Could I win a game jam with an engine I’ve never used?
Vložit
- čas přidán 3. 06. 2024
- Heya Pals!
Here's my post mortem and impressions of making a game using godot for Ludum Dare 54! I'm super happy with the result, and I was so lucky to have so many of you tune in and play the game. Thanks for your support, and I hope you enjoy!
You can of course play the game here: uppon-hill.itch.io/though-i-walk
Shoutouts to Kevin Thompson for writing this script for fast-tracking godot's geometry from a sprite feature: gist.github.com/kevinthompson...
Chapters:
0:00 - Intro
3:23 - Learning Godot & Impressions
10:11 - Jam Day 1
16:41 - Jam Day 2
29:24 - Jam Final Stretch
37:08 - Thoughts on the Jam
40:30 - Future Games in Godot
42:15 - Outro
Music: Gamechops.com
[Souls & Chill]
[Mario & Chill]
----
This video features clips from my stream. Catch it live: Weekdays 1-6pm AEST.
Twitch: / adamcyounis
Twitter: / adamcyounis
Tiktok: / adamcyounis
Discord: / discord
Become a Patron at / adamcyounis
Download assets and games from the stream at uppon-hill.itch.io/
Later, pals! - Věda a technologie
Making a video on how you would improve the animator would definitely help. The community is 100% actively looking for feedback and somebody might actually pick some ideas up and implement them.
At least this is an engine where the motivation is to improve the engine. FOSS is the best.
100% THIS is the biggest strength of open source. If you can provide a efficiently structured case there's more than a dozen people willing to try and implement it
big agree i would love to see your take on this!
There is an older github issue (#214) about the problem he mentiones where someone suggests sandboxing and it has been closed saying that they decided that a system with reset- tracks would be the better solution. Then they implemented it and it's the garbage we have today. So it doesn't look like there is hope for a better animationplayer.
no they wont, the devs are stretched thin, barely any fulltime maintainers outside of core
For the animation UI I believe you can just make a proposal in the Godot Issue tracker. If you make a video, and post it there as an issue/ feature proposal people can discuss it and implement the changes.
Really interesting to see Godot through the eyes of someone who hasn't used it before, impressed with what you managed to get done! Your feedback on Godot's UI and UX would be very valuable 😄
Hey! Make more videos!
Insert bojack horseman "what are YOU doing here"
For the people wondering why Adam wasn't able to place tiles in the learning godot section; when creating a tilemap in the scene tree, you get two menus at the bottom: Tilemap and Tileset. Unfortunately, those menus are covered by his twitch chat UI so you can't see exactly where they show up.
He was in the Tileset editor where you can import the tilemap.png and adjust some features and physics of each tile.
The Tilemap editor will allow you to actually place the tiles. I personally found the tilemap editing to be a lot more intuitive than unity, but I also did learn some aspects of godot from courses before just jumping in.
Yes! Please make a feedback video on Godot. This kind of feedback is really valuable. After the Unity dumpster fire Godot began improving at a more rapid rate. For example, 4.2. beta has a ton of improvements, my favourite being the baked 2D navimeshes.
I love your critics for Godot. I will encourage you to open proposals to make it better :)
Just when I think most gameplay styles have already been done before... That parallax function is beautifully creative, especially when you think of bending the world around you through control over yourself on the soul level, so the function fits the character too. Godot will sometimes crash on me too, but it could just be my laptop, even though it's new.
I just played the whole game and I think the reason it didn't get any better was because of a few obvious issues.
1. No music, so the game felt a little dead.
2. Since it had no music, the sound effects really needed to take the lead, but they were pretty uninteresting.
3. The controls felt a little bit clumsy. Specifically moving the soul to move the paralax often felt like a debug sandbox, lacking rigidity that prevents silly mistakes or accidentally crushing the character to death. I found myself repeatedly trying to get "comfortable" with the controls so that I could move things about quickly, but this was almost always met with getting killed or resetting my position because the only boundaries that exist are basically death and extending outside the game's limits.
4. No ending. Technically there is, but this very much felt like one of those slap-in-the-face Nintendo Entertainment System endings. "Congratulations for putting in all that work, now go screw yourself."
5. I didn't know how to make the soul come back to me and just had to guess. I didn't realize that holding shift was necessary to keep moving the soul. Although I figured it out pretty quick, it just added a sense that the game was incomplete, rushed or otherwise not quite polished.
6. I felt like jumping, while not _bad,_ needed a bit more control. It didn't feel great.
The positive in all this? The things that would need to be improved are not enigmatic or difficult to implement. Essentially, you ran out of time. Can't really help that.
