frustrates me as a player where if you don't intend to play a Mass effect or whatever 3 times you're going to miss out on stuff because weren't precog and realised mission X was going to have lots of Mordin stuff
just have him radio in with his comments!
With this all said, gosh, theres part of me that's considering the idea of adapting wallflower if the tools allow for it
do you plan to add eidolons?
not until everything else is done tbh, they're a can of worms. You should be able to get a lot of their custom rules working though with the normal trigger system ("when unit X gets below Y health, swap it out for unit group Z and raise the water level"), but stuff like ||the darkness layer|| would require special one-off mechanics that I don't want to have to think about until I have to.
#lancer-vtt message
Unrelated but I just wanted to share something really cool.
Oh that's so much fun
Spent some time today figuring out how I wanted to do glossary tooltips; I've not figured out how to do this sort of fancy in-text hover-for-the-definition subtooltip spawning you see in bg3 or rogue trader, so I'm going to try letting players tab through related concepts and keep it all in the same box instead:
Wrangling a single box is a lot simpler, anyway; it's a weirdly tricky problem to compactly arrange a bunch of variably-sized text boxes while keeping them onscreen without covering up whatever you're looking at
You don't see tabbed tooltips like that often. Can you have it so mousewheel them as well as tab through them? I prefer hotkeys as much as possible, but not everyone does
The UI is amazing btw, love the minimalistic
That looks amazing Olive.
I love the tabbable tooltips, and damn that UI is looking clean
I like it a lot! The nested experience is great, but this looks super clean too.
I actually want more greebles // visual identity in the UI, and have been thinking of this as very prototype-looking... good to know it's already doing ok lol
Oh that's a good thought! I'll try that. The thing that might sink that idea is if you're scrolling down in a list and mouse over something that pops this up, I wouldn't want the scrolling to be interrupted
For sure! The three challenges of making something like bg3's system are:
- the processing of text blocks to turn certain words into hoverable elements.
- laying out the boxes; sounds easy but can be tricky if you don't want to cover up certain parts of the screen
- figuring out the ux of how do you lock the previous tooltip so you can start mousing over specific parts of it
Think about the user inputs for that last one - it's hover over a tooltippable thing, lock the tooltip with a button press somehow, hover over the next thing in that tooltip, repeat
Whereas tabbing is just hover - tab - tab - tab
We will actually have one level of tooltip nesting because there's two kinds of tooltip: arbitrary lockable tooltips for things like weapon or system abilities (thing hovering over an ability in a hotbar) that you can tab to lock/unlock. If you mouse over stuff while it's locked, you get this tabbable glossary tooltip when you hover over stuff like range or damage.
It might make more sense with pictures 😅 I'll see if I can get a screenshot sometime today
It does feel a little odd that the tabs are the thing you’re hovering over itself + the things inside of it, feels like there should be a hierarchy there
here's that example of nested tooltips; hovering for an ability one, locking it, then hovering on that tip for (potentially tabbed) glossary terms.
and here's LT's soon-to-be-award-winning dialogue
(for some reason the translations aren't coming through, lol)
id vote for it
I like that. Could you have both methods; Tabbing through info, but also lock and look? More ways to get your information is rarely worse. Compared to similar games; I'd expect to be able to right-click on a button to get the locked info screen, then right click anywhere to clear it. Mouseover and tab for info appeals a lot to my love of hotkeys
slotcanyon.intro.100.cloak
truly, some of the most impactful writing of the modern age
this is looking so slick!
Horus tooltip that just crashes your game
Horus tooltip that deletes your save
Horus tooltip that gives you hrt
The possibilities are endless
This is perfect. I need to make my own one of these for my own godot rpg shenanigans.
im just going to say slotcanyon at people now
oh wait slotcanyon sounds like a vague insult
perfect
"quit being a slotcanyon.intro"
Fantastic, thank you so much
Exciting news: I spent today mechanically finishing the full Mourning Cloak license. All weapons and effects are mechanically ingame — e.g. bonus damage from variable sword and hunter trait, invis and terrify from the hunter suite, teleportation from the CP/fold knife/singularity motivator. It still needs graphics and testing, of course (and the Intangible status doesn't do anything yet) but doing a license dive has come back very promising for how easy it is to add new content to the engine.
One (sorta) down, 27 PC mechs to go.
Exciting!

Selective hearing: "I'm almost done"
Dev hearing: "And then there's the NPC classes, tiers and templates" *scream
Glad to hear it's easy!
Ok, genuinly curious, what do you think will be the hardest license to import?
Going to guess Goblin with all its wierd hacks
That's what I'd guess too, though Hydra could be spicy
I anticipate goblin's CP being a bigger problem than its hacks... though I need to solve the same problem for the mule harness
stuff being unique isn't too bad if the complexity is entirely within its one script. like duat will take some doing but all its state and stuff should be able to be mostly handled by the action itself
Is "units sharing a space" also an issue for deployables like the omnibus plate?
ye
mostly for figuring out what thing to target when you target that space
so a UI problem; stuff will have to handle waiting for the player to choose instead of just querying the map
it would be fun to make a betting pool, but I don't know the metric
Is "Units attacking their own space" any harder? It'd come up for jockeying and the webjaw snare
(The extremely common use case of Goblin on Sehkmet I imagine is already covered by the AI)
Page 246’s misspelling of “Near-Thread Denial System” is making me giggle.
Not intending to sound rude
Wrong channel? 🤨
Huh, the kickstarter update doesn't seem to have gifs for me
Yeah I just saw as well. KS changed their system and it seems to have worked in the preview but not published version ☠️
It's mirrored at https://wick.itch.io/lancer-tactics/devlog/682706/3d-map-editor-more-ui-and-mourning-cloak-license
That's a night-and-day difference, haha
The gifs worked in the email though? Wacky
Combat popcorn is a great term!
Does it become a Döner Kebab if the lines are rounded off?
Working in IT I've said to confusion enough times "Click the elipsis." Meatball menu forever
The gifs for these updates never load right on Tumblr. But only the gifs for these updates on Tumblr. Gifs elsewhere? Fine. Other things on Tumblr? Fine. These animations specifically on Tumblr specifically? Nope, they just load forever.
I assume the super condensed version of the updates is "doin' fine 👍," by the way.
XD yeah lol
I do not know enough about any of this to have anything to say other than "hooray, keep up the good work." So hooray, keep up the good work.
hm interesting. I wonder if I should optimize them to get em smaller or something for tumblr? I'm looking at it in an incognito window and some are working and others are not
I have no idea why they keep breaking on Tumblr. But it's Tumblr, I don't expect anything to work right anymore.
Best I've got is a stopgap "here's a link to the itch mirror of the update, where we know things work" at the start of the Tumblr version. ¯_(ツ)_/¯
i'm surprised you actually get them all into a post on tumblr without tearing your hair out, uploading images there is a nightmare for me
yeah... the news that its devs are being downsized to a skeleton crew is not the worst when it comes to the enshittification process (see: them giving up on tumblr Live) because there's less of a financial incentive to grow, but is obviously worrying in terms of maintenance and basic functionality.
I was also surprised, I kept expecting to hit an upload limit. It was like 50mb of gifs
I think it would be extremely funny if "oh thank whatever that they're going to stop breaking everything when they ~improve~ the site" made the value go back up.
I mean, really, when's the last time "hey we're adding a bunch of NEW FEATURES" wasn't met with groans of disgust and wails of despair?
My gf says that the ads on it keep getting shadier and shadier, which I imagine is another downside
i mean more to the effect of "if i try to add multiple images to a post at once, ever, all but one of them will fuck up. every time" and have to agonizingly do it one at a time until the little loading ellipsis goes away
holy shit olive this update looks so cool!
Thanks Kat! 😊
Every time it feels like we're moving so slow, but then these updates are like oh huh ok I guess we are making progress
Me whenever olive drops a new update, and I'm astounded at the progress (This happens every time)
(this is my 26th broken phone)
Personally i think you should do a taco menu
No clue if that exists
I just like tacos
That’s where you try to open one sub menu but all the sub menus spill onto the screen at once
Have posted in a while but saw the latest update and wanted to pop into to say you’re doing a fantastic job, Olive! 
Gotta say, it's looking impressive, and glad to hear putting in an entire frame 'only' took a day!
I'm honestly excited with how fast the development of it is going! I can't wait to try this out with some friends.
No more slotcanyon.intro 😔
I would have also been satisfied with Toasts or Chips
combat popcorn my beloved
shoulda put them at the top then lol
anyway, had a UI idea for how to handle Goblin core and Mule Harness: disappear the sprite then add a little floating window over the "host" with a portrait
somethng like this
Doesn't handle PCP though, best I can think of to cover both is a small popup list after selecting a space so you can select from items that occupy the same space. Feels a little cumbersome but as long as "the ground" isn't one of those options and it hapens all the time, it should come up irregularly enough
I can't actually think of any NPC side flat deployables like PCP or effects like Goblin where they share a space, so maybe it's a non-issue
oh that's not a bad UI solution. The thing I was really fearing though is the engine side of having multiple valid targets in the same space — I've taken lots of shortcuts throughout the code saying "give me the unit (singular) at this space, favoring mechs over deployables" and that's been good enough.
