/v/ - Video Games

God I should get someone else to write the kernel to my battle system. Blackpilled af from constantly failing it.

>>16418145

What exactly are you trying to make, and what went wrong so far?

>agdg dies

>I become productive after a months long hiatus

Coincidence? I think not

So what are the projects currently?

Reposting from previous thread in the hope that some anon who knows his way around godot can answer:

>try to make a snake game on godot to ease into gamedev

>make one tilemap for displaying the snake/food and another for displaying the walls

>for now using a hardcoded array to define the level, stored in the script of the snake tilemap

>set other tilemap as child so it can inherit the array

>script the wall tilemap to automatically fill the playing field with walls taken from the array

>can't access array

>after some research manage to fix it by changing the inheritance of the wall script from the wall tilemap to the script file belonging to the other tilemap

<the set_cell() method in the wall script works with the correct tilemap without needing any prefixes

<the tilemap isn't mentioned anywhere in the code

What the hell? What exactly is the difference between scripts belonging to different nodes when the contents are identical?

>>16418387

I'm confused, post a picture of your node setup.

>>16418353

I'm still technically working on Autumn Willow (that puzzle game) but it's been so slow and lazy I might as well be a nodev at this point

>>16418353

Demo Day is 6 days away. We will all find out real soon what all the lurkers [spoilers]including myself[/spoilers] have been up to.

>>16418560

What's the puzzle about? I mean, mechanically speaking.

>>16418616

Good job, newfag, you really nailed that spoiler in spoiler

>>16418628

spoiler inspoiler

>>16418639

fuck.

>>16418640

oh wait, it was obvious

>>16418647

Tell me your secret.

>>16418628

>calling me a newfag when I do the 8chan part of spoilers correct, but the part that's identical to cuckchan wrong

>>16418669

I literally spoilered how to do it in my previous post anon. Why don't you guess?

>>16418669

Two asterisks on both sides, like a normal spoiler, and cuckchan-type bracket spoilers in between them.

Now just like make game.

>>16418677

>>16418683

I don't deserve this power.

>>16418684

can'tyoujustuse4asterisks?

>>16418694

No, because the second pair of asterix will close the first pair, ending the spoiler.

>>16418538

Alright, I slightly misunderstood how "extends" works, I had assumed it meant inheritance of a specific object instance so that methods that change object properties unambiguously change the one instance that was inherited from. Apparently it inherits from the general object.

Now my question is what makes object specific methods apply to the specific instance the script is attached to.

I don't think it's relevant any more, but here is the basic setup:

Node2D

L__TileMap (Tiles of snake, background and food)

L__TileMap2 (Tiles of walls)

Attached to TileMap is a script that extends TileMap and, among other things, contains the hardcoded array with the level layout.

Attached to TileMap2 is a script that uses the array in the above script to set tiles in TileMap2 to draw walls. This is done through the set_cell command.

I needed to replace the "extends …" line that was automatically generated with "extends res://locationoftheotherscript.gd" so that the script could use the array from the other script, so I was confused why set_cell worked correctly without any prefixes despite TileMap2 I didn't read that the line I changed was "extends TileMap", not "TileMap2" not being mentioned once in the code.

>>16418677

newfags can't nest spoilers

>>16418697

dudeIamusing4asterisks. Itworks

>>16418752

But notice how each one of your words is a separate spoiler and requires to be hovered over individually whereas mine shows several sections with one hover and only hides the second layer. Does that make sense?

>>16418756

I noticed shortly after my post.

>>16418719

Extending your TileMap script for your TileMap2 script is not the proper way to do what you want. Variables in gdscript are public, so you can just use get_parent().walls_array in the TileMap2 script to access the array in the TileMap script. The reason set_cell worked in the TileMap2 script is because your TileMap script extends from the base TileMap class, which is what defines the set_cell method. You should probably change your TileMap2 script back to inheriting from TileMap (the built-in class, not yours).

>>16418760

>Variables in gdscript are public, so you can just use get_parent().walls_array in the TileMap2 script to access the array in the TileMap script.

Thanks.

>The reason set_cell worked in the TileMap2 script is because your TileMap script extends from the base TileMap class which is what defines the set_cell method.

That's not what's confusing me. It's obvious to me that the method set_cell would be inherited from the other script that extends TileMap, but why does the method affect the TileMap2 instance?

>You should probably change your TileMap2 script back to inheriting from TileMap (the built-in class, not yours).

That's probably true, but doesn't really answer my question.

>>16418647

sauce on pic?

>>16418796

Thanks luv

>>16418817

Like Stealth Inc.?

>>16418786

If a method is called without prefix it is assumed to work on the node the script is attatched to and is basically the same as self.set_cell(). If the script that calls set_cell() is attatched to the TileMap2 node then that is what is affected.

>>16418353

I'm working on Trigger Control, a stealth game with a focus on realistic firearm handling and modeling of computer systems. As of late I haven't been working on it as much as I would have liked, but I am still working on it.

In my case I am still deciding on what engine to tackle (Godot or Unreal) before considering development of some templates, probably a shooter with some pretty unbalanced stuff inspired by SAO-GGO (minimal RPG elements affecting player loadout and attributes), if anything basic mechanics will come first but I also need to implement some player avatar customization to apply the gimmick.

>>16418890

Thanks, that does some amount of sense. Is there a quick way to check where a script is attached to, or is that something I should clarify via filenames and comments?

>>16418353

>start working on an FPS idea I had for some time

>write own movement code for the sake of learning

>works on surface level but doesn't allow for good control in the air

>realize I have to discard most of it and rewrite

>haven't touched the project in a week

>>16419413

So, will it be called Dragonscape, or Rune Quest?

>>16419554

>Runescape with a top-down adventure feel like a GBC game

Holy s hit do it

>>16419556

What is it about?

>>16419554

The 'game' is a generic rpg to test the engine on and the midi playing isn't part of it. When it's finished I'm planning on a sci-fi themed game moreso than a traditional rpg.

>>16419556

just saw screenshot while scrolling /v/. what game is this about? got a link?

>>16419413

Aren't you the nodev that was looking for someone to do music for you, specifically NES music? I could possibly assist you in making a soundtrack for your project with no costs.

Would you like it if I sent you a portfolio?

Here's a sample of my work. The vid related reminded me of this piece: https://files.catbox.moe/wpyn1e.mp3

>>16419573

That setting is absolutely perfect, because I always do a lot of "spacey" pieces. In fact, listen to this piece embed, it has three different compositions in one. Although, I used the VRC7 chip with the base 2A03 chip, so it is technically not NES authentic but FamiCom authentic.

>>16419557

Taught about that actually but I want to do something original.

>>16419596

yeah and that would be great! I really liked the sample.

>>16419626

That sounds awesome, I have these folders from my MEGAupload that contains more chiptune stuff with some contemporary music.

The Unattainable Night(my latest album): https://mega.nz/#F!TaRWTSjI!QIUUAam_u_0c191WEVAk6g

Miscellaneous Works: https://mega.nz/#F!vLQUHIIK!iCsf-CFcpRZzKAz-Yktg4A

>>16419359

>>16419324

It's coop too

>>16419364

I wouldn't have scrapped it, I would just use state machines instead.

>>16419670

Thanks man, love it! Do you have a contact? If not I put occasional updates here. Once my exams are done I can put more time in designing the actual game.

