Vapourware Codexian Game Development Thread

[zwanzig_zwoelf]

That pixel art is only two colors right? It's a fantastic art style... very minimal but also packed with detail at the same time.

It's not pixel art, just 'normal' art processed in a specific way. You can check the source images here.

I came across your game (on Steam) via non-codex sources and went "wait isn't that ZZs game."

The hype is real! You better deliver

I don't know if it's experimental nature will click for most fans of the genre, but there's only one way to find out.

I'm a big fan of the genre but gameplay wise blobbers peaked with Infinite Adventures IMO. Something with better graphics might come out but I don't think it'll be outdone in terms of character tinkering.

So something more experimental is a welcomed change. Your game gives me Paper Sorcerer vibes, which I enjoyed even though I didn't finish.

Never got around to playing Infinite Adventures, but I liked Paper Sorcerer. As for my game, the inspiration came from Wizardry IV, SaGa series, Megami Tensei series, and Dinosaur.

 

[duskvile]

I started with Unity and I got stuck. Everything is so slow when clicking on icons, creating C#script, compiling the script. I learn by video and his Unity is much faster.

 

[zwanzig_zwoelf]

Since Demon Lord Reincarnation generates characters by specifying a class, and the desired power level, and then picks a random name from a predefined list, I've decided to allow players to get a chance to add their favorite name to the game.

Yes, this is your chance to add Sir Dong to the game.

https://store.steampowered.com/news/app/2395150/view/3698065064749084546

 

[Justinian]

have a background and some basic run/idle animations

at this rate the prototype will be done in about a year

 

[TheDeveloperDude]

Do you have to make a tutorial for your game?

 

[Twiglard]

I'm finally implementing lighting.

But with this implementation, objects are covering themselves with their own shadows. I'm going to add raycasting or figure something out.

 

[Nathaniel3W]

I pretty frequently hear that it's annoying when trees and other things block the camera. Today I finally had the time to work on a screen-space screen-door effect.

I'll have to make some changes to all of the materials of anything that might block the camera, but that's just a matter of copy/paste at this point. And I'll have to update the cutoff distance for when to apply this, so that it doesn't affect trees that are behind the player. Overall though, I think this is getting pretty close to usable.

 

[beardalaxy]

You guys are doing some pretty cool stuff! I unfortunately don't have access to reactions yet but I at least wanted to say I hope that all of your projects go well and to keep up the work!

I'm making a traditional, turn-based RPG that has a large focus on world building using RPG Maker. It has been in development for 7 years, though not always active development.
There is an itch.io page for it to help you get the basic gist: https://uselesstorch.itch.io/godsdisdain
Other than the tilesets and some object sprites, all of the artwork is original. I created the soundtrack and the sounds are all custom as well.
The main story is completely done, only needing a little bit of polish. It's around 6 hours. You can tackle the bosses in any order and then reach the final boss at the end. The bulk of the game will be in the side quests (+ optional dungeon exploration), of which I have written a lot of but am aiming to have about 50 of in the final game.
Development is steady at the moment, and I'm trying to just get the side quests done so a beta for the game can be released. I'll definitely keep you all posted when a beta is possible and I decide what the best way to distribute it is (closed VS open).
The game will be free on launch with optional donations, and I want to get it on Steam as well with some added features and the potential for some small, paid expansions in the future.

I had a friend actually recommend this site to me as a place to post about the game because it is less strangling than other sites while still being active. I'm looking forward to posting more info about the game with a dedicated thread soon!
It is an incredibly ambitious project, so I can't even really give an estimated timeframe of when it might be finished. I certainly bit off more than I could chew, but it'll be done... eventually.

 

[Nathaniel3W]

Hey beardalaxy welcome to the Codex. As someone told me when I first arrived, leave your sanity at the door.

Keep up the good work. Best of luck with the game. Biting off more than you can chew is standard procedure for solo game devs.

 

[what am i doing]

I pretty frequently hear that it's annoying when trees and other things block the camera. Today I finally had the time to work on a screen-space screen-door effect.

View attachment 37346

I'll have to make some changes to all of the materials of anything that might block the camera, but that's just a matter of copy/paste at this point. And I'll have to update the cutoff distance for when to apply this, so that it doesn't affect trees that are behind the player. Overall though, I think this is getting pretty close to usable.

I quite like that effect so far.