Like if you click on a tile to choose which unit to activate next, there'll now have to be a special side-flow to see if there's multiple potential guys at that tile in case it's a mule harness stack.
and that'll have to happen everywhere anything ever wants to target a tile for any reason
it's fine, it'll just be annoying to have to add because of one or two abilities
would something like coding mule harness and goblin core as the passengers existing in a separate zone, and they're targetable whenever the base unit is targetable work?
that's even more shortcut that might be better solved by allowing the engine to handle shared space though, I think
If it's any consolation, it also applies to the Atlas and Forge-2 emplacement for a total of 4 abilities
You could make it so you activate the attached unit by clicking the portrait rather than the tile, if you went with something like the image? Then you don't need to add anything to the space select logic, just a UI element. Would probably need to be bigger than the example but
Again, nice solution for the player side, but the space-target logic needs to work for the AI as well
This probably won’t work/cause more issues, but would it be possible to make it, code wise, have the attached goblin share the same ‘target’ code so when an enemy attacks, it mirrors whatever is done to the main target…. If that makes sense?
I think we're getting a bit into the weeds — I don't think the solution will be hard, it'll just end up being an annoying refactor when we get to it.
And I dunno what the specifics will be until then
🙂 will probably make a good tidbit for a devlog
The earlier a change the cheaper it is in dev time. But sometimes you just have to wait and see what you need later on because other things are going to change too.
Always enjoy a devlog!
what's the asset pack used for the mechs? Trying to use it for my table
Doing a batch of late campaign sticker/coin orders today, and I just found someone who exploited the small tier/add-on loophole to get like a five dollar discount and then added back a dollar tip. People are strange
Maybe they were in it for the thrill of the exploit
Maybe they think kickstarter takes a smaller amount of tips
can you ask them
Such strange behaviour for such relatively small amounts of money
https://lancer-tactics.backerkit.com It's not playable yet though. Last update said estimated 3 months before backers get super-early-access.
Noice!
Been having a pretty slow week, but today I added x-ray shader for seeing units and movement lines while they're obscured by terrain.
Looks effective
ooh it looks gorgeous
I was just expecting basic outline but it's got this cool vfx
oh and it works with arrow too?
Are the black lines showing obstructed LOS?
oh god fuckin bless
godotshaders.com is a godsend, they have so many good vfx there: https://godotshaders.com/shader/x-ray-vision-effect/
That looks so much better then a simple outline
Damn that looks sick!
Yesssss that looks so good!
Gonna need to save that link for later. Looks like a good resource
i like the digital effect a lot. To the point that I'd kind of like to see some subtle effects throughout? Kind of give a little "we're actually playing on the commanders holo display" diagetic effect. Almost certainly awful for readability but hey
literally anything that implies a menu is in universe I will fall over for
This is a BIG thing carpenter and I have been trying to think through for the last few weeks
Where do we draw the line? Are character sheets diagetic too? If so, what are they in-universe?
COMP/CON strikes a nice voice of being a sorta "you're in a faction-themed computer UI's registration software"
but do we really want to take the stance that the player is a commander looking at a holo-map sending orders to allies? If we're making a general-purpose Lancer engine, I don't know if that's always the narrative positioning we want
The more generic it is the more situations it fits, but you lose that excellent diagetic flavor that comes from specificity
The balance we decide to strike will dictate a LOT of artistic decisions; it was spurred in this case by our musician asking us "so... what sort of space is the player in during downtime?"
We haven't come to a strong decision yet.
and we can kick a lot of cans down the road, most UI theming can come as a late layer of polish
Maybe make it something that can be modded? Like, have one thing that you work on and is the 'main' UI, but have the option for people to make their own UI or make 2 or three of your own?
But that would probably be alot of work
I'd personally lean towards diagetic but excited for whatever you go for
Do what Xcom did and give the player a cool psychic commando for the final mission of Lancer Tactics 2
I really like the UI style, and one thing that could be added, since its a feature in the COMP/CON too, is having different UIs the player can choose from, one themed after each manufacter
Maybe even changing it depending to wich character youve currently selected (this one might be complex in terms of visual clarity as it could confuse the player, especially if the layout changes)
I mean starcraft did that and it ruled
but also "I don't want to make a decision, I'll implement them all and leave them as options" is in many cases the worst decision you can make scope-wise
Tv Tropes actually has a cool list with examples
plus that doesn't solve non-UI stuff like music being tailored to specific environments
oh that's very cool
From what i heard in the demo, the music slaps
it'll be very different than the demo!
we're working with a new artist. I think it needs to be more mellow, the demo is kinda intense
more into-the-breach-y
Into the breach is an amazing example for games like this
Mechanically and artistically too
Its just a great game
YESSSSSS
In other news, I got overwatch working today!
Oh thats sick!
yahoo
Btw little question me and a friend i showed the project to were wondering, will this game be usable as some sort of VTT for lancer, where a GM can like make a lvl with custom stuff and all, invite his players and have everything playable?
nah multiplayer is too big a swing for me to try and tackle, unfortunately
the best we might get is local multiplayer
and if we manage that, it might be possible to set up some sort of remote-play thing
Steam has remote play built in
This has been stuck in my head since it was raised. Online multiplayer doesn't look significantly harder than any other part of game dev. https://godotengine.org/article/multiplayer-in-godot-4-0-scene-replication/
But also multiplayer wasn't a promised feature. I think if you add multiplayer you should take a swing at networked connections, but... do it later, once more of the game is done. Or maybe even after 1.0 if you have the budget for it.
If you're going to add multiplayer to a game typically the times to do it are "as you are starting development" or "never", waiting until a game is done is asking for trouble
As the game has already started development with no multiplayer planned, retooling it to have multiplayer would inevitably be a large effort because you need to shift how so many things are handled
I think it's probably a lot harder than it looks from the outside
I think the best that can be swung for is “remote-play friendly local multiplayer”, which I’m sure is still non-trivial
At any rate, the “Nice to have’s” pile is tall and there’s stuff which much higher demand on it
Ultimately its up to olive, it wasnt a kickstarter stretch goal and theres already a ton to do
"Online multiplayer doesn't seem hard to add" seems like one of those things that game devs hear moments before the sort of disaster you read a tell-all about years later
A good rule of thumb in my experience is that if you think a thing is easy to add to a game, it probably isn't
Yeah, online is ALOT. There is a reason why when rollback is added to a game later in life, it takes ALOT of time and money
Any given game engine has a lot of features to help support multiplayer. The problem is that while it often seems simple at a basic level, in practice as things get bigger and more complicated, getting things to work consistently across servers and clients becomes more difficult, depending on what you’re trying to do
I can’t speak for this game or godot, but as an example I’ve dealt with working with Unreal, maps (lists of key-value pairs, aka dictionaries) just don’t replicate. Any client-facing functionality that depends on data you store in a map on the server? You have to get creative in how you get that data on the client in a way that takes up minimal bandwidth. And that’s gonna change on a case by case basis
if you weren't already thinking about it when you started a project, chances are that many of the solutions you have made to various game systems are going to turn out to be fairly unsuitable for networked gameplay where you need to start caring about client/server communication
I mean, multiplayer is easy. When its your friends turn, hand them the keyboard
Play by email
Competive mode, you must wrestle over the ability to take a turn
vordt of the boreal valley plays
It's this exactly. Getting a multiplayer demo up isn't too bad, but having async comms between host and client put a lot of constraints on how you can structure things and I think ends up having a multiplicative effect in the work required across the whole game. It's notoriously one of the thorniest features in game dev, and why so many wannabe MMOs crash and burn
"Play by mail" mode where it's possible to, like, send over a code of the move taken and its outcome, similar to chess notation, so the other person can synchronize game state, would be pretty funny.
Modders adding multiplayer to Skyrim was a feat of heroes
Holy crap, I want to play this game while arm wrestling now
That was sorcery and you can’t convince me otherwise
That said, I'm cleaving closely to best practices in keeping view separate from game logic and it's already saved us once in the switch to 3D so maybe it might nevermind I need to stick to a strict policy of keeping multiplayer expectations at zero
Good move. You've got tons on your plate already.
Well you can still coop by passing controlable mechs around
Yeah I'm feeling whelmed these last few weeks in trying to tie things together into the first EA demo. Lots of boring stuff like getting resizable fonts working and adding game settings to persistently adjust audio levels // organizing sfx into fmod channels so they can be independently adjusted, etc
The plumbing that doesn't matter until you want to hand it off to someone
yeah there have been a number of fan mod projects to add multiplayer to singleplayer games and in general every time you read about them it sounds like an absolute nightmare
this is why you hear about in so many game retrospectives about how the multiplayer part was handed off to someone entirely separate, 'cause that shit's so much extra work.
you also hear a lot of "the c-suite demanded we add multiplayer and it fucked everything up"
This happened to Mass Effect 3 (which somehow magically ended up not sucking to everyone's great surprise) and Spec Ops: the Line (which was just sort of there) among other things
I think "a strict policy of keeping expectations for X at zero" is one of the smartest things a game dev can do in terms of workflow planning
it's still wild to me that 1) me3's multiplayer was good and 2) that i miss it
Really caught lightning in a bottle with how me3 multiplayer turned out
a.k.a. the old "how to end up on Wha Happun speedrun strat"
I once got host migration for a hololense project working once
Unless you walked too far away
Even if the connection was perfect
If you moved too far away in the real world it couldn't keep the assets synced
Never again
Code notwithstanding, I always see the suffering of devs handling off turn reactions in turn based adaptions and lancer has lots, and that sort of thing gets worse in multiplayer
Like high level bg3 is an ocean of "do you want to force a reroll" on every single attack. Then have it happen for every player. Then have it happen when a player stepped away because it wasn't their turn.
And now everyone is waiting and you probably have to implement a timer, which end adds an unpleasant pressure all the rest of the time
A nightmare
I play bloodbowl. "Do you want to take this interupt" type things have like.. 5-10 second? timers, happen regularly enough you need to pay attention, but rarely enough that your attention can wander. It's a long enough pause that the other player gets frustrated at it interrupting their turn and also completely screws you if you dont make the choice.
All interactions that flow nicely in a table game.