>>16419848

You could message me at muscises@cock.li or at my soundcloud profile, and if you have any requests as you are working on the project I'll try my best. Appreciate it, and hope for the best for you.

>>16419755

So risk vs reward.

>>16419755

Everything looks like really tiny versions of those "Madness" characters.

>>16419813

>Having friends

Look at this normalfag

>would've had something for demo day but decided to overhaul a bunch of mechanics and ended up not having enough time to work on them

m-maybe next time.

>>16419143

>imgui

my nigger.

>>16418353

Just recently ended my dev hiatus and started working on my hobby project again.

No concrete title atm.

Been solo dev'ing this project for a few years now.

Basically it's a project that's currently focused on AI (hybrid AI approach, slightly nuanced), and exploring this in new ways (neat mechanical implications); in a way that has to account for my current skill based mechanics (fromsoft esque).

I was also doing stuff like dual contouring on the GPU (nearly finished) w/an elder scrolls esque cell/chunking system.

>>16419085

HTN is pretty great, and is generally a lot more efficient than GOAP due to being forward-propagating rather than backwards-propagating; i.e. pruning large sets of tasks via the composite/method/primitive layout w/o a necessity of a best path heuristic alg.

So, you made the right decision there.

An annoying part you should look forward to is world state, and getting the correct (important) data to the AI without it turning into a cluster fuck.

Also almost forgot to mention that creating a domain GUI is extremely helpful; as otherwise crating new AI is a bit of a pain (w/all the p-cond, effects, expected effects, child tasks, etc etc).

>some dude released an open source HTN implementation

I'd recommend to read the game AI pro article (multiple times) if you want to jump in (I think I found what open source one you're referencing, it's basically a 1:1 to the spseudocode of the article).

You could also use that open-source code as a stepping stone to programming your own.

The reason why is because that article provides a few approaches (each with their own caveats).

So there's different approaches that cater to the different needs of the dev (more than one way to build the nuanced parts of it, that solve different potential weaknesses, and so on).

>>16418353

I'm not sure I'll actually finish the game before demo day, but I've been working on a brainfuck interpreter that maps part of the memory to a screen, and want to make snake with it. I'll probably at least release V1 of the interpreter plus source code on that day, though.

>>16419556

>Ludum Dare was this weekend

>had plenty of time to participate

>no one here posted a reminder

God dammit, I even like the theme for a change.

>>16420696

> Also almost forgot to mention that creating a domain GUI is extremely helpful; as otherwise crating new AI is a bit of a pain (w/all the p-cond, effects, expected effects, child tasks, etc etc).

Part of what made me dislike the GOAP solutions. It's quite hard to organize in code-only implementations.

Visual domain editor is on the to do list, but I'll tell the dev that he should probably start with a visual debugger as that's more pressing to anyone who actually wants to use it.

>>16412636

>But I figured I couldn't make it as good as Tyrian

You certainly can make a better game than Tyrian anon, you don't need any of that pansy shit for a good shmup.

>>16421097

It looks great!

Can you scrub her back?

>>16421342

While looking at a painting of Mt Fuji?

>>16422630

patreon.com/PunjiStudios

twitter.com/PunjiStudios

for anyone not wanting to click shill links, a quick rundown of the game is as follows

>A disorienting arcade style "shooter" where you need to make your shots count because you only get one and you control its trajectory the entire time. taking inspiration from movies like Wanted, and games like Superhot. The games main objective and goal is to reach the boss room at the end of the level and eliminate the target by avoiding collisions with anything that might stop your bullet.

>As you ascend the tower, each boss has a significantly larger floor plan, each floor is also randomly generated, making it that much more difficult to reach them. Currently we only have straight tiles to the boss, but are adding branching rooms to allow for full turns and dead ends.

>>16422708

>8MB .gif's

use webm's ya dingus

>>16422716

yeah you right, they were dying on the upload.

and they are chugging on the load now haha

>>16422708

This will give people motion sickness.

>>16421342

Youll be able to scrub her hair. Arms too if i have time.

>>16422751

the framerate in the gif causes some uneasiness with myself, but when the game plays at a stable framerate no one seemed to comment anything aside from being disorienting because the world is spinning making everyone slowly turn their heads to try to compensate.

>>16422865

Precisely.

>>16422905

as in confusing, not motion sick

you can also control the spin with the side buttons on the mouse to spin faster or slower.

having the playering spin makes it so the level is forever evolving instead of the same patterns every time you come to a repeat tile, you'll be coming at it from a different rotation.

noted none the less, thank you.

>>16422708

>>16422710

Damn you just gave me a massive surge of nostalgia. Played a very similar game on my cousins laptop more than a decade ago.

>https://missile-game.bwhmather.com/

>>16423006

had completely forgot about the missle game…

very similar in style haha.

>>16418817

Kind of a progress post. Realized that I had to make it so clicking anywhere on the screen that wasn't a menu button had to hide my menus, fixed that and have alt keys working with input. Some more polish and I can get to working on settings and game start options.

>>16423426

it's all from quickly recording footage from a 48 hour gamejam game. which was also how long the game's been worked on.

Is there any good literature floating around about how to do good randomly generated levels?

>>16423590

What kind of levels? I have some PDF's on procedural floor plans that look like normal building floor plans. If you think you can get something out of those I can post them.

There's also this wiki.

http://pcg.wikidot.com/

Honestly if you want to make good procedural levels get good at making hand-crafted levels in the genre you want.

>>16423627

Since this also covers /vm/, what I want to do right now is randomly generate a number of rooms in Super Metroid. What I'd do is run a script or program against the vanilla rom, do the things, and then spit out a new rom to play, much like how the item randomizers do it, but on a larger scale. I know how romhacking works, and I've been tinkering with stuff over the last few months, and can comfortably read and write a good chunk of data structures within the game itself (though only on the rom file, nothing with live memory)

It would be a ""roguelite"" in the vaguest sense of the word. But basically, you'd choose a game length, go against maybe 6-8 mini/bosses across 3-5 areas, and all of the items may or may not appear depending on settings. So obviously, I'd have to factor in what abilities a player has access to for any part of the game.

My naive approach was to evenly distribute item locations across a 2D plane, and then figure out where the items are located, then generate room shapes/areas and give each one a "purpose" (connection, enemy challenge, treasure, one-way exit from a boss, etc) and also track abilities and movement required for each one. At that point, the room data would be generated

For each individual screen, I was thinking of reading the raw room data and based on connections (eg open north or south passage), it would figure out the block types on that screen, and add them to a "dictionary" for that screen type. Then, it could use this as a bias when it generates the rooms.

It doesn't have to be perfect, since this is only the proof-of-concept stage but I don't want to shoehorn myself into doing something retarded

>reading code I wrote half a year ago after not doing any dev since then

>it's like looking at a smarter version of myself, can't understand anything despite it all being autistically documented

fuck this shit

>>16424013

>pdf

Oh wow this came at such a perfect time. Thanks anon!

>Add 3000 LoC without being able to test

>Decide to test a little bit just for the hell of it

>Oh fuck why is my computer frozen

>INTEL-MESA: error: ../mesa-18.2.8/src/intel/vulkan/anv_device.c:2005: GPU hung on one of our command buffers

hahahahaha kill me.