To be fair, idk if most game jam games have satisfying endings, I think most are sort of prototypes that are hopefully a bit polished. Like many game jam games might have 3 levels in a game that may have 10+ if it were a finished game
Awesome to see more and more people using Godot!
Even in the past few years, the engine has received a ton of updates; especially in the animation and tileset areas you were mentioning.
With more active users and contributors, it won't be long before these systems are more fleshed out and up to standard with other tools! :)
Loved to see a whole game dev process in 45min. Gave me a nice overview on what has to be done. And love your organization tools
just started using your palette and it's AMAZING -- perfect desaturation as brightness decreases, friggin love it
This is such a great video! I was eating dinner while watching and had a blast listening to your approach on the game, music, art, and of course the engine. I will echo what others have said and say I'd love a series of deep dives. :)
Was fun watching the process of this on stream! Good work!
Excellent video. Perfect pacing and feel with music and editing, clear explanations and interesting narrative with the Godot vs. Unity aspect.
And the game looks really cool. No wonder it scored so high in atmosphere.
Been waiting for you to turn the streams into a video! Really loved the design of the game both mechancially and artistically.
Awesome game! Great to see you jumping into Godot as well!
What a great video! I think the feedback to the Godot team would be awesome, I agree that the tilemap and animator UX need work and I would love to see it improved
… i love your content 😭! I’m an artist/sort of game dev, i haven’t published anything but I’ve been tinkering on things for years now. Your stuff is just sooo inspiring, thanks for posting honestly.
Your main game mechanic was way cooler than mine lol. Good job on the placements! I wasn't too far behind in overall, and even got #36 in fun so that was a cool outcome, hopefully we see more videos of you doing game jams, they are fun to watch alongside Insignia :)
I love Game Jams. Such a great opportunity to improve your skills and be rewarded with a epic game!
Subbed! You are incredibly talented and I hope to see more Godot development content 👍
I found a plugin called "better terrain" and it is much better than the default title editor. Plus, the guy who made it has done excellent video tutorials showing it off.
This is a sick mechanic and the art is beautiful. It would be cool if you made a full game from it
A lantern would be a good idea for when the soul is away from the body, having the white soul jump between the lantern and the player based on which is the focus of the controls. That could be their "sword" :o
Really cool and interesting video Adam 👌
Congrats!! Great idea and execution
GOOD LUCK WITH NEW ENGINE we're rooting for you
I love how you're facing problems in my Engine of choice and the fact that I've been through it all before in your Engine of choice Unity makes it even better xD
The idea is insane wonderful!
that is really kick ass concept pal, niiiice!
1:50 wow.. just wow... Also thanks for getting straight to the point in your video no fluff intro crud
My favorite artist playing with my favorite engine, what a delightful video ^^
Haha, I started getting into Godot recently. Nice to see I wasn't the only one who was very confused by the animation saving thing. :D
Definitely helps accelerate the creative process with music to have some templated arrangement of instruments to jump in to; my creative juice tends to fade as I fiddle with the UI and prep for making music
Interestingly the 2 things you called out as not great are being worked on for 4.2. And 4.1 put some improvements in.
Great, ain it
Just for science, seeing you test UE5 with PaperZD would be extremely interesting.
I've used GD, Unity and UE and I think the 2D/indie part of UE is something people are too harsh about. It feels like a thing people just say at this point without having ever tried it.
UE for "2D", not actually true 2D but 2.5D instead even if it fully looks like 2D brings a lot of benefits, like lighting, physics and other things.
And yes, when you sleep your brain takes the input of the day and processes it. This is why sleep is so important.
The parallax layer mechanic was such a good idea @adamcyounis . Enjoyed this vid. ✌
Don't know what your process is for making music, but would recommend preparing a sample library of sounds/loops before hand, and just doing some sound design with some of the basic synths and saving those presets. This is a good way of practicing the music part without feeling that you have to make a song, you can just explore and have fun, and then you have your own library that you know that you can reach for different things if you are feeling uninspired! :) Cool game!
Great video! Thanks!!!
Having a full CICD pipeline set up to automate builds to itch is super smart for a jam. Definitly soemthign more should prepare in advance
Loved watching it live! I kinda hoped you would call the game Genuflect since it sounds like a cool word, but yeah meaning isn't really fitting with whole game.
Tbh I would love a video of you talking about how you would improve animator and tile system for godot. I might be biased tho since I'm a designer so I nerd out on these things lol
gr8 video! godot content is fun to watch
Pleaaaase do a video on how you would handle tile editing in godot. You have been a huge inspiration in terms of workflow overall and im building a level editor in game maker atm trying to replicate the auto tile system from Tiles.
impressive result!