Three days late, but the major thing I remember about the multiplayer of Spec Ops: The Line is that the devs straight up called it a tumor on the disc.
Idea for multiplayer: whenever you lose, the other person can now play as Luigi until they lose.
i support this because i wanna see luigi goomba stomp a goblin
Aaa!! I didn't realise I'd also be getting stickers after the initial campaign! Got some things for my laptop now.
today was a "have fun with shaders" day, vis a vis working out how to represent the "off map" area. I don't want it to be an Into The Breach puzzle box; I'd like to have it feel like it connects to the outside world, but also I do't want to have to deal with off-map actual terrain. A weird shifting texture seems like an OK compromise.
looks nice! the little touches you're adding is gonna make the game look so gorgeous
Looking fine
The edge just going to be a solid wall, no Icon-style insta-death modifiers?
a literal "fog of war" 😎
you could try like, blurring/fuzzing/dithering the map edges to blend it into the outside map texture too?
Oh I like that idea! I'll try it
mmm I'm not convinced it's better than keeping a clean break
adding alpha blending isn't much better
(compared to this clean break)
but I like the concept of integrating the edge with the map... maybe outward is the solution rather than inward, like leeching watercolors outwards. That sounds hard enough that it's probably a final-polish step though.
Having the edges bleed into the map is artsy. Having the hard edge is more gamey. I think it's OK for games to be games - I'd go for hard edges
maybe one cool thing that could be done is adding a little text barrier that appears saying something like "out of radar", kinda like in armored core 6 where if you get too close the edge of the map you see it
a thought - there was some discussion previously about how diagetic you want the UI to be - if you were to end up going in a diagetic direction, you could do some kind of digital effect, like the computer is bleeding out the colour or fading to like, digital noise
Yeah! That's a little what I was going for with the jerky pixel-y quality of the clouds. It's probably on the subtle side, but I think that's a really promising aesthetic approach for this.
The hard edges look a little better imo
Yeah was thinking maybe give the outside a staticky look or fading into plain grid or ‘outside of sensor range/battle field’?
got ramps working! the plan is to have odd ramp shapes later, but they're not mechanically different than these basic ones so that gets kicked down the road
I have watched this gif on loop a good 10 times trying to absorb every bit of info
Things are really ramping up
whichever backer has been pressing whatever button that sends me this message has just been making me spiteful about writing the next update. They're monthly because that's the pace that works for me. /rant
I haven't seen that particular popup, but Kickstarter sends something similar when there's been comments on your project and it's enormously stressful. It's not just you
Ah I like seeing the comments one, it's nice when people have stuff to say themselves
Yeah that seems bloody rude of them.
why is it a big red banner and not a small notification icon 💀
I can understand why they'd make it a big red banner, and I can also understand why it'd be annoying for it to be a big red banner.
What if it was a big green banner?
If it's something that can be triggered at you by any one of thousands of random internet people with zero cooldown, it shouldn't be anything.
It's not clear to me what exactly cooldown it has. I've gotten one every few weeks for the last month or two.
the "start writing" button really annoys me because it sounds like a command more than an invitation
can it be turned off? that is just ridiculous
Yeah, I'm sure they put "start writing" there with the idea that it makes it easy. We'll reduce friction! We'll close the loop between creator and backer!
In practice I'm 100% with you. The message my brain would see is, "stop slacking and procrastinating, get to work! 😠"
Can you get rid of it using something like uBlock?
Just post an image of your git history /j
posts one screenshot
"stop asking for updates or this is what you're gonna get, see you at the end of the month"
"Every time you ask for updates another mech License is removed from the scope. Now silence"
I want to read the intention in good faith; that someone is excited for the updates and wants to send encouragement, not knowing that kickstarter's notif for it is so odious
"The end user doesn't know how odious the platform is" is a safe bet in 2024.
Eyyyy check out these isometric friendos! https://www.kickstarter.com/projects/wickworks/lancer-tactics/posts/4063845
They got different facings!
heck yeah they do!!
we can get away with just having a front-and-back by not caring that their sides flip when we mirror them
oooooooh!
The games version of carcinization is getting closer and closer to Minecraft. Love removing and adding blocks.
just gonna go full on deployables to play as much minecraft as possible
In the end, everything is Minecraft
Kobold/hydra for maximum mining and crafting
SEISMIC RIPPER KYAAA
extremely aggressive ear ringing
Fun fact: it's bad accessibility to have "ear ringing" sfx in a game; it can set off tinnitus much the same way flashing lights can set off a seizure.
love the color scheme, is this also like a preview of mech art customization?
They're the working colors for showing the different customizable areas. I think the current approach carpenter and gen are working with is having 3 modifiable color channels: the primary modifies the grey, secondary the red, and then a tertiary for energy or highlights like the tips of the goblin spines.
we haven't actually implemented any color-customization for them yet, but it'll use the exact same recolor-shader tech as we do for the pilot portraits
HOLY SHIT Y'ALL GOT GEN FOR THIS?
Oh my god
perfect casting for doing mech sprite for this
this is exciting news, genuinely
You're gonna be happy to know that Winter Scar also has those in a few places
i would love to use these arts in my actual games
omg are the NPCs getting same treatment by any chance
hm I wonder if Gen would be interested in releasing em as a noncommercial asset pack like musicians do for game OSTs
I'll make a note to think about that far down the line
Yooooooo, I love Gen's art, they're a great fit for this
I just hope they get to flex their muscles and do some larger pieces as well, those are amazing as well
oh i was just thinking png export like the character art
I think that was one of the features, right? or am i misremembering
would be so cool if in the future the game got support for custom mech sprites, imagine using your own retrograde minis
i think it's not in the scope bc pixel art doesn't compress well, among other reasons
Sprites! Geometry! Rotation! It's looking fantastic.
Fantastic update
Flying in LT is a status that increases your effective elevation and lets you ignore terrain/climbing movement costs
I'm currently considering collapsing hovering and flying into one thing to simplify it
I feel like at my table we usually forget about stuff like the "fly only in straight lines" rule
Mayyybe just have the difference be that some mechs have an extra buff that keeps them from falling when immobilized
It does feel like an indirect nerf to the Dusk Wing, since it rides or dies on being the only mech with hover
Dusk Wing is the only frame in the game that can fly by default, so I'd say it makes out just fine tbh
Yeah it’s also still heat free flight
Crack Shot 1 might be a little buffed with that hover change? Hover is still either rare or a big investment though
And the duskwings I’ve played with didn’t like to sit still anyway
Oh looks like I misread/remembered the hover rule. The only difference besides the straight line thing is that non-hovering characters have to shuffle around a bit on their turn to stay flying?
That seems doable.
Yeah if you don't fly on your turn, you fall
I do want to use this to just like... build wallflower
Now you’ve got me thinking the same thing
Hey, I don't know if someon already asked, but is there any plans to make a multiplayer version? 0_0
multiplayer is out of scope
I'm curious how hard it'd be to hack it in.
Given that it's out of scope for the original developers, I'm going to guess the answer is "very."
It has been done for other games before, a fact that continues to baffle us, but.
Yes, the answer would be very we think
Depends how you do it. it can range from very easy (you physically hand the second player the keyboard) to extremely hard (large servers)
with most options being closer to the "very hard" side of the spectrum.
however as stated, there isn't any plan to implement it for this game 😅
I am an experienced (and have been paid professionally) multiplayer game dev. It is a nightmarish endeavour that colours every inch of the development pipeline and sours everything it touches and I do not wish it on anyone.
“Your game is multiplayer now” is one of the most cursed spells to be banned by the wizard high council. “My game is multiplayer now” is not banned, but it does incur a terrible price on the caster
i keep petitioning the council to additionally ban "arbitrary multiplayer focused franchise installment" but they refuse my letters on account of "not being a magician"
I still recall the tale of the multiplayer stuff for Spec Ops: The Line being outright called "a tumor on the disc."
Ty! I found the game and was curious about that. I decided to back up the project anyway after played the demo. I think the soloplay experience solve the time "time problem" of lancer so well that even the stuns are not frustrating like in the TTRPG version. Very excited to see the beta 🙂
Do we think that size 1/2 units should be able to reach up and hit another standing on size 1 elevation with a threat 1 weapon?
They count as height 0 in terms of line-of-sight so they can't see over size 1 things (bc LOS is reciprocal and size 1 objects should obscure them), so it seems a little strange that they'd count as height 1 for things like threat or engagement.
I also think the image of a napoleon fruitlessly waving its sword at an enemy on slightly higher ground is funny
The way I see it, a size 1/2 still occupies one space, and threat is drawn from the edge of the space, and a mech that's on top of size elevation is still within 1 space of that space, just a bit diagonally, so yeah
My gut thought is that 0.5+1 = 1.5, which still puts the reach high enough to get to the top of the size 1
chill thanks for the sanity check
I don't think we need to bully Napoleon any further.
I guess the "size 1/2 units are worse at engagement" angle is already covered by the normal engagement rules
is their base still size 1 like normal? for edge to edge calculation purposes
yeah "size 1/2" is implemented by being a normal size 1 unit but with an extra boolean that downgrades their height when checking LOS or causes them to be treated differently under specific circumstances like comparing unit sizes for engagement
plus their sprites are a little smaller
I was wonder if that or "Size 0.5 in the backend but with special exceptions for some things" was the easier option to program
Yeah keeping the size stat as an integer is waaaay simpler
okay i'm actually really curious
Is lancer tactics going to handle grapples?
It just hit me that grapples are weird
I just had a several hours meeting with Carpenter talking about how we're gonna handle them. I thiiiink we have a good theoretical base for them, but I haven't started implementation. We need a general attached-units system for MULE/goblin CP/Lancaster tether cables anyway.