 

[NoMoneyNoFameNoDame]

It has occured to me I have not shared the game's updated trailer here yet.
It's already out of date, but it's close to how the game is.


https://cdn.cloudflare.steamstatic.com/steam/apps/256952481/movie480_vp9.webm

While i'm here i'll point out the current update that came out.
Patch 1.09b

-New/Change: Dying by lack of O2 is now reported to activity log.
-Change: Removed some annoying Infinity Abyss from Enlightened Temple Way.
(Reminder: Enlightened Temple Way is the respawn site for those who have Respawn Mode enabled)
-New/Change: Added additional way of toggling the in-game Task text to the Paged Journal.
-New: Added Infinity Abyss navigation course to the Tutorial.
-Change: Adjusted the Infinity Abyss detection boxes down some.
-Change: Selective Teleporters such as ladders no longer display
text on how to use them manually.
-New: Death by suffocation now has a unique death screen graphic.

Spoiler

And now I will regale you with a story of how I added a new section to my tutorial.

People can die in my game from many things. The things that kill you in 1 hit are quite rough for some new players to get use to.
So to clarify what these are, the Infinity Abyss as I call them, I added a small navigation course to pass through to exit the tutorial.

Now since these normally kill you, I realized for the sake of tutorial you should just warp the player back to the starting point.

The end.

 

[Nathaniel3W]

NoMoneyNoFameNoDame I get that you're an auteur. You eschew convention. You break all the rules. Etc. But an 11-minute video of clicking through menus is not a trailer. You post on here telling us that people give you useless feedback and criticism, and you complain that no one buys your games. Now pay attention because I'm going to give you some useful criticism, and it's going to explain why no one buys your game:

You need to learn how to make a trailer.

A trailer should be 60 seconds. In that 60 seconds, you need to convince the audience to buy, or at least wishlist, your game. In the first 3-5 seconds, you want to convince the audience to finish watching the trailer. You do that by leading with the most interesting, action-packed, intriguing part of your game. Then show the player what kind of game you're advertising. If 15 seconds go by, and I don't know what kind of gameplay to expect, then it's a bad trailer.

I'll watch 11 minutes of gameplay of an exciting game that I've been looking forward to. I'm not going to watch 11 minutes of anything if it starts off with 30 seconds or more of menu-clicking. I don't know how long the menu-clicking lasted because I didn't finish your trailer. If you want to show off your game, then you need to show it off with the understanding that your audience has other things they can do with their time. If I'm watching your trailer and I decide that cat videos are more interesting, I'm going to stop watching your trailer and go watch cat videos. If I'm watching your trailer and I decide that telling you why the trailer sucks is more interesting than watching your trailer, then I'm going to stop watching your trailer and tell you why it sucks.

 

[LarryTyphoid]

Game's already taking up 10% of CPU usage, and all it is right now is just walking around a wireframe maze. I probably did something wrong. Oh well. The textures aren't in an atlas or even an array so that's probably not helping.

I spent like a week trying to get a menu to work because I'm retarded and set the variable used to track which menu option the player has selected to the same value every time at the beginning of the loop. Real tard hours there but I probably didn't catch it because of the fucked up way I've got user input set up with SDL2: the SDL_PollEvent function that checks for input is being looped through in the main source file, and whenever I want it to deal with specific inputs like traversing the menu or moving through a maze, it calls on functions from other files which have their own switch statements checking for the input being fed to SDL_PollEvent. I really don't know if it's right to have all of the user's inputs basically being fed to a single event receiver, but then again, each part of the game having its own SDL_PollEvent sounds stupid too. So fuck it, I'll just keep going no matter how screwed up it is. I don't think you need to worry about that stuff so much when it's a turn-based game anyway.

 

[zwanzig_zwoelf]

I am finally approaching the final stages of development. I aim to wrap it up by the end of the month and release it in late July.

The good thing is that my TODO list is shrinking rapidly, so I don't expect any delays or last-minute issues -- at this point, it's a bug-free, almost-finished game.

 

[Justinian]

submitted a prototype of my next game for pixel art gamejam (with 4 minutes to spare ). very barebones, but it's a decent start.

https://justinianiii.itch.io/enmas-daughters-demo

enemies will have randomly generated body parts, that give different stats. currently only tentacles and horns are available and they don't do anything (yet).
Primary weapon will be a puzzle box that reconfigures itself into a gun, that can then be upgraded further during run.
each daughter will also have her own abilities and melee weapon, + active spells (Hell Verses) will be selectable at the start of a run.
The first (and currently only) daughter of Enma, the king of the underworld, persephone, createas a nightmare cloud periodically. will provide buffs/debuffs as it is upgraded, although in the prototype it just hurts enemies.
more daughters will be added, based roughly on less prominent goddesses from various mythologies.

 

[MLMarkland]

Dashiell Hammett style single location detective RPG investigating the death of Thelma Todd at her PCH cafe.