>>16424305

Ok I figured out the problem. I was using the same staging buffer for both my indirect draw commands and vertex data, so there was probably a data race which caused junk data to be fed in for my draw command, causing my shitty integrated Intel GPU to idle (and also making my screen apparently hang because that's also how my WM is rendered). I just fixed it by using two separate staging buffers since I didn't feel like messing with synchronization anymore than I already did.

>>16424485

Too late I sent the hitman already.

>>16424507

oh fug.

>>16425928

Should have made a closer shot, at this distance I can't say anything more substantial than "looks pretty good".

>>16425928

looks pretty good

>>16425928

Don't seem wrong, but yo should check how shoulder topology is made because that's always a bother when you are starting

>>16426427

vuvuzela clearly

>>16426427

The one that sounds like a mixture between a kazoo, flute, and saxophone? Probably an exotic instrument of some sorts.

>>16426427

It sounds like something between a flute and a reed instrument. You can search through lists of Chinese and Japanese wind instruments and see if any of them sound close to what you're looking for.

Speaking of music, what's that African drum instrument that makes that bassy bu-woomp sound?

And that one flutey instrument from jungle or Asian soundtracks? Most frequently heard in mgs3 I think

>>16426427

After some looking around it seems like one possibility is the Chinese Suona.

>>16426487

There are a couple of African Drums that make that sound, but it could be the Bougarabou.

>>16426516

No, it's usually a slower, elastic and elongated sound f that makes any sense? I'll try to find a song

>>16426506

>Chinese Suona

I don't think so. It sounds like a wind instrument that has multiple 'pipes' at diferent pitches similar to bag pipes (I don'y know music lingo). The chinese lusheng is the closest thing I found

https://invidio.us/watch?v=G6JJUd8MfKA

1:11

>>16426487

The are japanese drums that do that bu-hwoop sound.

>>16426487

Maybe you're thinking of the tabla?

>>16426427

One thing to note is that at the beginning the instrument is being played in fifths and it sounds like later it's being played in octaves, which might mess with the timbre. If you asked me about a generic Asian-sounding wind instrument I'd probably say shakuhachi flute.

>>16427038

Embed related.

And the latest of my vulkan woes:

>INTEL-MESA: error: ../mesa-18.2.8/src/intel/vulkan/anv_queue.c:47: execbuf2 failed: Invalid argument (VK_ERROR_DEVICE_LOST)

>INTEL-MESA: error: ../mesa-18.2.8/src/intel/vulkan/anv_queue.c:247: vkQueueSubmit() failed (VK_ERROR_DEVICE_LOST)

>>16427038

Yeah Tabla is closest to what I'm thinking of I think

>>16427045

I'm guessing why the driver/GPU is crashing is because of some kind of synchronization error, and I've narrowed it down to a vkFreeMemory before I reallocate a buffer to expand it because when I remove that line my program magically works. The problem is, I can put as many vkDeviceWaitIdle()s I want around it, and I still get the same error.

>>16427045

>Going anywhere near intel vulkan drivers

Don't do this.

>>16428244

Also just making sure to point out, I just need an idea for determining a numeric value, not the actual world gen.

>>16428244

You simply set the seed at the start of your procgen function. Said function should take at least two variables, the world seed (generated once at world creation) and an X/Y offset to see which chunk you're generating. You do have to ensure your PRNG generator always returns the same values even when multithreaded, so you should either have one instance that gets locked to one chunk at a time, or one instance for each chunk if you want to fully multithread even the chunk loading.

Then again, if you want actual user interaction with the world you're going to have to save chunks to disk anyway, so you'd only need to gen them once.

>>16428366

Meh, I disagree with you there. Like I enjoy this board and the people on it, so even if I made a game that got popular I'd still use this place to argue about Sonic and other dumb shit. Like /v/ is fun, so I have no issues coming here. Even if I had to hide my game because of some bigshot publisher, I'd still post here anonymously.

>>16428289

No, I don't want user interaction with the terrain.

>>16428459

That's cheating, you should at least ray freehand.

>>16428430

>>16428244

I'm confused,don't most languages come with a RNG function? You'd just make a global seed like anon said, then some kind of offset/sub seed for each chunk

>>16428369

Almost anon.

>>16428491

That's entirely dependent on how your RNG works, you dumb nog. If your seed is 64 bits, you can have for instance a 32 bit world seed, 16 bit x seed and 16 bit y seed, then just say totalseed = (worldseed << 32) + (xseed << 16) + yseed;

I didn't assume that was your problem since it's so fucking simple that you should really know how to do it before you can come up with a decent RNG. This is honestly like saying you don't know what a modulo is.

>>16428369

Should be ready today or early tomorrow.

>>16428463

>>16428663

Fuck off carlos

>>16427809

I have no choice; this is the only vulkan machine I have. I'm thinking to work around this I should just pre-allocate big blocks of memory of certain types and then divy that up for my various buffers, then free that memory only at the end of the program.

>>16428879

Why?

>>16429150

>Was it an engine?

yes and no. Savage Empire was built in the same framework as Ultima 6. I don't know if you could call it an engine as we would understand it today (being invclusive of a large scale toolkit like Unity or Unreal), but the underlying code is much the same and there is an editor for the U6 derivative games

to my knowledge LotR is a different beast entirely, and the only thing shared is the sprite style

>>16428244

Just use a two-dimensional noise function, you dingus: https://en.wikipedia.org/wiki/Perlin_noise

>>16429250

Game Maker isn't that bad

>>16428244

If you're going the >>16429277 route, use OpenSimplex.

Also, this is unrelated to any project I'm currently working on, but I had a vague idea on how it would be to make an erotic text adventure game. I remembered I have no idea how to put queries in the text dump (eg for flags and stats) and quickly gave up on it.

Is there an engine or whatever that's good for that?

>>16429281

It's pretty shit. Framerate is still tied to room speed and it's overall ridiculously bloated. There's a bunch of shit wrong with it, too, but it's not coming to mind right now. I just remember the new redesign looking like absolute shit.

Godot's not that great either, between being unfinished, lacking in documentation, breaking functionality between updates several times and overall being dick-sucked mostly because it's FOSS, meaning it can do no wrong in the eyes of the typical Linux crowd. That has some potential though.

Making your own engine is by far the best option, though.

>>16429295

Greatest problem with Godot is the OOP.

>>16429312

OOP is fine for the paradigm, especially because it makes it far easier to separate your designers from your programmers (eg. I was making a game with a non-programmer friend a year or so ago, and due to the design of Godot, I could write supporting modules that he could entirely work with without code. He could make behavior requests and I could work those in while he worked on other things).

I think the biggest problem with Godot is using their own language instead of making the core programming language something better that already exists. GDNative and NativeScript (I'm still not exactly sure the difference) let you integrate other languages, but it's annoying that the easy and seamless core language isn't as easy and seamless as it could be. As a professional Python programmer (I know, calm down, it's not my primary language, but I do use it at work), using GDScript always feels like an uphill fight to unnaturally do in a weird way the things that Python already makes easy.

I'd just prefer Python or Lua or something. Even Ruby. GDScript gets the job done, but it feels like a bad hack in a lot of different ways.

>>16429312

I'd argue it's their insistence on using some proprietary made up bullshit programming language, and one based on python at that. Last I checked, they had experimental C# support but they're not really bothering to put any real emphasis on making that usable.