It'll get a bit hairy when there's like a chain of units grappling eachother and moving around like a big piece of adhesive flotsam, but I think technically viable
It's also an open question how to figure out what are valid move tiles when grappling. You shouldn't be able to walk someone else through a wall, but you should be able to walk them off a cliff
If anyone ends up in an illegal space, the game will have to kind of shunt them somewhere else. I suspect this will be an abusable surface area for speedrunners
And what about if you're moving through a friendly unit's space and then lose mobility?
That reminds me, are you expecting Size 2+, which I have seen a conspicuous lack of in promotional materials, to be a big headache wrt illegal spacing?
Same illegal space shunting mechanism
Yeah they're handled mostly already but are a continual challenge
jericho cover deployables show up as 3D meshes & you can stand on them. the carcinization towards minecraft continues apace.
that's so slick
looking good so far!
cant wait for a public test, but i actually CAN wait tho
take your time
cooool
Cute bug I just found: after terrain destruction, all units (incl. deployables) recheck the ground height under them so they'll move down if the ground is shot out from under them. When the Jerichos do that, they say "oh heck, the ground is taller! I better move up to stand on it!" — not realizing that the taller ground they're seeing came from themselves.
Literally pull themselves up by their own bootstraps
Ahh the classic "Ah wait it's self inclusive"
Great job, Olive! The game seems beautiful so far
Infinite height glitch
*stacks cover on itself for height advantage
All these previews are looking gorgeous!! Your efforts are really paying off. :]
with the carcinization towards minecraft is it possible to make a tower of mines and other deployables? is there a maximum height?
Alas, no, we're not doing stacking multiple units; the way I fixed the above situation is by having units/deployables that add height always stick themselves to the ground, ignoring all height-adding effects (including their own). Since most of the engine continues to be a 2D game under the hood, there's not a good way to track which unit should be lower on a stack — units don't remember their z-level. Whenever it's needed, it's calculated from the current ground level + height-adding-effects + if they're flying.
There's currently a maximum ground height of like 6 blocks to make room for large setpieces like the hangar going even higher up but carpenter has petitioned that I increase that substantially. There's a performance cost of doing up-front line-of-sight calculations for more layers, but I think we're still within reasonable bounds for that.
how do y'all plan to handle line of sight and cover with height and flying?
I vaguely remember you mentioning that flying will be handled differently but if we're getting terrain height of up to 6, I feel like that may change some things
Have you put any thought into the Kobold's trait and cp? I know it's not core book content, but I do remember it being one of the demo frames
The cover calculator does a center-to-center raycast looking for blocks that give hard cover and adjudicates based on the position of intersections (adjacent blocks can give hard cover, further ones are just soft).
It's really friendly in comparison to the prickly LOS system, which is still basically the same as the 2D system but with more layers. When I need to see if one tile can see another, I check sight clearance using the higher of the two tiles. This means no overhangs/bridges. And flying works the same as if they were just standing on a higher block.
Not really... but I think we could fake stacking blocks by making special versions of the slag deployables that are taller if you target the same tile multiple times during the CP
testing out a topdown mode to fix not being able to reach units nestled between two tall buildings
I spy both Gen NPCs and the Drake 👀
(though I'm pretty sure the NPCs were also in that last gif)
oh this doesn't look nearly as cursed as i thought
huge fan of the how flying is representated here
Yeah, this looks great!
It is a bit hard to tell things are flying in top down. Maybe there's a little note or modifier you could show next to the mech in that case?
That's a good point. Carpenter was talking about maybe going full straight-down so the elevation edges disappear too and give that information with colors or borders or something... I think the ideal would be something like how Homeworld has a blue-schematic "tactical" view that you can switch to.
Sigh lol we're back to making a 2D game, pack it up
if you're still doing the "flying as a status" thing, you could put an icon on it, right?
put a tiny little goku cloud under the feet of the flying sprite
this is only like 45% a joke
there's always the standby of 'flying units get an obvious dropshadow'
It's kinda like 2.5D, but not. 3-0.5D
Like Kingdom Hearts
i realize this may have come off as like, idk an attack
I meant that I expected any sort of angle change to look super cursed but how you did it looks very good
all good, I had similar fears initially
Maybe flying units need an effect like jet wash?
also not a bad idea!
I'm rotating the sprites in my head to imagine how the non-flight units would look like soaring through the air
I know you are doing flying as a status, but I'm interested in what that means mechanically?
I guess in the sense of like, can you melee a flyer, kind of thing
it means they're at +3 elevation above the ground
so to melee them you'd have to have enough threat or like be standing on a cliff high enough that's next to the flier
or be a melee barb lol
they also ignore all extra movement costs from the ground // elevation changes
To fly is to not be ground
@alpine sorrel how do yall plan on handling the sprites for the core bonuses that change your size?
Goooooood question, we've talked about it and we're gonna see what looks best when we get to it — either resizing the sprite (possibly ugly-ly) or just giving it more spaces to take up without changing the graphic.
If a straight x2 scale fits we'd probably do that, but it might not work for the size 2->3 change
Size half->one probably has no visual change tbh
yeah resizing sprites is a big ew
tbh
the why would you do this option is to make new sprites for every possible player size up to 4 lmao
casually quadruple the number of sprites why not
Wow I've never actually contemplated a size 4 barb
so fucking unwieldy lmao
the few size 4+ homebrews ive seen all have some trait that lets them circumvent the problems of moving as size 4
like being a big nano slime blob or flying or whatever
Pixel art is such a pain to work with. I don't know why we're still doing it
Gamemaker's toolkit just posted a video where he learned this firsthand once it moved beyond a tech demo and was like lol forget this and converted to hires art: https://youtu.be/UlzgvZqig40?si=_yGqu1EL4l-Y2gzT
🧲 Wishlist Mind Over Magnet on Steam! - https://store.steampowered.com/app/2685900/Mind_Over_Magnet/ 🧲
🍿 Watch the trailer - https://www.youtube.com/watch?v=71MzN_vYEgo 🍿
Developing is an on-going YouTube series, where I share the step-by-step process of making my first video game: Mind Over Magnet!
In this episode, I look at the work behind...
(I say this with love — I think it's very pretty, just obviously a style suited for a family of physical devices we've left behind)
Luckily, Fomorian frame only goes up to size 3, so Barbarossa won't be an issue in that regard.
Iirc barb used to be size 4 in some old playtests and it was rough, so it might even be the reason fomorian frame has that limit
Worth considering the Ship Template when thinking about max possible sizes, scaling up NPCs and all. Although I'm not sure what the plan with templates is so ignore me if it's not relevant
Oof, yeah, I hadn't actually thought of that. The only solution I can think of is to make a sprite specifically for ships and use it for any NPC class that gets the template. But it might also end up being a thing that we shoot for for the 100% port but ultimately not get to.
If any template was gonna get a unique sprite, it would be ship
pixel art good
the good news is you don't have to do anything close to this
its just the default size and size+1 (max of 3)
fomorian is the only "get bigger" effect, and only increases one increment to a max of 3.
no size 3 dusk wings or other silliness
Now I'm realizing there needs to be something for Sag + Fomorian Sag
That reminds me Olive, you solicited opinions in letting any size 1 NPC double as an Everest sprie. That sounds like a cool and easy way to expand the options, so long as there isn't some secret big overhead to doing that
maybe like a "main" ship sprite that then gets 5 variants for the various roles? (or 6 we guess for Bio)
that is 4-5 more sprites than 1 though so like. y'know.
and the thing about pixel art is that making something twice as big typically takes about four times as long from the larger area, so even a single size-4 sprite is a reach
honestly it might be easier to have it be a 3D model, now that I think about it.
the line thickness increasing proportionally with size sometimes looks really weird
do we just like pixels for nostalgia? discuss
Dumb suggestion I mean entirely as a joke: reserve the 3D model for NPCs with the Exotic template
ha that's good too
I bet paper mario did something with the eldritch horror of beings with a third dimension
i switched from hex to squares just for this one metavault scene myself
not asking for this btw, not even as a joke
it was kind of a nightmare to figure out a good way to do it on vtt, I'm sure it's 1000% worse to do it in LT
In Super Paper Mario, Mario can shift into the 3rd dimension, and doing so is physically dangerous to his health and he can only do it for a limited time
In Sticker Star mundane, 3D real world objects are alien artifacts that nobody understands the purpose of, that you can use as weapons
In Cassette Beast, all characters are 2D pixel art who live in a pixelated texture 3D environment. The Eldritch Archangel in the game range from 3D model, 4D model, paper cut collage, 90’ Cartoon looking guy, and a moving painting.
Amazing
cassette beasts is such a good game
Heard both good and bad things about it- Supposed to be good but it's Slow
Can't be slower than gen 4 Pokémon
No idea
You can consistently get FTKs or ZTKs in cassette beast if you get the right set up
I do think pixel art has worn out its welcome. Kinda disappointed me when octopath traveller went for the retro look, you could had this beautiful pop up story book look without the "grainy filter"
Variety is the spice of life. Pixel art games are, should be and will be always in the mix and that's a good thing
True, when it's additive to the game experience. Something like hyper light drifter for example
boooo hiss
:v
ill always appreciate good crisp pixel art
Pixel or not pixel, I'll love this game anyway
it's also a low barrier to entry if you wanna make your own assets
of course you can make amazing pixel art, but also it's relatively easier to make a decent one compared to other styles
Honestly gotta disagree, Octopath is beautiful and the art style really adds to it, imo
That's a fair point
I dunno, i get that often they are try to invoke the retro look of older games, but it just makes me think of those games rather then what I'm playing.
It gets me a bit with sea of stars as well because it just makes me think of chrono-trigger, which is annoying because Sabotage Studio are epic.