I struggled with the Reset animations when I first switched to Godot too. Now I always add my first keyframe for each property using the key icon in the properties panel. If you do it this way you will get a popup asking if you want it to create a Reset keyframe automatically.
bro just gave me motivation to learn godot
You explaining your learning godot was more important than all of the tutorials combined
Very cool game concept
I know they're working on fixing the animation player, but for now there's a plugin that can bridge the standard sprite animator to the animation player. This means you can set up your animations in the sprite animator and then export them to the animation player without having to manually set up the timeline. Its a decent temporary solution, since animatedsprite2d doesnt work with the animation tree.
Something really cool about Godot is that since it is open source the problems that you're having might have already been solved by another user, or you can even create the plugins that you need yourself. Obviously it takes being more comfortable with the engine itself before you start editing things, but the fact it is open source makes it so that the portions that still need polished can and will be polished overtime.
Ah Shucks!
It's a bit rough that you and the animation player didn't get on an equal footing.
The bezier curve thing is baffling to me too but overall it's quite intuitive to use within the engines context once you get the hang on it. Linking up functions or variables even to animations is super quick and simple, changing multiple properties in one go is quite fast too.
Though Godot itself has this distinction. If you wanna animate something that you know beforehand where it starts and ends, the animation player is good to go, if you don't know, however, tweening in code is much better.
I would think to simply animate your camera target, a tween would've done it within seconds of you figuring out the syntax.
Nice work!
Yeah this was incredible. That was awesome
You really inspire me :D
Ahh, that's annoying! You ran into the packed scene circular reference bug! It's pretty awful, just found that on a Discord server and was able to help someone identify it, but it's not fun at all to see that corruption. I imagine it was even worse for you, being so close to the end.
Also, a large bunch of the community would *LOVE* for you to find ways to improve the TileMap/Set editor and the AnimationPlayer. The former is my biggest gripe right now. I don't do bezier curves very frequently, but I make TileMaps all the time.
Outstanding
Very creative gameplay
22:58 - if it's only one color, you could make a separated sprite in all white and then modulate it to any color you eant. I use this all the time
There is in fact an in and out function for the animations to help smooth them out
Do you have a video talking about how you make your schedules, what is your thought process when you want to make one... how do you make a plan? I am so curious about this subject... I don't know if this is the task of a Project manager in a team but I think the way you organize your work is very impressive!
I'm so fucking glad you did a video with some Godot to it Adam! Love your ability to explain and guide, you're a legend. I've been learning it myself and it certainly has its issues.
Make this an entire game! preatty please
i admire you so much.
amazing!
Ooooh this video gonna do good
I've recently switched over to Godot and feel like I must've got extremely lucky not to run into those issues. That animator timeline one in particular sounds *awful*. I feel like the issues with the simple screen transition animation probably stem from the engine being made by very tech-y, code focused people, who would find it much more natural to code anything like that via script, but that's no excuse when the tool is there to be used.
Congrats on the great jam result, and I can't wait to play it to see what your soundtrack sounds like.
I would like to see how you would build the title maker or animation system
I use godot quite often and the animation player thing (with the reset) that adam mentions in the video gives me terrible anxiety every single time
"I platform by faith, not by sight "-Paul probably
It's open source, dude! Feel free to participate! Especially on the animating stuff, where you are an expert on. Godot Go do it!
Could you do a video showing how to set up a Git like that for your Godot project?
A visible aura around the character would be awesome to further respect the state change.
The better questions is: "Does AdamCYounis make a game that pushes the medium forwards in a Game Jam with Godot?" The answer to both of those is a resounding yes. I wonder how you would make a Full Game out of this prototype. Would you make it into a horror game, where you have to capture the Souls of others who lose themselves to the Parallax? Would it be an atmospheric game about piety?
The possibilities are endless and I can't wait to see what is in store! If I may ask, how do you come up with game ideas that are so creative. I'm coding a game for a Halloween Game Jam and it's basically just a side-scroller. It feels a bit lame for what I'm going for, even though its my first Game Jam. How can I make something that plain be as grand or interesting as your games?
Really great work Adam, you did great and congratulations on the result!