>>16429293

Twine or Ren'Py will work fine.

>>16429322

Yeah it's basically a gimped version of Python, the inability to use multiple inheritance and other OOP stuff is annoying.

>>16429343

Is this using boids? I assuming you're using some kind of spatial grouping on them

>>16429358

nah, it's just 10 groups of 100 pawns each, they never change groups

>>16429326

>>16429336

>>16429322

It's not even that difficult to use GDscript. It's strongly based on unity so the majority of it's techniques will apply here as well.

>>16429322

I criticize OOP because I want an ECS engine that's highly vectorized and automatically multithreaded. Until that exists no one can make the spiritual succesor of SupCom FA in Godot.

>>16429326

>based on python

You mean in syntax, right? Isn't GDScript super fast? I'd think that was the point.

>>16429383

>I criticize OOP because I want an ECS engine that's highly vectorized and automatically multithreaded

Well, maybe you should look into a real man's engine, like Unity

>>16429184

Thanks, Anon. Found a couple editor options actually. One seems to be python based which is great since I've done some dicking around in Python, mostly for IRC bots.

I did watch some game play vids of them both, and yeah I don't know why I thought of them as so similar. it's just the slanted view and character sprites really. I spent endless hours as a kid playing them both, my sister and I even nagged and nagged my Dad to buy my family 486's first sound card to hear the digitised battle sounds. I guess after 20 years I was rusty on the mechanics, so I apologise for that. Now I have to slap some test stuff together in Asperite and see if I can figure things out.Thanks again!

>>16429373

It's not that it's difficult, it's that there are a lot of little things about it that are very surprising. The way that some types handle are especially surprising, and a script itself being a class is weird as hell. You have to re-learn a lot of things, and there is a disconnect between some things that should be identical between script and non-script (for instance, connecting signals can be done through a script or the interface, and is shown differently for each, when it should probably be enforced to just have one canonical way).

Some syntax things with GDScript are also really awkward, like the distinction between a dictionary and an array.

Testing out new ideas in GDScript are also annoying, because there's no way to just run a standalone GDScript from the command line or anything like that. Having a REPL for gdscript would be good. Take something as simple as format strings: http://docs.godotengine.org/en/3.1/getting_started/scripting/gdscript/gdscript_format_string.html#format-method-examples

In order to test these examples out and try it for yourself, you have to have a runnable Godot project and work it into some particular node's script in a particular place and print out the result. At least it's got a decent enough debugger, but having to use a syntactic element (the "breakpoint" command) typed into the program itself to use it is really fucking weird, and makes it awkward to properly debug a running program.

When you get used to it, GDScript is actually a pretty decent and useable programming language, but it has a lot of very specific and weird behavior and a lot of annoying limitations that would have been avoided by using something else that existed already.

Using GDNative is doable, but then you have to completely program your own implementation or trust one that you find. C# is alright, but not perfect, and also still considered an experimental language for Godot.

For an amateur game dev, it's good and gets the job done. For a professional programmer who expects certain things from a programming language, it's a huge compromise in a lot of different places (though I do love a lot of things about it, like being able to develop modular scripts that non-programmers can interact with through export variables).

>>16429545

I've heard of that, but I don't want to pay them.

>>16429550

I see. I'd be interested in knowing whether any open source engine or framework goes that far. Never looked.

Personally, I don't have a problem with paying the Unity devs, because I'm a filthy capitalist. I like the idea of paying others to do work so I don't have to. In this case developing and maintaining an engine that has more features than I could realistically implement myself.

>>16429336

>Twine

I installed it and it just feels like a barebones html-enabled text editor. Was there even a sample project with it? It didn't feel right

>Ren'Py

Originally I hadn't tried it because I assumed it'd be mostly coding (and I assume it will be), but it looks like an IDE of sorts right out of the box. Probably going to go with this. Thanks.

>>16429568

FOSS is the future, old man! I don't like the idea of making games in C#, and I've heard so much shit about Unity that I'm jaded by now.

>>16429716

Twine is probably better for just text, and it's more friendly if you don't know how to code and want something friendly

you'll want to get the SugarCube 2 format, which I think comes with a sample

>>16429893

What's wrong with C#?

Are you complaining because it's not "as pure" as C/++ and runs on a virtual machine, or because of Microsoft's hands in it? Mono exists, for example.

>>16429893

Unity is an excellent engine. Sure, it has its problems (as all software solutions of this scale that have to cover such a wide area of use cases have), but it's by no means bad. The go-to anti-Unity argument is to hold the engine responsible for the shovelware made with it, which is complete bs. If anything, the fact that those completely incompetent "devs" can produce something vaguely resembling a game with Unity is a testament to its usability. They wouldn't be capable of doing that from scratch with older engines like Source.

>>16429913

Mostly prejudice, but my biggest issue with Unity is the bad notion I have of it from years of anons complaining.

>>16429924

Also payment. I don't want to pay them.

>>16429893

I also love FOSS, but I won't sacrifice the quality of my games to be FOSS at all costs. I use Blender, I support Linux and I plan to make parts of the games I work on open source. That's good enough for me.

>>16429924

if you want to use C# Godot is completely FOSS

>>16429924

C# is probably one of a very select good things that Microsoft has made over the years.

(Visual Studio is also pretty good if you're into that sort of thing, but there are plenty of FOSS IDEs available too)

>>16429969

Isn't C# just Java 2.0?

>>16430051

It is exactly like Java, but good.

When it first came out, C#1.0 and C#2.0 were literally identical to Java as much as possible, to coax Java programmers (probably mostly pajeets) onto their new product, and it worked. Keywords, namespace stuff, it's basically a 1:1 carbon copy, the only real superficial difference is that Java uses naming likeThis, while C# uses names LikeThis.

There's a few other differences too, like how any function in Java is extensible by default (a security nightmare) and generally wonky language. C# is essentially Java but done properly. I'm actually stuck on C#4.0 because of technical reasons*, but throughout the years, there have been a large number of performance and VM optimizations.

*: I have Win7 with Windows Update telemetry neutered. Because I prefer to use VS, I'm using VS2010 Express C# edition, which is hardcoded to only support up to 4.0. However, newer versions of VS need that Universe C support pack or whatever that Win8.1 provides, but I cannot install it, even if I enable Windows Update again.

>>16429250

Unfortunately I realized this only when it was too late.

>>16428184

Made a new level to test some things. Have a video. Not making demo day I think, too much to fix and I want this to be something worth playing.

>>16430452

Maybe add easing at the end of the arm extension.

>>16429924

Maybe do your own research?

Instead of listening to the past autists that suffered from the dunning-kruger effect.

Yeah, /v/ is a savant when it comes to gameplay, but development… wew lad do they have no fucking clue.

Life advice: Nobody can convince you to change your mind; you're the only one can do so.

Otherwise, if you "ask" what decision to make, you're just looking for social affirmation of the decision you've already made.

Make up your mind to be able to change your mind, research it, then make your decision w/critical thinking.

>>16430840

>Unity

<Unity

You're asking for the world, anon.

>>16429150

Like many other good old games LotR V1 has open source implementation. You can check one out.

If by some chance I would be able to compile open source implementation of Golden box engine by demo day I'd show it. There very little of my work there yet, thought. Big idea is to make tools for Godot similar to those of Golden box engine.