The messenger is one of stand out games with pixel art because it's uses 8-bit and 16-bit to show time travel. And the writing is amazing.
Evoland 2 did similar. BW 8bit. Color 8 bit. 16 bit. 3D for different ages
I do pixel art, and yeah some of it is the nostalgic factor, but I really enjoy the limitations of the style. The limited color palettes and low resolution can be inspiring, and require some interesting techniques to try and work around them. It almost turns some pieces into a puzzle, where moving or changing a single pixel can completely fix or ruin the whole work. Also dithering just scratches a certain itch in my brain.
Those same limitations can be a crutch, making it easier to get into, but there's still a lot to learn to do it well. It's it's own medium with it's own tropes and styles and techniques to learn. Not that there's anything wrong with it not being a style of art someone personally likes! I can certainly understand it being over-represented in a lot of games, leading to the style overstaying it's welcome for many.
I think especially from square-enix that it's just for that nostalgia, I don't mind when it's genuine expression because there's a passion behind it that gives it character. Square doing it seems like baiting.
For literal decades in the video game industry squaresoft was like one of the go-to examples for quality pixel art in video games, saying that them using pixel art these days is just nostalgia baiting is like saying that lebron james winning basketball games is just pandering
If anyone in the industry is allowed to make pixel art rpgs it's them
There are plenty of avenues you could look at and reasonably say that modern day square is trying to leverage nostalgia, like the fact that they're going to be spending a decade+ remaking a 27 year old game, or their lackluster ports that bring little to the table an emulator couldn't except with worse fonts, but making new games with all new pixel art graphics is just making a game that has art direction
god forbid someone use an artstyle
I do agree with almost everything with that, I didn't want to expand this out to a greater discussion on square-enix as a company since it would getting away from my grips with the what I feel is the overuse of pixel art.
The specific grip with square-enix and their use of pixel art is that it doesn't feel like the style itself is evolving, they have a specific style which to me is a safe chrono-trigger like style. octopath 1 and 2, live-A-live, star ocean are all games that to lose their identities visually because they are so much like chrono-trigger.
I ragged on sea of stars a bit, but that studios art style still has quite a bit more of identity outside of chrono-trigger which is clearly it's inspiration. I guess what I'm trying to separate is inspiration from pandering.
I will check by basis now since I probably should have stated this at the start; I don't enjoy games that feel like they want me to look backwards as the reason they are good. It's why when I look at modern games using pixel art, especially games with large budgets, I'm critical if they are choosing the style for an artistic/limitation reason or for pandering.
I do wanna present what are some strengths with pixel art in addition to what Ada and Suji said. it is in my opinion, pixel art is fantastic for maintaining clarity in fast paced games or games lots of visual information. Replaying some of the old sonic games as well as mania highlighted this, playing games like katana-zero, hyper-light drifter or Risk of Rain returns would be incredibly difficult without that visual divide which pixel art is known for
I'm not saying it an invalid style or doesn't have strength, it clearly does on both fronts. I just question when it feels off, when the choice feels doesn't have some reason outside of invoking the past.
I feel like its a bit unfair to say the style "hasn't evolved" and then mention octopath 1 and 2 who, personally, by all account actually use the pixel art to amazing effect
not even mentioning the boss fights, which use this effect of giving them more detail which really makes them stick out
Pixel art requires a ton of effort, more so than what a lot of people give it credit for
There's some really interesting stuff around making pixel art work, because a lot of the old pixel art was specifically engineered to create specific effects on old crts - If an artist understands the mechanics, they can use it to imply detail that simply doesn't exist in the raw image. Super cool stuff, and absolutely not the same as antialising it.
But that means pixel art now isn't really the same technique at all except at the most rudimentary level, old sprite styles can look pretty rough on modern displays, (and somehow no retro games or emulators seem to offer crt shaders that look remotely like the old blending you used to get.)
That might be using greater bit depth or some specific shading techniques to get the look right, but I think it's really cool how it's different and it's really not the same thing it was back in the nes/master/snes/megadrive era
octopath traveler also doesn't look like chrono trigger
like, dracs eye there. One red dot creating sinister narrowed eyes
if Live a Live shares some resemblance with Chrono Trigger, well, there's a reason for that considering Live a Live was made by square in 1994 and Chrono Trigger was published in 1995
It’s feeling like this convo has drifted away from the thread topic to a wider discussion of pixel art in video games? Would it be appropriate to move it to another place?
I'm done, tbh
Valk right, apologies I do like to debate.
Three solid days of work; we got data serialization for save/loading (mostly) working! It was very satisfying; everything in the game can neatly fold and unfold itself up, to and from json.
The remaining thing to do is to figure out how we want to handle pilot portraits in the JSON; you can see the pilot's picture become "unknown" after loading. Our options are to regenerate the portrait image (will make loading a little slower), try to encode the image in text (will make the save jsons a little less convenient to manually edit), or just give up on it being json at all and save the game in a binary format. Not sure which one yet.
here's a savestate, if anyone is curious. a nice an compact 12kb for the above scene
How long you talking on regenerate portraits? In a scene with 4 players, and 6 structure worth of enemies that would only be 10-20 portraits
I think smaller saves are better
almost certainly less than a second
it's just a bit more finicky because they're currently cached by essentially taking a screenshot while in the portrait maker; we don't have machinery for generating an image from the data while in a different mode
I remember doing mod work on SS13... Load times add up fast
You're crushing it Olive
Make all portraits be a combination of pre-set face characteristics. To then load a save you need to mix and match those characteristics (as stored in the JSON) to unlock it. This saves needing to cache the image and also gives your game Facial Recognition software /s
My understanding is that you couldn't store it as a blob file
Oh?
Shot a message on twitter but I didn't realize there was a whole as thread dedicated to this:
This is incredible you are incredible keep kicking booty
ah i meant to put a "?" at the end, you storing it as an image. But off the top off my head - if you store it as a face you can't swap it true, but it does mean if i force a new asset into the game it could replicate outside of the modded save?
If manual editing of the JSON is a valuable feature, you could store all portraits as a JSON blob with a key value to link to it's actor in a top level array at the end of the JSON. Then all the human-readable stuff will be at the front, and when you hit the base64 morass you know you've passed all the useful data. Personally I'd still lean on regenerating the portrait from saved character creator parameters, but if it's a blob I'm absolutely going to experiment with swapping that data with arbitrary images for my pilot portrait.
Planning for a "Feature not a bug" already XD
It is a good point though; allows custom portraits instead of forcing players to use the in-game portrait maker
I love the portrait maker already, but I know people who would feel constrained
That's got the "look into supporting for its own sake" tag anyway, I agree it's valuable to be able to have strange one-offs. I'm leaning doing regeneration as well, but at some point add an escape hatch where you can feed it a file path to an arbitrary image.
or a blob, I dunno
if you make the portrait into svg instead of png (nightmarish task imo) you can probably store the entire svg file inside the json
that's what the blob thing we're talking about basically is; represents image data in plaintext
a lot more condensed than svg would be
Spitballing; what if you cached portraits to their own folder so they don't have to be added to the save file? Editable, doesn't have to be regenerated
ha, I literally just did that to address the slight loadtime issue! I don't save the cache to file; after it regenerates I stick the portrait into memory (they're small pngs, it's probably fine) keyed by a hash of the portrait json. It'll have to reload em the first time you start up, but I think it'll be a lot faster when like you have 100 pilots in your roster.
i'm calling it the backstage photobooth where pilots line up to get their pictures taken
license mugshots
It's not too difficult to load an external file as a texture in Godot, and you already have a file picker. Once the image is in memory, converting it to a blob is the same regardless where it came from.
I wonder if you could save out the images as Godot resources, use a generated hash/uuid as the file name, then refer to that in the save data JSON?
{"portrait": "abc123imauuid"}
I guess the save us not all encapsulated in one file then, which might not be desirable.
yes, I'd like if possible to keep these saves as a single file as much as possible... and I think that that'd be about equivalent to persisting this cache on disk instead of memory. Which we could do no problem but since pilots/unit objects come and go, I think a persistent on-disk cache would pretty quickly become a graveyard of portraits with pilots long gone.
which is kinda cool I guess but a bit messy
"tile":"(14, 16)"
Does this mean every time you store a vector in the json blob, on the loading side there's a utility function to parse this string back into a Vector2 / Vector2i?
No judgement, json just makes my bones itch; I had this problem working on the lancer map creation tool for mr. ralf interpoint and eventually threw up my hands (and my breakfast) and switched tostore_var(giant_dictionary)
good eye! yes, exactly, Godot can automatically do the conversion to vector-like string but to get a vector back out I have ```
static func is_serialized_vector2i(value) -> bool:
return (value is String and value.match("(*, *)"))
static func deserialize_vector2i(value:String) -> Vector2i:
if is_serialized_vector2i(value):
var split:PackedStringArray = value.split(', ')
return Vector2i(int(split[0]),int(split[1]))
return Tile.INVALID
not having to do that would be another benefit to switching to a binary format instead of json
and I have some general serialization/deserialization utils that I can feed a resource and a list of props in to handle the basic variable types while turning a resource into a dictionary:
static func serialize_attributes(resource:Resource, serial:Dictionary, attributes:PackedStringArray) -> void:
for attribute:String in attributes:
if attribute in resource:
serial[attribute] = resource.get(attribute)
static func deserialize_attributes(deserial:Resource, serial:Dictionary, attributes:PackedStringArray) -> void:
for attribute:String in attributes:
if serial.has(attribute):
var value = serial.get(attribute)
if is_serialized_vector2i(value):
deserial.set(attribute, deserialize_vector2i(value))
else: # normal property
deserial.set(attribute, value)
Whenever I think about having to write something like this, my stomach churns, but having seen someone else write it, I'm always like "huh, that doesn't seem so bad" 😆
yeah, being able to do reflection-like stuff in gdscript like iterating over attributes is a neat feature of the language!
it really is! I ran into a case earlier today when I didn't know up-front the list of attributes I wanted to serialize (a resource type that has like a dozen inheriting types that all do slightly different things with different attributes) and I was actually able to write something up that is able to step through whatever resource I fed into it and serialize all their attrubutes automatically. It saved me having to write a ton of teeny boilerplate functions.
it definitely started to get arcane though, with weird hardcoded internal ints as type values 😅
eg this snippet
24: deserial.set(prop, load(serial[prop])) # Texture2D
28: deserial.get(prop).assign(serial[prop]) # array, in a way that handles enum conversions
_: deserial.set(prop, serial[prop])```
uh oh
if you start hallucinating about enums, stop at once and consult a physician
Olive loses it
Experimenting with increasing the elevation cap to 10. It's nice but definitely is risky with increasing the voxel space, gonna have to keep an eye on performance.