On a side note: The cool thing about Godot being an Open Source engine is that if you are not happy with a system/feature, you can fork it, change it and use it! Or even merging it back to be available for everyone. Of course game development and engine development are different skills, that is the beauty of Open Source, just something to keep in mind :)
Can you make a video about what it is about the tile map issues ? I actually like it better than Unity once I got used to it,
I started using Godot earlier this year to build a 2d platformer. The tilemap was a stumper trying to get my ramps to work and getting the layering correct. I have zero experience in anything but used Heartbeast's tutorial to get a functional first level built except with my own assets. I never finished the tutorial series unfortunately. I got distracted trying to figure out how to make controllable riding platforms. I was going to put nubs on mine so it was a bucket as well. But ran into the issue that the riding platform wouldn't collide with the walls of my tile map. I wanted rideable platforms you could control with ui keys that were contained to the area in which they were placed. Or a hover vehicle only controllable while in an area2d attached to the vehicle. Or ui controlled elevators. I couldn't figure any of that out and searched around but couldn't find anyone having the same issues. I plan to get back to it as a winter project. If you started doing godot how to videos and could solve niche issues like those I'd be interested.
I'm not sure how helpful this is, but I believe if you made platforms with a characterbody or kinematicbody, you can use is_on_wall to detect left and right boundaries in your tilemap. I believe that as long as the tilemap has a collision physics layer, that should do the trick.
@@Chspas Thanks for the quick reply I had already made the change to character body but still had the original node type referenced in the script not sure if that was hurting my situation. I ended up getting it to work later the next day basically doing what you said... I pasted in the wall jump function from my player script and deleted all the lines I didn't need until all the errors went away and it worked.
Now I'm onto trying to debug this rideable "enemy" that turns around at walls... I angled up the raycast so it wouldn't interact going up steep ramps and got it so it wouldn't fly straight anymore instead of going down the steep ramps. But now that it's all functioning alright I noticed if the player sits just right on the edge of the collision shape where the raycast points the flip of the enemy throws the player into the air with a wild velocity. It's quite jarring anyway enough of my troubles thanks again.
For non isometric 2D games, mappging / level creating got into a place where I do not see myself using any other tool than LDtk...Quality of life....
Is that some smooth plin plin plon in the background?
Would be interested in hearing what your specific issues were with the animator and the tile map editor. I honestly have found both to be leaps and bounds better/easier than any other engine I've used. I'm not sure I understood the issue with selecting a Bezier curve to start, is that more because you'd rather it be default so that you don't have to remember to click it every time?
Personally I just haven't used Bezier tracks a ton because you can already adjust easing with a regular property track. Wouldn't curves on top of that just be easing on easing? Or is it more the workflow of doing it by feel that makes Beziers more appealing? It's funny to me that you liked Unity's better when I found that to be utterly painful.
I don't get the automapping concern at all. It has automapping that works and is fairly easy to create. You can also split it up by terrain type and assign variation/randomization/collision all pretty easily. Not sure what happened there for you, but the functionality is definitely there.
In regards to the animator, there's just no reason to have different kinds of tracks. it seems to be a result of the bezier tracks being a more recent addition, but I don't think it's suitable to ask the user to choose one or the other when conceptually they represent the same kind of data. I found the inability to natively select, copy and paste keyframes along and across tracks with keyboard shortcuts very displeasing as well.
Regarding auto-mapping with tilemaps, there are two classes of solution here. Those that work by checking the neighbours to the subject tile, and those that work with find / replace operations over regions. The latter is far more powerful at allowing the user to define complex mapping rules, as seen in software like Tiled and LDTK. Particularly, any logic that goes further than one neighbour deep is impossible with Godot's solution, but is kind of foundational to nuanced tilemaps in higher fidelity games.
@@AdamCYounisThanks for responding and taking the time to explain. I'm unfamiliar with the specific aspects of Tiled that you're referring to but it sounds like even then it would put Godot at a similar level with most game engine based tile editing. If I'm comparing apples to apples I'd much rather use Godot's tile editing over GMS or Unity ever again.
The animator issue makes sense now that you've explained it but also could be a result of attempting to keep the engine beginner friendly and broken up into specific components with specific use cases. That's more of a guess than anything but I really haven't found the animator to be a pain point at all. Maybe I just got lucky but I also find myself really only using property or method tracks 99% of the time.
Thanks again for the explanation and hopefully they work out those details.
Also I agree that I've had some issues with the copy/pasting/moving of frames and the changing of values. So in that regard I do think that part can be less fiddly than it currently is.
May I know the name of the project management schedule you are using? Love the UI!
With Godot and your art skillz, you could win it for sure!
Whst was the tool you were using to show what tasks you did when?
Godot 4.2 got released today, one of the things it got was hot reloading.