>>16430930

Less about unity and more about your decision making process, and the way that blends into your goto heuristics for decision making.

You've said it yourself, you've already made up your mind, and I touched on what you're doing with your line of self-leading questioning.

That's fine and all, however, the way you think is self-detrimental; which is my point.

Use critical thinking more and also realize that confirmation bias exists.

Overall tho, u do u, no skin off my back.

>>16429924

You can pirate unity pro if you're concerned about DRM

>>16431105

If you want to make a game, no. If you want to work on an interesting engine/learning project, sure.

>>16431105

From someone who's spent the past 2 years writing an engine instead of writing the game he wanted to make:

Probably not. "Vulkan" is more wrangling with hardware aspects–memory, concurrency–than what most people consider graphics programming (i.e. writing fancy shaders).

>>16431114

I do want to make a game, but I am also interested in learning more about programming.

It kinda ruins my cred if I like to do all these impressive low level nonbloat programs then use a pre made engine.

>>16431032

Ok, I've taken some time to articulate my full thoughts, since until now I've just half-explained myself retardedly.

Regarding the percentage payment, the complexities of accounting are something I wouldn't want to make worse. This year I generated three fines all revolving around a single mistake with my earnings/tax statements that my shithole country misinformed me on. It's one thing to have some account that steam dumps the money on, but to have to state sales numbers to Unity periodically? Missing one month and making a mistake in stating earnings have already caused me so many problems that I just don't want to deal with it, regardless of greed, which I do have in plenty amount.

Regarding development, I don't want to put all the effort into learning Unity just to someday have to pull all that effort again into migrating to Godot, which I've always planned to end in anyway since I want to switch to FOSS on principle as much as possible. Migrating tools is a major undertaking. This is why I've stayed with 3ds Max even though I've intended to switch to Maya for the longest time.

I'll admit I've been discouraged from Godot lately. Ever since Unity announced it's ECS-style stuff, Godot seems to have been left behind. Godot's 3D capabilities are still limited. Its devs make it worse by focusing their efforts into accessibility with their visual node programming crap instead of delving into deeper features.

Lastly, I'm studying CS and the idea of an open source engine I'll be able to add things to and modify really appeals to me from an long-term effort investment perspective.

>>16431128

If you try to do everything you won't finish anything.

>>16431141

That's really cute dude

>>16431141

I don't like, how water doesn't give distortion. It looks fucked. I know, it's simpler this way, but could you at least make water opaque?

>>16431128

>It kinda ruins my cred if I like to do all these impressive low level nonbloat programs then use a pre made engine

Then use your low level non-bloat programs to make a game instead of starting over again

<but I haven't finished any of them

Then you haven't made any impressive non-bloat programs, and have no cred to lose.

>>16431179

I can add some distortion. I'll have to bump up reflections.

>>16431105

Are you looking for this?:

https://github.com/GPUOpen-LibrariesAndSDKs/V-EZ

But it looks like you keep giving up prematurely. You have to be able to do all of those things to make game. If you don't want to use an engine but don't want to make one either, I guess you could use OGRE?

>>16431154

Good writeup.

>Regarding the percentage payment

To clarify, just for unity that is, there's a yearly licensing fee that's non-optional IFF you had made 100k+ earnings/funding (for your game company or game sales if you're the sole proprietor) in the previous fiscal year.

Sounds like the main issue is your country's tax system, and that's implicit if there's any money made.

>effort into learning Unity just to someday have to pull all that effort again into migrating to Godot

Check out decoupling design patterns; such as the dependency inversion principle.

Decoupling patterns allow for the migration of your high level codebase from engine to engine, but of course you do need to rewrite the low-level execution code.

It's worth knowing and practicing even with a codebase that interfaces with an engine you like.

Although, with Unity's ECS implementation being highly engine specific it's questionable if that's possible w/o creating your own ECS API in the engine you'd be migrating to.

>which I've always planned to end in anyway

Humm, it seems that Godot matches with your philosophy, but it lacks some core criteria for committing; such as an ECS API/design.

You could always use an open source library to build your own ECS API into Godot; which meshes well with the long-term investment perspective.

However, my godot knowledge is surface level so I can't say how difficult this would be (maybe its been done already?).

Although, the limited 3D capabilities are worrisome even with that issue hypothetically resolved.

>This is why I've stayed with 3ds Max even though I've intended to switch to Maya for the longest time.

Sorry to ask an antagonistic question, but isn't maya closed source?

Wouldn't it be better to switch to blender since it complies with your philosophy?

>discouraged from Godot lately

>devs focusing their efforts into accessibility

Humm, from an outsider's perspective the accessibility part really makes me suspicious of their motives; i.e. more (less skilled) users = more patreon bucks, but then again, more funding equals more time they can spend working on the engine itself (as to implement more difficult features and so on).

So it could just be a long-term tactic so they can have more funding (for the talent, so they have more dev time) to implement the features they really want to implement.

Godot Pros

>matches philosophy, FOSS

>hypothetically can solve ECS issue w/library

>long-term payoffs

>no licensing concerns

Godot Cons

<questionable engine direction

<limited 3d capabilities

<initial timesink to resolve no ECS issue w/own API

<questionable compatibility of own API due to updates & godot API changes, i.e. must maintain homebrew API as engine changes

Unity Pros

>ECS

>cutting edge 3d capabilities

>no need to maintain homebrew API

>short to mid-term payoffs

Unity Cons

<not FOSS

<licensing concerns

<need to learn API (initial timesink)

<potential of requiring codebase migration

<Unity

>>16431252

Let Godot be your last choice. It is a very bug-filled engine and I've barely scratched the surface. There is a reason why 2d in 3d is what I went with.

>>16431283

Godot Cons

<Bug ridden

Unity Pros

>Surface level bugs are fixed quickly

>>16431277

Is it written in HolyC? No? Garbage. idk, but it uses python which is a big no-go for games imo

>>16431305

It's written with C++ but there's an optional scripting feature which uses python most of it's addons are written using C++ but i want to know if i can change it's renderer from DirectX9 to maybe OpenGl or something although the faggots and sicrra abonneded the engine to make another shitty java script engine

>>16431315

My first question would be: why?

Tho, not sure about your question. If it's open source then I suppose it's possible… but is it a reasonable investment of time?

>>16431318

>Why ?

i want to publish games on other platforms other then windows and at the same time i can emprove the engine renderer and maybe and Vulkan

>Is It open source

Yes but there's no documantation to it

>Is it a reasonable investment of time ?

this is weird but i have no coding expeirence what so ever in C++ and i'm total beginner so yes i have autism.

>>16431252

>You could always use an open source library to build your own ECS API into Godot; which meshes well with the long-term investment perspective.

Yes, that would be the plan, though I'd stick with OOP for the first game or two. GDNative allows C/C++ modules to be created, so one can offload as much as one wants to native code and use GDScript only for very high level stuff, or potentially not at all.

>the limited 3D capabilities are worrisome

That's my greatest hangup right now.

>Maya and no Blender?

I should have mentioned it too. Maya has been the aim for a long time, but I have gotten more and more confident in Blender, I just haven't really looked into it yet. Like with Godot, I fear there might be something crucial that Blender lacks that I won't realize until much later, though that seems unlikely. I work in VFX so the plugins are a big deal (FumeFx, Krakatoa, Rayfire), but once in gamedev it won't matter. I've heard the UI and pipeline is very different though.