Also I think we're gonna have the top-down mode square up and lock the orientation to be north = up while in that mode. Not sure if we want to do 85deg where you can see the heights a little still or a straight-down 90deg which looks tidy but more opaque vis-a-vis elevation.
I like the 85, personally. Elevation is a lot more easily readable (I can read it on the 90 but I gotta think a little), and I kinda like the slight tilty to the view
Gives overhead drone vibes
ye, also like the 85
It gives a little more information without significantly obscuring any tiles, and doesn't look bad either
also in favor of 85 for the same reasons
Honestly I don't think 85 is any meaningfully easier to read than the 90, and I do like how the 90 looks
I could see a desire to have both top-down locked to north and top-down with your current iso rotation (i.e. only a vertical rotation on the camera, no horizontal rotation). Would it be possible to have both as options?
What do you think a good control scheme for that would be? Right now it's a dedicated button to toggle the topdown mode on and off.
Ah yeah, that makes it tricky. UI or keyboard button? A single toggle is simple and elegant, it might not be worth complicating that.
Keyboard
this makes me wonder, are you planning to have some sort of easily readable display of exact height on a tile? E.g. a little box on the hud that displays what you're currently mousing over in terms of like, elevation, difficult/dangerous terrain, etc.?
Oh that's not a bad idea; I could maybe add an elevation number to this terrain box? Except it only comes up when there's non-standard terrain... hm.
could be a toggle?
Could it be a number in the corner maybe?
corners are pretty primo UI real estate
I'm sure humans have figured out many, many good ways to have maps depict elevation.... I should just do a research dive on it at some point.
Alt to show more info? That's a fairly common idiom. Could either only show while you're holding Alt, or pressing it toggles the extra info display.
Could also have that toggle things like HP bars? 🤔 Threat ranges?
Added to the list
If your doing an 85 for thr overhead it might be worth looking at what a.. 10-15 or so degree shift from North looks like too, imo it might be a bit more readable because you'll get height information from 2 faces
Another thought, what if height intrinsically was tied to colour, with the tile getting a tiny amount more or less saturated. Then in the pure top down view you'd get a little impression of heights at a glance.
I've always liked how creeper world did it - the little column of yellow bars in the bottom-left that's also got a number on it
here's how a couple super robot wars games did it, though not near as much terrain info to be had there. in the first, you either tap the cancel button one tile at a time for terrain info, or you can scroll all around the map for it when you're selecting a unit's movement to check everything. you get a popup box for info that could shift around the screen, sometimes paired with one for units if you're doing movement and you cursored over a unit in your range. (sometimes these would glitch out and overlap in the same space, whoops.) the second, the cancel button toggles a sort of investigation mode where it brings up a less obtrusive persistent bar at the bottom for terrain info, and cursoring over units gets a much fancier quick-info popup box.
not sure how doable it is, but could you put an elevation number on the top tile of every square when someone holds a button? Would make for quick and easy comparison of relative elevations
I'm sorta hoping that for quick comparisons, they can just look at em in 3D mode, otherwise what was the point?
fair point!
Belatedly, I've always been curious how save/load systems worked in video games!
goes from very simple to "oh gods no"
depends entirely on how much state you want to save: a platformer can get by with very simple data, a skyrim needs a way to store the state of every cheese wheel in the world if you touched it once.
Skyrim, incidentally, bungled that up such that every PS3 copy invariably crashed on boot past a certain playtime
yup
be glad if the game you're working on is turn-based; having natural sync points where you can just serialize the game state makes it so much easier than trying to deal with a continuously interactive, real-time game
This is also why you see things like checkpoints or save/load only being available in the menu
Data management is one of the larger problems to solve for an MMO, for example, and is often why you see them sharded.
This is also something that can make mod support tricky; how do you handle the possibility of loading a save that was made when a missing mod was enabled? It’s one of those things that gets handled differently depending on how the game’s data is organized and what the developers feel fits their game best
just did an import of a buncha updated assets from Martina, so now the rest of the hair and clothes options are recolorable
For like the last month I feel like I've been saying "we're getting close to a vertical slice" but sub-tasks keep getting generated the more I playtest, but Carpenter and I just had a meeting where I think we've outlined the rest of the things that I need to add for it. Today we got a UI for getting structured/stressed.
skyrim also has some quick save/quick load nonsense used in any% speedruns
aargh! discord scroll
Exciting!!!!
does this help solve your issue with portraits disappearing on load?
no, that had to be handled with a "photo booth" that regenerates + caches them the first time they're loaded during a session
this just finally updated the assets so they could be recolored; the original asset needs to look weird and red like this so the game knows which pixels to recolor to what channel
I know this is probably wildly beyond where you are right now, but have you thought about enemy AI at all? I'm always curious about how it's implemented.
ohh I see! reminds me of picrew and retrograde
actually I guess this stuff shows up a lot in a lot of places
I think I've overengineered the current solution... this is the diagram I made while putting it together and I don't super understand it anymore. I think the idea is that it makes a "heatmap" for each tile on the map based on how much it would like to stand there, calculated by what abilities/weapons it can use from there + if it's in a command zone tile - some decay if the tile is far away. It then goes to the hottest point on that heatmap and does the highest-rated action from there.
I think it was supposed to be flexible where its behavior is dictated entirely by the stuff it's equipped with rather than a top-down objective. It does its best making that heatmap with normal weapons (each point of damage they could do is +1 score for that action) but it could also be overridden by special abilities to prefer targeting allies, for example.
I should have started smaller and built outwards, because one of the things on my to-do list right now is figure out why the berserker keeps frag sigging instead of using its axe
and it's hard to navigate this system because there's a lot of stuff I added that there's no actual examples of yet
prop for not segrating assest/hairstyle by gender
It is a sore point for me and likely many others too
here's one of the tests for the AI heatmap that's generated with an AI unit at (0,0) that's equipped with a pistol and a potential target at (3,3). The 2s in the heatmap are because it likes being further away when using ranged weapons.
I think Weight based AI sound like an opportunity for a lot of cool customization moment like having sliding scale of how much an AI want to do something like "attack in melee more" or "support ally more". Which can be used by mapmaker
it's more bottom-up than top-down; instead of an AI unit that's set to "support ally more" it's an AI unit seeing "oh, I have an action available that's highly rated for supporting allies. guess I'll do that." Here's the current fields for customizing an action's behavior via those weights
This fucks pretty hard ngl
and was what I kinda hypothesised would work myself. Worst case scenario could you bar some enemies from certain actions? As a GM I never have -1 or worse SYS NPC's frag sig, for example.
berserker too based to live, needs nerf
is the heat/state weight analysis based at all on potential target structure / stress outcomes? I could see some AI naturally realizing that it's more optimal to use a second quick action to invade 🤔
exactly, the default tech attack behavior checks the unit's tech attack and scales the action's base score on it. Looks like I have it floor out at 1 even for negative tech attack bonuses, but otherwise they would literally never use it.
but maybe folding in "structure/stress potential killshot score bonuses" would make a negative base score fine
I imagine defensive behavior might be hard to implement... how would you get a NPC to dive into cover?
True. I like to spike my own NPC's chances to hack when a player is in the dangerzone. To punish reactor crimes, ya know
I haven't done it yet, but I assume I'll be able to scan the map and give tiles that have cover from enemies a bonus score
I don't know if they'll be smart enough to tell what players they should take cover from.
Its certainly abusable, but if you put have a way to expose it to their decision making, you could make the benefit depend on the size of the player's largest mounted gun
Means that they'll make bad plans against howitzers but idk what can you do
call that a feature 
I'd also like to introduce some amount of randomness/jitter to spice it up a bit but I'm not sure what's the best vector for that. Add a small amount of noise to the final heatmap? Have it sometimes pick the second- or third-best options?
The latter sounds more impactful
yeah having some randomness is probably good.
Just have the cover weighting ignore arcing/seeking weapons. 😉
I had done a decent amount of pondering on an AI system for a tactical game, that took into account mission goals for longer term decision-making and tactics for planning with other units with a similar weighting system; but seeing this and what needs to go into it even without those aspects reminds me of how naïve I was being, not surprising since I've never done any game programming let alone complex AI lol. Your mention of possible overengineering just reminded me of some videos I've seen from "AI in Games" about FEAR and Into the Breach, where sometimes simple systems can seem more complex than they really are. E.g. FEAR's enemies don't know anything about each other's actions or decisions, but they don't really live long enough for that to become apparent. Hopefully that implied complexity comes about here too!
Me, triple-checking that weapons can only have one weapon mod: "OK, I think we're safe linking weapons to weapon mods as a singleton rather than an array; there's never going to be more than one."