I think being ranked "12th" is kind of the point where you can make games that most people will enjoy, and as long as you're consistent, you'll develop a group of fans that will shell out the five to twenty dollars for all of your projects. Unlike, say, sports, it's not a zero-sum game, where if one guy takes a spot then another person is removed from that spot.
In this market, as long as you're in the top %20, you will always find work, always find an audience.
That's where you are. You won't always, maybe even will never, hit number 1, but you'll consistently be in the top 20% or better, and that's good for both meaningful output and income.
You should be very proud of where you are, and I'm sure you're aware of your skill level.
I aspire to be at this level in my novel writing. Of course, like you, I also aspire to be better than simply good enough. You go above and beyond with your pixel art. Every aspect of your style, when it comes to projects that you have time to put proper effort into, pushes quality to your limits, and therefore is actually superior to something like 95% of things out there. You're just not satisfied with quick work and good-enough art and animations. You want to make something of genuinely exceptional quality.
This is how I am with my novels. I don't want to merely write a "good story." I want to be on top of the craft. I want my stories to be fun to read at the level of the very _sentences,_ where word choice and cadence and narrative voice enrapture the audience immediately and sweep them into a tightly written story with deep and likable characters. I want every paragraph to introduce an idea and then pay it off. Constantly, the audience is being engaged, challenged, reminded of previous ideas. _Bam, bam, bam!_
It's tough to do when you have an outline that's nearly 100,000 words all by itself.
Anyway, the point is that we're not satisfied with adequate, and I hope that I'm able to one day be, in writing, at the level you are with sprite-based games.
I think the problem is that Godot is a Frankenstein monster of developers who contribute different features. Some of those features are extremely good, and some are not, but most have poor usability because devs aren't usually trained to do UX. GIMP, Blender, Krita, etc., all have poor usability. This is just what we should expect from open-source.
43 mins. We are here for a loong time hehe
PLEASE do a video on how you'd do an animation system. I love godot, but I also can't stand the editor. Luckily it's early, so any ideas you have can probably be implemented!! :)
What was you setup for working with a pixel art game? Was it pixel perfect (with the viewport stretch mode) or did you use the canvas_items mode?
In this case, pixel perfect with viewport stretch.
I’m curious, what’s the tool you’re making the trello/Kanban board or whatever it’s called
I love Godot to death but there are a lot of really weird bugs. Like if you yield create timer and the object leaves the scene, it always crashes--this is probably the most common error. The best way to circumvent this is to simply create a timer node and yielding until that timer finishes
Oh, it was weird watching adam do the preparations for the jam, I always thought that it was considered cheating and I've always started from scratch in my jams
Consider that "from scratch" is vastly different depending on what engine you're using. I had an FSM, a parallax script and some ideas, that's not really much to begin with, especially since both have equivalent Godot systems that come out of the box.
In any case, the rules state the following: "You’re free to use any tools or libraries to create your game. You’re free to start with any base-code you may have. At the end, you will be required to share your source code."
Hi, did you use Godot version 4? Because the tile system does better now and its progress is very good since beginnig.
Let's be honest, Godot 3 tilemaps were convoluted and limited, but Godot 4 tilemaps are even more convoluted, and on top of that are non deterministic. To the point when you just *have* to use third-party addons to make it usable.
And if you are using third-party tools anyway, why not just use Tiled export functionality as it supports Godot 4 now? It's far easier to build your levels in Tiled and just export to Godot. And since this is exactly what Adam's doing in Unity for his game, it would've been far easier to do in Godot.
Oh you're a ux designer... Now the video makes sense. We think similarly this was cool.
have you heard of ldtk for auto tiling?
26:41 I think the RESET frame would have saved you a lot of time in terms of the animator editing your serialized fields. This is something that I also struggled with and found really unintuitive about the animator
it's the subtleties that people don't notice what makes something feel great. Because if it's not there, people don't know why but they just don't like it lol.
time to make popcorn
What do you use for tile maps? If unity and godot arent good enough?
Automating tile mapping is a huge part of my workflow, and Unity / Godots solutions are only suitable for automating the edges. I personally use Tiled map editor, which allows for far more control and customisation of the mapping rules. I've also heard good things about LDtk. There are certainly plugins to allow one to import Tiled maps into Godot, but it'd be nice to cut out the third party if Godot's solution were good enough out of the box.
So what about that figma template, hmm? I actually have a team jam in a week so that would be handy
3:22 Why 4.0? Newer versions improved on the tile maps and animation system, two of the critique points you had?
I probably should have said 4.x. I used the most recent version available in late September, whichever version that was.