>accessibility part really makes me suspicious of their motives

Clearly they are prioritizing wider adoption. Not sure if it's about money itself or they're just playing the long game. If they could get much of the mobile and 2D indie market they'd be much more secure for advanced 3D, which is probably too difficult to get in a reasonable amount of time so as to be priority #1.

>>16431283

>There is a reason why 2d in 3d is what I went with.

Like anubibro says, It would be smarter (and all-round more practical considering my skill level) to start simple in Godot, with a 2.5D game at most. Thus, the temporarily limited 3D and OOP aren't immediately relevant. With a more humble initial aspirations but plans for the future, my table would be:

Godot Pros

>FOSS

>No licensing

>Long term payouts

Godot Cons

<Insufficient documentation

Unity Pros

>ECS

>Mature 3D

>Short/mid term payoffs and big plugin/module market for things above my head

Unity Cons

<Propietary and closed source

<Licensing and legalese

<Unity

You have helped me figure things out by challenging me, I am now more secure in my decision. Thank you.

>>16431328

Solid plan.

Happy to assist, and I hope we have more cooperative socratic style debates in the future

>>16431330

Too tired atm to get involved in the needed discussion. I'm sure someone will sort u out with what needs to be said tho, lurk a bit

>>16431330

Why do you want to use Construct Classic? It's old and doesn't meet your requirement of supporting platforms.

It's open source, so yes, it is possible. Yes, it is a lot of work. No, it is not worth your time.

You wont just be porting the d3d graphics to opengl, you'll have to port other components, like window creation, input handling, etc. as well.

At the end of it, you'll have written enough code to make your own multi-platform engine.

I'm guessing you have some nostalgia for toying with Construct Classic in the past and want to make it available to other platforms. That much is fine, but you can forget about accomplishing that if you don't learn how to program first.

>>16431105

You want to learn how things work under the hood or to make game?

Pick only 1

>>16431141

I want one

>>16429411

britbong as in the britbong(eggbong/mrbtfo)? or just a british accent?

>>16429343

Do you use some kind of spatial data structure as well for the navigation?

>>16431982

>drag and drop

Does anyone really even use that for Game Maker anymore? It's a legacy feature, but 99% of all tutorials are for the GML language. You can talk about the quality of the language, sure, since it's not good, but I don't think Drag and drop is really that valid a criticism.

>>16431988

I've seen too many people admit to me that they use GameMaker's drag and drop feature to quickly push out a prototype.

>>16431982

>fine depending on what game you are making

>get you laughed at by real programmers

Real programmers don't give a fuck what tools you use as long as you get the job done you poser faglord.

>>16432002

Second method is superior because I like symmetry.

>>16432006

Wrong, the first method is a lot better

>>16432006

consider suicide

>>16432002

I accept both as long as they are consistent within a project.

Personally, I started with having { on a new line when I first learned programming, because that was the first style I've seen and it stuck with me. When I went to study CS, I begun to put the { on the same line as the expression, because it saves you tons of space when you need to put some code on a sheet of paper during an exam. I just started using it everywhere, because why not?

I watched a series of interesting talks about programming (yes, there are more than just the DOD talk by Mike Acton) and one argued in favor of staying within column 80 and why it's not an antiquated concept. He made good sounding points about programming vertically as opposed to horizontally, so I gave it a shot and I liked it, but putting the { on the same line doesn't work cleanly with a vertical layout. So I'm now back to putting the { on a new line and I think it'll stay that way, because I put it there for reasons other than taste.

>>16432065

The only sound argument for 80 columns is rotating your screen to increase how many lines of code can be on screen at a time.

>>16431141

>ears

Is she an elf or something?

>>16432065

>staying within column 80

This is new to me. But the inline { wouldn't change much and still conserve vertical space. I don't get it.

>>16432152

>This is new to me. But the inline { wouldn't change much and still conserve vertical space. I don't get it.

I think they think that you should put the brackets in a separate line from the rest of the code. Something like

function()

{

code

more code

};

versus being

function()

{code

more code};

which is what I use.

>>16432156

>function()

>{code

>more code};

>which is what I use.

Terry's spirit have mercy on your sould

>>16432162

The anime is shit though, why do you expect everyone to have seen it?

>>16432156

Actually the question is between

> function(){

> code

> }

versus

> function()

> {

> code

> }

I don't know what the fuck "{code" is, don't ever type that again anywhere.

>>16432162

I know Shinobu exists, I just haven't watched it and didn't know she was some elf with liver issues. I haven't watched all anime from >10 years ago. If it's so good, how come I've never heard of it outside of /v/? :^)

>>16432171

function() {code;

more code

;

}

;

;

>>16432023

>>16432041

>>16432065

Using a newline curly helps visually when reading the program. When you have conditions nested inside loops nested inside functions, it's hard to tell exactly where you are. The braces give a space to say "here's the start/end of that".

It's all preference, the most important thing is all files use uniform formatting.

>>16432156

Why not put everything in one line? No point in wasting those precious newlines.

>>16432181

Seriously, what possible merit can there be for newline { ?

>>16432162

>>16432171

>>16432176

Anon was right about programmers getting into retarded arguments over trivial shit.

I like my code being as concise in space as humanely possible. This might go back to my first programming foray being in Basic.

Might as well be screeching at me for not using the dark mode in my IDE. This shit is why I'd never even consider making my work open source, it would just result in endless autistic screeching, not because of functionality but because it doesn't look nice to you.

>>16432185

> When you have conditions nested inside loops nested inside functions, it's hard to tell exactly where you are. The braces give a space to say "here's the start/end of that"

That's what indentation is for.

>>16432187

>not using the dark mode in my IDE

Don't your eyes get tired?

>>16432185

>Why not put everything in one line? No point in wasting those precious newlines.

Usually a readability issue. If I can, sure, I'll put it in one line but if there's multiple things going on, I'll break it up to make it more readable.

>>16432192

>Don't your eyes get tired?

Not really, no. My eyes get more strained from dark themes, actually.

>>16432152

>This is new to me.

The column 80 rule has its origins in technical necessity. Way back the monitors on your mainframes couldn't display many characters per line, so programmers did well to layout their code in a manner that ensured that you didn't have to scroll horizontally to see what the code is doing.

>But the inline { wouldn't change much and still conserve vertical space. I don't get it.

The problem of the inline { isn't about the column 80 rule specifically. It's about what happens when you use a vertical layout for your code and you have argument lists. You make the beginning of your function bodies harder to find, because they have the same level of indentation as your argument list.

int Something(
    char *Blah,
    float Whatever,
    double Banana,
    char *SomeText,
    int Index) {
    int Result;
    int Temp;
    char *Buf;
    // ...
}
// vs
int Something(
    char *Blah,
    float Whatever,
    double Banana,
    char *SomeText,
    int Index)
{
    int Result;
    int Temp;
    char *Buf;
    // ...
}

>>16432204

Oh I get it. If my arguments were long enough that I'd multiline them, I'd also put the brace in a new line without thinking it. It's inline as default for me though.

>>16432204

To me that makes the header and the body look like separate things when reading fast. I tend to do something like this if I need to do vertical arguments:

int Something (
		char *Blah,
		float Whatever,
		double Banana,
		char *SomeText,
		int Index
	) {
	int Result;
	int Temp;
	char *Buf;
	// ...
}

>>16432090

>The only sound argument for 80 columns is rotating your screen to increase how many lines of code can be on screen at a time.