OPCal waiting in the wings: 😈 😈 😈
(it's not technically a weapon mod but is best modeled as one, so now I gotta make the system be able to handle multiple things modifying a single weapon)
I could hack it in as a one-off but then where will that leave AutoStab... and here we go, more sub-tasks are being generated when I'm trying to tie off the character sheet work for the vert slice.
autostabs and, presumably, core powers that buff weapons too?
tuga, sherman, raleigh, etc
or does the limited duration of those effects change how you model them
it's more the "this only applies to one weapon that you have to choose"
those CPs all have their own thing going on and don't require a player choice in the character sheet
I imagine this also applies to the Tagates' thing
how's that work, again? I don't have it on hand
When you equip a weapon you can choose to make it a Rifle instead
ohh yeah hm... that'll be a later problem, but my gut is that yes it'll be a weird "weapon mod" that you can apply to multiple weapons
lol it'll require a broad search-and-replace for all the places I directly fetch the weapons' types and instead make weapon types a derived state that needs to check its mods
is tagetes core?
nope, hence later later problem
And the Viceroy making Launchers two things at once can't make that easier to program <_<;
ohnnoooo right
lancer is a complicated game, everything can be modified by anything 😩
Atleast if you get the Viceroy working you can apply that solution to the Mimic Gun?
This is reminding me how I wish that weapon types were an array in CompCon
why isn't everything immutable like clojure
Or fourth- or fifth- or sixth-best options 😄
Signed, someone who likes easy difficulty settings
@alpine sorrel I think this might be interesting to you, re: portrait maker and Black hair? https://www.reddit.com/r/Games/comments/1cficeo/as_a_black_gamer_i_dont_care_about_anything_else/
yeah we're being pretty conscientious on this front, have already hired a Black sensitivity consultant specifically to take a look at the portrait maker once we get it a few more steps forwards.
Thinking today how it'd be cool if people end up being able to eventually make and sell larger LT modules as like 3rd-party DLC. I know modding culture is that most mods are free, but we have an interesting parallel where players are buying 3rd party campaigns for the tabletop game, so why shouldn't there be the same deal for LT campaigns?
Since we're already planning on supporting being able to import other people's maps, it'd just be a matter of those people putting those maps behind like an itch.io paywall.
I think in-app purchases are beyond what we wanna do, but we could probably have an in-game curated list of paid/free community content like CompCon does.
For the KS version, yeah.
A later steam release is likely after all the obligations from the campaign are cleared
i wonder how steam does paid mods
i know it's been done before
It doesn’t anymore, really. That feature got pulled after a lot of negative backlash
Unless it’s been re-added and news about that is buried under older search results
But as far as I can tell that’s still the case
Still, I imagine itch being the primary distribution platform (if I’ve understood things correctly) for the base game would make using itch for paid campaign mods easy enough
i think optics would matter for this. If it's "people can sell mods", it's bound to have some negative backlash. however, if "third parties can sell mission packs" or expansions or something along those lines, it might go over better
I’d have to do more research to fully refresh my memory, but I recall it being more complex than just the principle of paying for mods being objectionable. I vaguely remember complaints about how much of the sale revenue went to Bethesda and steam? So there could be other differences that inherently make this more palatable
for one, distribution of third party content would certainly still need to be made through itch io. steam doens't really stop you from installing external mods, so you can just buy them on itch and manually install on your steam version
That's a good point... gotta get some sorta "third party DLC" stamp for them.
A sikh friend of mine always keeps an eye out for turbans, speaking along these lines. Would be nice to see too
re the DLC - An incredible idea.
quick question about how the scripting possibilities for campaigns - does your scripting as it stands allow for pop up images or animations? My circle involves a few comic artists and I dabble so it's something of interest to be able to do
something like a visual novel esque segment? That sound pretty fun actually
I love the idea of community made story DLC. I think that idea of having a currated list in the game (with accompanying links) is a good idea, but I would also have a link to an itch.io tag so people can easily find the rest. I do expect you'll get some backlash regardless, but hopefully the fact that you're not Bethesda and that your main market are people that are, as you stated, used to buying story packs leads to it being an overall positive.
I hadn't thought about this... the visual novel scenes will have backgrounds you can set, but I hadn't considered popup image panels. I'll write that down as a potential idea.
So there is prescident for this?
a friend is playing NWN - and they actually do have "Premium" modules, modules that people can sell
YMMV if they'll take off
Oh neat, I'll take a look at how they did it, thanks.
kinda wanna see someone use that avenue to make a big mission collection with a purposefully wonkily-drawn mech for the cover (think some of the more awkward 80s battletech and transformers art) and labeled as L-ZONE in throwback to those questionable huge 90s map packs for doom and duke nukem and the like
Does anyone have the exact dimensions for the campaign challenge coins? I wanna get a bezel or something similar for mine, and I'm not sure about getting those right by hand.
Just checked with my calipers, it's got a diameter of ~38.22 mm and a thickness of ~3.35 mm.
I wouldn't trust that last decimal place, I had a decent amount of variance in measure depending on where I checked.
Not sure how much of a workload it would be on you and the team. However would it be possible for interested parties to submit a DLC for you to test and subsequently release in an official capacity?
Also, "Black Reliquary" (a mod for Darkest Dungeon) has its own page on steam. Not sure if there is something similar but paid that could work.
Looks like 1.5in is a standard size for commemorate coins, so that tracks
oh that's wild, I've never seen (albeit free) DLC listed on steam by a different developer than the main game. good to know that's possible. I wonder what the communication + agreements between that team and red hook studios is.
However would it be possible for interested parties to submit a DLC for you to test and subsequently release in an official capacity?
Hard to say before we get to that point. I'd prefer to not have us be a bottleneck for people releasing stuff in general, but if we get to the point where someone has made something big, it'd be cool to be able to highlight it.
I've seen portal 2 mods on steam store, and I doubt they contacted valve for a deal
in a similar vein there's Slay the Spire: Downfall, which is a mod that has its own store page.
There’s a series of total conversion mods for Elder Scrolls games (Enderal for Skyrim, Nehrim for Oblivion, etc.) that have Steam store versions which give a separate executable to launch. They’re free though (for obvious reasons).
Oh shit downfall has a store page
Wow a lot of people have paid valve $100 to basically get a VIP workshop page
some of those mods are basically new games
but i wonder if there's a deal you get for that kind of thing
Some people spend a lot more than that on their hobby/vanity projects. 😉
it costs $100 to make a steam listing? didn't realize that before 😱
Better than having to run a greenlight campaign (back in my day...)
The video for a talk I did on LT's project architecture just went up: https://www.youtube.com/watch?v=T9f5POJywcA
Dive into Godot with the creator of the adapted CRPG "LANCER TACTICS" for structural tips on systems-heavy games!
Recorded live from the Portland Indie Game Squad Talent Talks in March 2024.
View the playlist here:
https://www.youtube.com/playlist?list=PLRf17HFnsZ8cK-3jOqAKXxKfmeiT9mo9Q
For more information about the Portland Indie Game Squad ...
Slides are here if anyone's interested in taking a closer look: https://docs.google.com/presentation/d/1OJypkW5yHjDOFg1ShUMAL-_NMEI3SHl8b1Na9pqldw8/edit?pli=1#slide=id.g26badd7a696_0_1
this was really interesting to listen to, thanks for posting it
Got the alternate actions panel ironed out, complete with WA's shock effect
👀 What's this for? Just anything where you go "want to add another effect to the one you just did?"
Kinda like in BG3 where you have multiple options for an action e.g. upcasting a spell; we don't want to add an icon for each WA option in the main picker, so the engine collects all the possible variations on an action when you start using it and lets you switch to them. To complicate matters, sometimes there's multiple dimensions of how an action can be changed; something like WA on a throwable melee weapon would have to let you independently choose both the attack option and whether you're throwing it or not.
here's the flow for it; it was a beast to sort out. When you pick a new action, it essentially has to cancel the in-progress previous version of the action (since we want people to be able to start using the default version without having to make an explicit choice first) and swap in a new modified version of the action (e.g. with knockback) and start running that.
whenever I discover a new exciting bramble like this that is required by Lancer's mechanics, this tweet from tom during the kickstarter echoes in my brain
but hey I also added in a lil takeoff and landing animation for flight
Finish lancer tactics, conquer hell, become the devil
lance six billion tactics
tac 6 billion tics
Not to be confused with tic 6 billion tacs 💊
No one has ever finished a game of that. Takes a long time to get 6 billion in a row
Apparently, they make 55 billion Tic Tacs a year. So it wouldn't take that long.
nah someone'd accidentally nudge one and mess it up
and ra forbid someone add a toe
Nice Talk 🎉
I have a question about how functionality is being organized in that giant node graph of doom
One of lancer's rules is "weapons aren't allowed to have more than one mod applied to them" and I'm curious about where that logic lives
Is it a question of wandering up the graph to the closest common "ancestor" of the two interacting functional nodes and chucking the logic there?
I.e. is the Kit responsible for ensuring that weapons only ever have one mod?
Are there other situations like that you've run into where quirky interactions between multiple separate nodes in that graph have caused you headaches, or has it been pretty smooth sailing?
Sorry if this is Top Secret or too thorny to explain quickly, no worries if you are too busy, thanks again for linking the recorded talk!
I was just talking about that, specifically! The way it's currently modeled is that a GearCore (e.g. a weapon) can have a pointer to one other GearCore (i.e. a weapon mod system). I thought I'd be safe having it be a single cxn but now I need to have other things also be able to modify a single weapon (opcal) so I need to modify the doomgraph so there are multiple "mod" cxns. I used to be able to just see if there was a cxn when checking if the mod slot was filled, but now I'll have to step through all the attached modifications to see if any are a true weapon mod.