 

[ERYFKRAD]

Dashiell Hammett style

 

[MLMarkland]

Dashiell Hammett style

I think the move is to figure out ways to reinforce the player having an inner monologue as the detective having to actually figuring things out and solve them rather than a bunch of Sam Spadey voice acting from Jeffrey Dean Morgan or whoever.

Otherwise it’ll just come down to whether each player likes that specific version of the 1930s LA PI we picked to do.

 

[Nathaniel3W]

Ugh. I've been working on this for almost a week. But I think it's finally working now.

That's a screen-space effect on the meshes' material. It was a huge pain to implement. I had the principle figured out in a day or so. But then I had to test with windowed/full-screen, different aspect ratios, and different resolutions. I have no idea why, but on one iteration, it would be the right shape and position when starting the game, and then when I change resolution, it stays the right shape and position, but then when I change the resolution back, then it was messed up. Then I realized that with the hole staying one size (in screen space), then the hole is too big when you zoom the camera out, and it's too small when you zoom the camera in. So I had to redo the whole effect with a mask that scales with the zoom.

And just watch: Now I'll probably get a bunch of bug reports from my players saying it's not working on their machines.

 

[MLMarkland]

Ugh. I've been working on this for almost a week. But I think it's finally working now.
View attachment 37526
That's a screen-space effect on the meshes' material. It was a huge pain to implement. I had the principle figured out in a day or so. But then I had to test with windowed/full-screen, different aspect ratios, and different resolutions. I have no idea why, but on one iteration, it would be the right shape and position when starting the game, and then when I change resolution, it stays the right shape and position, but then when I change the resolution back, then it was messed up. Then I realized that with the hole staying one size (in screen space), then the hole is too big when you zoom the camera out, and it's too small when you zoom the camera in. So I had to redo the whole effect with a mask that scales with the zoom.

And just watch: Now I'll probably get a bunch of bug reports from my players saying it's not working on their machines.

Study matrices transformation in rendering and you won’t ever have a problem with this stuff. Moderate to easy linear algebra.

It’s about transforming one coordinate space into another coordinate space (for example gamespace to screenspace)

https://www.haroldserrano.com/blog/...ness of a matrix,arranged in rows and columns.

 

[Nathaniel3W]

MLMarkland thanks bro. I really do need to brush up on my linear algebra. I'll put that on my reading list. But honestly, in this material, Unreal Engine already had a single node that was supposed to all of that for me. There just ended up being some other complications, including Unreal Engine not refreshing the shader's awareness of screen size.

 

[MF]

Can't you hijack the built-in occlusion culling mechanism? In Unity I'd use layers to determine if my shader should stencil or not. I'd make a foliage layer or something more generalized and limit expensive culling calls to that layer.

Either that or use staggered raymarching to see if anything from my object of interest is still hitting the camera and cull accordingly.

I can see how it would be a bitch to implement as a screen space effect, but why would you want to?

 

[MLMarkland]

Can't you hijack the built-in occlusion culling mechanism? In Unity I'd use layers to determine if my shader should stencil or not. I'd make a foliage layer or something more generalized and limit expensive culling calls to that layer.

Either that or use staggered raymarching to see if anything from my object of interest is still hitting the camera and cull accordingly.

I can see how it would be a bitch to implement as a screen space effect, but why would you want to?

Better frame rate if done properly, for example as a full screen screen space post render loop effect

 

[MF]

Better frame rate if done properly, for example as a full screen screen space post render loop effect

Fair enough. Are you supporting mobile platforms, Nathaniel3W?

 

[Nathaniel3W]

MF MLMarkland
Good questions and interesting ideas to consider.

The game is not on mobile. Minimum requirements include a GTX 750 Ti video card.

Screwing with the occlusion culling would require digging into the C++ engine code in UE3. Not saying that wouldn't be a good solution. It's just that I wouldn't know how to implement it, and I think it would be really difficult.

In Himeko Sutori I traced (that's what we call it in Unreal) from the focus point to the camera and if the trace hits a static mesh tagged for it, I switch the material to something translucent. I would tag big meshes, like houses, to switch materials. Smaller meshes, like lamposts, don't become translucent.

I didn't like that effect because it depended on collision. If you have a bunch of foliage that's purely decorative, the trace goes right through it. The way I have it now, if any actor at all--static mesh, foliage, whatever--has this material, it will automatically cut out if it's in the middle of the screen, and closer than the player. It's not even dependent on any post processing. The material's transparency mask cuts out in screen space. (That's probably related to why its shadows are screwed up, so I had to turn them off.)

Why did I do it this way? Because it's what I know how to do. I can make materials. There are probably better solutions. But if I'm the one who has to make it, this is the one that works the best that I knew how to implement.