LOC isn't the be-all end-all metric for efficiency. Having more stuff on-screen doesn't help, if its structure doesn't lend itself to being parsed quickly. It's not that 80 is some sort of magic number, but it's a good guideline. Vertical code layouts (whether they're 60, 80 or 100 col/line) have some inherent benefits over horizontal layouts. >>16432102 mentioned one: on my 1440p monitor I can view 3 files at once (or the same file in up to 3 places) without anything disappearing behind a scrollbar.

Embed related is the talk I mentioned. Interesting, charismatic guy and he covers other subjects as well so I recommend it. The part about proper naming and what the correct level of abstraction is is absolute gold and should be beaten into every programmer's head, if necessary.

>>16432236

Watch the talk. He mentions that one too. It has one downside mine doesn't: your beautiful alignment of the char *Blah with the opening parenthesis (assuming it wasn't coincidental) isn't stable under refactoring. If the function is renamed to something with a different length, the alignment gets broken. And no, "I can fix it by hand" isn't an argument if you either use an IDE (which can rename the function from within another file, which would force you to review every single refactoring site by hand whenever refactor anything) or if you work with other people, because you can't expect them to be that autistic.

>>16432156

>function()

>{code

>more code};

You meant ()=>{code more code}; right?

>>16432204

>Way back the monitors on your mainframes couldn't display many characters per line

I assumed it was for dot matrix printers

>>16432308

Possible. I only know its origin is technical necessity, so I assumed its monitors. But maybe it's both, because old monitors were narrow and had really low resolutions.

>>16432289

>And no, "I can fix it by hand"

You guys actually do this? Manually identation?

Ever heard of Document.Format or equivalent?

Ctrl+Shift+F in Eclipse and I believe Ctrl+M+D in Visual Studio (long changed to eclipse pattern because I came from years of Java)

>>16432320

Of course not and that was the point.

The problem with auto-formatting systems is that there's a limit to how configurable they are. From everything I've seen, clang-format is the gold standard in terms of flexibility, but it's not available for every language and even it has its limits.

>>16432333

Looks like a perfectly fine alternative to mine to me. You can chalk up using this one over mine to personal preference.

>>16432356

Eclipse's is pretty top notch

Visual studio's is so so, but enough for my preferences

>>16432002

The ones who get picky about those are people who have never completed a single project. Please disregard them.

>>16431154

Godot is very limited and buggy for 3d. It's borderline rage inducing to use since it fucks up the models that I can export easily in Unity/UE4. You're better off working in Unity/UE4 and transferring that workflow into other engines.

>>16433404

Are you drawing everything all at once, or do you create a preset number of animated sprites from what they have equipped?

>>16433429

the second one

>>16434692

Looks like I'm not the only one making a Theif inspired stealth game.

>>16434692

Looks more like they have instant vision and chase a little too fast. Sound should be fine but you have to remember that in order to escape sound detection you have to move slowly, which means you need time to slowly move away from the last spot they heard you before they can reacquire visual.

>>16434789

>>16434692

If you wanted to reduce difficulty while adding realism, you could find some way to make sound-location less precise. They know they heard you from a general direction and move that way, but they don't actually know where you are.

>>16434784

They move kinda… angularly.

>>16434870

not entirely sure what i'm doing wrong with it

in the shader

			float4x4 quaternion_to_matrix(float4 quat)
			{
				float4x4 m = float4x4(float4(0, 0, 0, 0), float4(0, 0, 0, 0), float4(0, 0, 0, 0), float4(0, 0, 0, 0));

				float x = quat.x, y = quat.y, z = quat.z, w = quat.w;
				float x2 = x + x, y2 = y + y, z2 = z + z;
				float xx = x * x2, xy = x * y2, xz = x * z2;
				float yy = y * y2, yz = y * z2, zz = z * z2;
				float wx = w * x2, wy = w * y2, wz = w * z2;

				m[0][0] = 1.0 - (yy + zz);
				m[0][1] = xy - wz;
				m[0][2] = xz + wy;

				m[1][0] = xy + wz;
				m[1][1] = 1.0 - (xx + zz);
				m[1][2] = yz - wx;

				m[2][0] = xz - wy;
				m[2][1] = yz + wx;
				m[2][2] = 1.0 - (xx + yy);

				m[3][3] = 1.0;

				return m;
			}

			float4 toQuaternion(float yaw, float pitch, float roll) // yaw (Z), pitch (Y), roll (X)
			{
				// Abbreviations for the various angular functions
				double cy = cos(yaw * 0.5);
				double sy = sin(yaw * 0.5);
				double cp = cos(pitch * 0.5);
				double sp = sin(pitch * 0.5);
				double cr = cos(roll * 0.5);
				double sr = sin(roll * 0.5);

				float4 q;
				q.w = cy * cp * cr + sy * sp * sr;
				q.x = cy * cp * sr - sy * sp * cr;
				q.y = sy * cp * sr + cy * sp * cr;
				q.z = sy * cp * cr - cy * sp * sr;
				return q;
			}

			void setup()
			{
			#ifdef UNITY_PROCEDURAL_INSTANCING_ENABLED
				float4x4 rotation = quaternion_to_matrix(toQuaternion(0, positionBuffer[unity_InstanceID].w, 0));
				

in c#

    public static float vectorAngle(float x, float y)
    {
        if (x == 0) // special cases
            return (y > 0) ? 90
                : (y == 0) ? 0
                : 270;
        else if (y == 0) // special cases
            return (x >= 0) ? 0
                : 180;
        float ret = math.degrees(math.atan((float)y / x));
        if (x < 0 && y < 0) // quadrant Ⅲ
            ret = 180 + ret;
        else if (x < 0) // quadrant Ⅱ
            ret = 180 + ret; // it actually substracts
        else if (y < 0) // quadrant Ⅳ
            ret = 270 + (90 + ret); // it actually substracts
        return ret;
    }

and somewhere later

angle = vectorAngle(dir.x, dir.y);

>>16431141

Shinobu a cute. A CUTE!

Demo Day's tomorrow, but I'll be busy for most of the day. I'll make some time to post my project, but probably won't be around much after that.

Do we have a gameplan on who's baking the thread? I can't do it this time. There's a guide on http://8agdg.wikidot.com/demo-days so as long as you pay attention, anyone should be able to do it.

I should also probably update that page to add the newer autumn-edition image I made last time.

>>16435163

Pretty sure

>OP made that brainfuck thingie, which is more of a toy for programmers but whatever

>there'll be a new update for Beelzebox

>cookinganon and bro will have a new update for Everyday Life

>the anon who made something for Ludum Dare last weekend will probably post it there

Maybe/likely

>shinobudev might have an update

>that one anon that is working on a game that looks like Ultima IV posts occasionally, he might have something

>this anon >>16433404 may have a demo

>the other gamejam people >>16422623 though I kinda feel like they're new to these threads so who knows

>>16435231

Oh fuck yes, I've been looking forward to testing this baby.

Your development speed is really goddamn fast as well, didn't you start development on this in early 2018?

>>16435277

Unity or Godot?

>>16435291

>he doesn't know

>>16435319

YUME and YUME2 are still proprietary, though.