#1079952007915376711 message
So I guess the logic for ensuring single weapon mods will mostly be enforced by the character sheet interface more than the engine itself.
That sounds like if someone made a third party save editor, it could open things up for some fun, funky builds
there's a good chance that pilot saves will just end up being json so a text editor will do
any chance the pilot build will be comp con compatible?
Copy/paste 30 molten wreathes onto a combat drill
technically possible, not something high on the priority list
would have to make some sort of importer to convert between the formats
that's fair enough, I'm sure somebody would be interested enough to write something that convert formats if you don't do it
Actually, I realize this isn’t relevant to the vanilla product since in game there won’t be a way to do it, but I’m curious; would the same mod applied multiple times via mucking around with files actually have its effects applied multiple times? Or does some part of how it’s structured stop that
Not sure since I've not figured out mods yet, but from what I've read files can be overwritten by mods, so I assume for multiple mods the last one loaded would win
Oh sorry, I meant weapon mods via save tweaking
Oh 😅 yeah it depends on the mod but I think most of them would apply twice.
Adding knockback, which means I need to add cone calculations (to choose which angle to knock back), which means I'm adding a half-implemented daisy cutter (soft cover'll come later).
algo still needs some wrinkles ironed out
Oh that was a quick turnaround
bad and naughty pilots get repositioned by the cone of shame
Cone 7's, continuing to be ludicrous in the best possible way
I first entered it as 2d6 cone 5 because that's what seemed reasonable to me and was once again shocked when I doubled checked lol
3-18 is so much funnier to look at than 3d6
you look at it and immediately go "wait there's a chance I might do 3 damage"
knockback and fall damage!
YOU GET TO CHOOSE TO KNOCKBACK DIAGONALLY ?????????????????????????????????????? THAT SO COOOOOOOOOOOOOOOOl
that npc is cool!
Oh yeah, is that a berserker?
I keep getting distracted adding little touches; having the unit turn around to face your mouse as you target moves/actions makes it feel a lot more dynamic
It did look a lil weird in the previous gif without that
Doing amazing work, Olive.
it's come so far in so little time
power of day job
Nelson looks adorably chunky as always
I really like the sorta shark face the berserker has
New changelog is up! https://www.kickstarter.com/projects/wickworks/lancer-tactics/posts/4096488
mostly consisting of gifs that I've been posting here over the last month lol
I'm surprised every month at the "feels slow" // "look back and a lot happened" contrast
update post 
love seeing all the progress collected in one thing
Great work and thank you again for the updates
What’s that last teleportation bugfix?
It used to be that since teleportation only counts for one step, it would only cost one movement no matter how far you went. This was relevant for the Mourning Cloak's CP; it could teleport several times to go much further than its normal speed budget. Teleportation is now costed by how many tiles away the destination is, OR the height difference between the two spaces, whichever is more.
the second bit is so it still costs like 4 movement to teleport 4 spaces up onto a building
which seemed right?
or like so a fold knife's two spaces can't do that
i assume the updates will not happen to the downloadable version for a while as it's a demo and you're going to putting out a final product
Ahh, I get it now
I want to leave the pre-kickstarter demos up, and it'd be nice if they could keep living on that page
but also that's where it makes the most sense to put the alpha public demo of this
could leave a link to them or smk? I mean hey, the demos are wonderful, i had a lot of fun with them during a slightly slow lecture
Put new one there
And a link on the page "Are you looking for the old one? Click here!"
YEAAAAAA
It's a small thing but I like the little up and down dance the sprite do when selected
The importance of a good 'lemme at em' selection movement is nice
lol just found a bug with knockback, can you see it
you've heard of "running up that hill" now get ready for ramming up that hill
zoop
improbably dense ball of beserker grunts
will we be able to deploy an improbably dense ball of ronin in lancer tactics?
Ka blooey
ballin'
would chomolungma using core power in a cloud of hornets crash the game?
I know lungma isn't in the game but like, if someone were to add it in
or i guess what I'm really asking is, how long of a chain of actions can be handled in a reasonable way
I can't think of a theoretical limit. Depends what you mean by reasonable. This sort of stress-testing is what public playtesting will be for 🙂
hell yeah
I was gonna ask if Wrecks are the sole exception to the "Obstructions = 3D" rule, but then realized they're not obstructions, they're actually difficult terrain. I never knew that!
wait what, I thought they were obstructions as well and was planning on making em 3D
The way I'd read this is to have them be both obstructions and difficult terrain
...oh, but that's going to be hard for the engine in the case of a wreck being placed on top of another object like Jericho cover. I might have to have the shunt-to-a-nearby-legal-space check cover that.
Doublechecking the obstruction rules, it says " Characters are obstructed by any solid objects", and I presume that wrecks aren't that because they're merely difficult terrain instead. The alternative would be... An object with difficult terrain on top of it? Which doesn't seem right
I never had to reckon with this my GM played without wreckages
At any rate, wouldn't you have needed that check anyways, in case a mech got destroyed on top of a wreck?
Oh cool so this is the lancer tactics group chat. I played the demo and I thought it was pretty fun
Is there a release date for the full game yet?
oooo
ks goals are slated to be done end of 2025, though we've already hit some complications (moving to 3D) that may ahve pushed that back a bit
Yeah wrecks are odd but they are def meant to be difficult terrain as well
Maybe easier to look at them like fucked up water spaces I guess
I don’t think anyone would blame you if you tweaked Wrecks to play nice with the game, too
im a fan of wrecks for producing the exceptionally fucked phrase of "zone of hard cover" in most rulings of them ive seen
Tbf that's also in the Kidd license
Wrecks working like this is also an important part (imo) of a lot of npcs
Destroying a Goliath means you're making a big patch of difficult terrain
Imo it could be worth making it a zone of hard cover you can move through that can also be destroyed maybe?
Yeah how am I supposed to gum up the escort path with grunt bodies if they aren't difficult terrain ;D
(a thing that happened in my first Wallflower combat that was a cool visual, and mechanical hurdle to figure out)
A non-obstructing destructible zone of hard cover is actually pretty easy to do in the engine I think... that might lose the "objects that can be pushed and moved around" aspect but seems fine.
If I remember right wrecks only provide hard cover for pilots inside once destroyed, so they don't get immediately shot out in the open but are still targetable. For mechs they only provide difficult terrain.
So it's only sometimes a non-obstructing zone of destructible hard cover, as well as a non-obstructing destructible area of difficult terrain 😄
Completely ignoring the "objects that can be moved around" thing sounds like the way to though, pretty sure everyone forgets about it, myself included. Maybe can make it a target for Ram at a push?
(Pun unintentional)
page 82 crb, as cited above, indicates otherwise
(here)
they just. provide hard cover.
Are there any discord only qna rulings
Not from what I see in the lancer-faq website
worth noting that despite not being obstructions you still can't end your movement within them, as that's a function of being a character or object (for example, a size 1/2 drone doesn't provide obstruction to a size 1 mech but you still can't end your movement on the same space, and the same with allied mechs)
which hopefully means you won't need to code a special case for that (except for the pilot of a wrecked mech, and maybe the Atlas if you ever get around to doing that)
oh that's very interesting as well! .... so RAW it'd be you can't end on the space, they provide hard cover like Jericho does, are difficult terrain to move through, and you can attack/knockback/target them like any other mech/drone/deployable?
I think the engine can handle that all pretty easy; just means I need to toggle these terrain data bools appropriately that are applied to the space the wreck is in.
Oh boy it’ll be interesting to see how Size 2+ gets implemented with dealing with “walking over smaller stuff but being forbidden from ending one’s turn there”
(Anecdotally it’s a common headache for me without House Rules)
I take "Any spaces occupied by wreckage are difficult terrain" as implicitly fulfilling the "[Characters] can’t end a movement in a space that is occupied by another character or object unless specified" clause of obstruction rules, but that's just my reading, what's best probably really is what's easiest to program here
you can attack/knockback/target them like any other mech/drone/deployable?
yep, objects are valid targets for attacks and RAW knockback doesn't specify characters-only (though some sources of knockback do, like Ram). I think most GMs have a "you can't knockback a building" rule where some objects are fixed and other objects are loose
unfortunately the lifting and carrying rules are not particularly well defined, so figuring out how a wreck can be moved might require some judgement calls
TSS1 will do it for a size 1 or smaller.
after a day and a half of work, lancer tactics now has... drumroll please.... remappable controls. captivating, I know.
custom keybinds are something you dont realize how much you appreciate until theyre gone
God I remember spending a good week or two trying to get something like that working in early ue4
it's a surprising amount of work to get all the piping set up correctly; being able to swap inputs, validating that everything has at least one input, saving both the default and changes from that default (like if a slot ends up empty, I have to store the presence of that blank slot), handling the names and icons for each possible input (there are some very useful godot addons that help with that tho), and correctly animating the results to better see what changed/didn't change.
I left gamepad support as a later addition, but had to build the whole system to be forward-compatible with adding new devices like that.
I didn't write tests for it but I probably should have 😅
i hear it's huge for accessibility! Thank you for adding this
a really important thing to add for accessibility!
wonderful work even if it doesnt' seem like muhc
Doesn't Godot have inbuilt Gamepad Recognition?
Yeah but there's a world of difference between "can this game detect inputs from a gamepad" and "can you use the gamepad to play the game"
Mm. Silly question
Not at all! Hard to know the piping from the outside
And how much of it is prebuilt or not
oh yeah, @alpine sorrel you should check out Reverse Collapse: Codename Bakery if you want to check out a really smooth turn based tactics UI imo
genuinely really good and smooth ui in my experience
Can you write off game purchases as a business expense?
yes
Haha, that's good
benefits to being an indie gamedev