>>16435291

Custom engine. https://kircode.com/yume

>>16435231

Oh fuck that looks awesome.

>>16435252

>>the anon who made something for Ludum Dare last weekend will probably post it there

Yeah I'll be posting it since I've continued working on it since the gamejam and think I will be doing so for a while. Hope people don't mind even though I haven't really posted here since I stopped working on my educational game.

>>16435482

>Hope people don't mind even though I haven't really posted here since I stopped working on my educational game

Ah, you're the guy who was making that math RPG? I see no reason why anyone would mind, you've posted plenty of times in the past and aren't blazing in here with HERE'S MY DEMO DONATE TO MY PATREON PLS only to stop posting right afterwards

>>16435482

I post here all the time but pretty much never post progress updates on my own project because I just don't care to. And shitposting gives way more (You)'s anyway.

>>16435435

Impressive.

>>16436344

That is going to be heavily dependent on how you implemented tiles/objects. I see you have stairs, how do those work? Presumably they have an extra variable that points at an X/Y coordinate to teleport to. You can have similar extra data on signs that's just a string of characters. Or have a long list of messages somewhere, and just point at a message ID in the sign.

When will demo day start?

>>16435109 http://8agdg.wikidot.com/demo-days mentions "12am at timezone discussed in cyclic agdg thread" and I'd like to know which one that is.

>>16436344

There's a lot of ways to do it. For example make a separate array of all the sign contents, and when you read a sign, the appropriate text is searched from the array based on coordinates or something. That way you wouldn't need to store anything in the signs themselves.

>>16436417

midnight tonight, I'll have it up for a few days to a week if needed

>>16436426

>midnight tonight

my midnight or yours midnight?

>>16434038

Are you using hierarchical system to categorize food, or is it set-based?

http://wiki.c2.com/?LimitsOfHierarchies

>>16437224

Food got a ton of tags, for instance, carrots have

Carrot, Vegetable, Root, Tuber

Recipes have slots that requires/accepts/forbids tags

I think that counts as set based?

hierarchical system would be like how species are categorized?

Is there any particular engine that would be best for making a 3D fighting game? I know Unreal 4 seems to be getting used a lot these days (Tekken, Soulcalibur, etc) but maybe there is something better? Unreal seems to have a bit of extra input latency inherent to it and I'd like to avoid that.

>>16437241

Tags/sets are the alternative to hierarchies where one trait is higher than another despite being orthogonal

>>16437253

UE 4.19 reduced the inherent lag to ~2 frames. Which is obviously not perfect, but is better than it used to be.

>>16437253

make your own engine

Question about Blender 2.8. How do I bake a procedural texture to an image? I've tried the online tutorials but that just gives me a black texture.

The process I've followed that doesn't work.

>changed to the Cycles renderer

>created a new image in the UV editor

>added a texture node to my material node editor

>set the image for that texture node to the image I created

>selected the texture node

>in the render tab, clicked "bake"

>after the texture bakes I get a black texture that covers the area covered by the UV of the object.

>>16436632

What a satisfying take on pacman gameplay.

>>16435231

looking forward to it

>>16437308

That's good to hear, so I guess the extra lag is on the developer's side then.

>>16437429

Not viable.

From what I can tell, modern fighting games still tie their gameplay to frame rate. I assume this is to ensure absolutely perfect, consistent, reliable hit detection on moves and combos, but I'm wondering if such a practice might not actually be that necessary these days. Does anyone think a variable frame rate fighting game could be done well, or is it best not to risk it?

>>16436632

Your lawnmower sounds idle, gotta get a sound effect of that sucker at full blast, and perhaps dynamically switch to the idle sound effect if you stop moving for too long.

>>16437531

Nevermind, figured it out. I needed some light to be able to bake the texture, so I ran it through an emissive.

>>16437990

>From what I can tell, modern fighting games still tie their gameplay to frame rate. I assume this is to ensure absolutely perfect, consistent, reliable hit detection on moves and combos, but I'm wondering if such a practice might not actually be that necessary these days. Does anyone think a variable frame rate fighting game could be done well, or is it best not to risk it?

Online multiplayer will require your game logic to be deterministic. So you will always be stuck to a constant gameplay tickrate.

That said, there's no reason why rendering framerate needs to be tied to gameplay tickrate. You can simply run the game logic itself at a high rate (e.g. 240Hz) and then when rendering your render the last gameplay tick every time. Or interpolate between ticks, if you want that bit extra.

However, integrating this into existing engines might be a bit of a pain.

I'm gonna go ahead and post the demo day threads on /agdg/ and /v/ in a few minutes.

It's up.

>>16438187

>>/agdg/33714

>>16438190

>>>/agdg/33714

fug

>>16437759

It's a remake of an iPod Touch game i used to play 8 years ago

>>16437992

Yeah, that is up next. I've spent a lot of time on Freeaudio but didn't really find something that fits well.

What's the correct way to do interpolation and fading and stuff? Like let's say I want the camera to smoothly change position from point A to point B, where should the movement code be?

My current plan is/was to have some array of animation information, so if I want to animate some value I'd just add it there and only have to check that array in the game loop, but there's too many things that are not fixed in memory and thus can't be given a generic pointer to.

Or what if I want to have a fade out animation for sprites, should I add a special fade out mechanic to every single shader as opposed to just changing the alpha value?

>>16441060

You mean like sine wave?

so apparently there was already an actual 3d bionicle fangame that got abandoned and shared all of it's assets for free

http://www.bzpower.com/board/topic/13995-project-mnog-3d/

as long as i don't care about being sued by lego i can totally replace my assets with these

>>16441060

For something like fading a sprite I'd just lerp the alpha value over time but for more complex stuff you definitely should use splines like >>16441315 said. If you're using an engine like Unity just download a tweening lib and be done with it. I see no point in rolling your own tweens unless you have very specific requirements.

>>16441223

>>16441315

>>16441317

I'm not talking about the transition algorithm, I mean where do I actually put it without hardcoding fade-related data into every data structure and an if/else condition and a fading loop function for every separate transition I want to do? Where does the transition function get called from and where does it get it's state?

>>16441324

Really depends on the rest of your engine. An ECS system might have a fade component where a system handles all types of fading. Or you push an event that contains what entity and how to fade it. Or you have some kind of fading lib that you call from that specific entity/at your fade fixing point that will push a fade job to some queue. Or just run the fade code in your entity.

>>16441324

>>16441331

Another option is having a single "fader" object that handles the calculations, receiving the necessary information in the parameters of the method, something like:

class fader {

public 3Dpos sineFade (3Dpos startPos, 3Dpos endPos, float transitionRatio) {/*…*/}

}

//then

static fader myFade=new fader (); //one global fader, maybe?

onTickOrWhatever () {

myObj.setPos(myFade.sineFade(myObj.getPos(), myObj.getTarget(), animationRatio))

}

meh.

>>16442041

The fade from top to bottom? I like it. What's it look like at the secret passages though?

>>16441303

>being sued

unless you're stupid and take money for it and if the assets are all recreations (and not ripped straight from one of their games) worst will probably be a c&d, if they even bother with that depending where you are.

companies usually only get extra salty when you use their shit without paying or worse make money of their shit.

>>16442041

doesn't look bad, just a bit "off" for me, too "2d" compared to the rest if it makes sense? bit of structure or outline might do the trick.