I just want to say that i hope you do well on this platform. You are exactly what people need right now. I'm done with quick and dirty tutorials that convince you that you can make a full game made of bandaids. It makes it so difficult to actually make a game. It makes you feel like it's impossible to make the game you want. You feel it's only possible to make the simplest game you can think of, nothing actually good. No one shows actual architecture. So thank you. Having said all that, i feel like for someone that doesn't know that much about programming (even tho i had programming for two years in school) it's hard to understand what you're saying. When you start talking about simulation and layers and backend, stuff like that, it's really easy to get a bit lost. I only get the most basic idea of what you mean (honestly the hammer thing really helped lol) As you said, the people that know about these things aren't watching this video. It's the ones that have no idea that are here. So, can you recommend SIMPLE and SHORT books or videos to get the idea? I'm personally very lazy and not very smart so it would really help. Or better yet, make a video. The advantage of you making it is that you can control what you talk about, therefore giving a much better understanding of what you're saying. Also, personally, i'm very visual, and so are many others. I think it's a good idea to show how things in different scripts are related. Even if you just said something 2 seconds ago, it's hard for people to keep it in mind at the same time you're explaining something else. And then understand that well enough to remember it for the next thing you're talking about. Check out godotneers for a good way of showing things visually (i've only watched his data model video) Having said all that, this video did have a lot of visual aides Man i went for a long while lol, sorry about that Thank you again
First, as extremely talkative ("writative", lmao) person, I'll never feel bad about reading a long letter-comment. The more the better and don't be ashamed to do that, I find great pleasure in talking in general. Thank you for your support. Secondly, you are unbelievably on point :D. So, right now I'm releasing the first video on enemies subject, and I made it waaay slower as an experiment with format. Also somewhere in Sunday I'll do a voting on how people liked the slower pace. The whole enemies series will be slower no matter the voting outcome, for experiment purity. But make sure to put your vote, I'll happily bounce back to galloping tempo if people will say they liked that more ^__^. Considering simple and short books, sadly, nothing comes to mind right now. But, if nothing sudden will happen, I have a plan to make a series about gamedev and programming for people with zero programming experience. and I want to make it reeealy far-coming, like, from "what is reality, what is an abstraction and how do we describe one with the other?" word-river. Somewhere mid-autumn probably.
About the tempo: It does seem like a hard thing to balance. How do you make a well explained, information dense video, that doesn't alienate the people that are there to see it in the first place? Slowing it down and being more detailed is probably a good thing to try. But make it too slow, and people get the feeling like the video is very empty and they click off. You'll only keep the people that are actually following the tutorial and people watching for entertainment (true weirdoes lol) I guess both slow and fast are valid for different reasons The series you mentioned sounds amazing. It sounds like exactly what i need. I feel like it would be a good base for whatever you do in the future.
This is damn insightful, I love the way you differentiate between tools and frameworks. I've been wondering, as a new Godot user and a long time user of Vue, what a framework might look like to encapsulate the complexities I've run into. I'm excited to see what other content you have, thank you for sharing!!
Tutorials only isolating and solving a single issue is really an understatement. I feel like 90% of devs run from tutorial to tutorial and never really dig their heels into fully understand concepts in their entirety forever chasing a dream idea.
I wish I could understand English better, sounds like you really know what you're talking about. Also, I might have missed/missinterpreted some parts of the video but I wonder will you make a video on animation blending? Not about blending trees but on how to implement a custom animation blending system (something to smoothly blend between idle/walk/run/dash animations depending on speed, maybe an implementation of randomized/partually procedural animations (altho I'd imagine if the system is flexible enough it would'nt be hard to make an extension for it to do all of that))
Maybe I will, but not in the nearest future, and it won't be some "superior custom animation system from the ground up". Godot's animation stuff is actually pretty good and capable for animation purposes, the whole video was as essay on the attempts to shove your game logic inside your animations. Considering your question about smooth walk/run/sprint, the answer you seek is not animation blending, but animation speed manipulation. Let's say you have your run animation default as 4 m/s speed and your sprint animation as 7 m/s. If you want to have a representation for about 5 m/s movement, you can't just "blend" run and sprint with some coefficient. Because to blend animations is essentially to sum a bunch of 3d-vectors with said coefficient and animation of "70% run, 30% sprint" will be a gibberish. What industry does instead is it speeds up the character and animation at exactly the same rate to get the feet touching floor synchronized with character speed. So you have your run animated corresponding to 4 m/s, then to simulate 5 m/s movement you just need to play your run animation 25% faster. Then there will be some level at about 5.6 m/s, where 40% accelerated run animation gives up and you start to play 30% slowed sprint instead. If you are aiming to replicate said effect with Godot, check the AnimationPlayer.play method, it actually has 3 optional arguments, one of those is playback speed exactly for such needs. Or you can "scale up" all the animations of the player with its field AnimationPlayer.speed_scale. See more information here: docs.godotengine.org/en/stable/classes/class_animationplayer.html#class-animationplayer-property-speed-scale.
I see, I remember trying to make something in unity 3-4 years ago and a lot of "beginner" tutorials usually pointed straight to the animation controller, which is a pretty powerfull tool but working around it is a huge mistake which back then ended up with me drawing pentagrams in state graph and trying to use engine only libraries to access current animation state which was a disaster. MVC might require a bit more work to get started but makes things a lot smoother in the long term and even if something goes wrong it'll be a lot easier to fix because you know how your code works. Hope more people see your tutorials!
Yup. I also not so long ago faced a necessity to split my character into torso and legs to be able to add those sexy "torso only" actions like reloads etc. After bashing my head into blend tree for several hours and seeing how awful it is, I went into "fine, I'll do it myself" mode and threw together a system of two simple animation players masked by a simple Node on top directing and syncing them :D.
@@PointDown I've been looking through the docs whenever I get time. Haven't really dug into it yet but seems way more lenient and manageable. Do you have a video on blending the torso and legs manually without anim_tree? I'll take a look when I got free time.
@@pewpeee Yes, I do!) In fact, I have two of them: Main one: th-cam.com/video/Fsa2wxyQvzM/w-d-xo.html And a sequel: th-cam.com/video/huwW1JZt0BI/w-d-xo.html I am also finishing the edit of the last update with a lot of bugfixes for several controller's systems including that double-animator's design. It will be out tomorrow, so, what I'd recommend is to watch those videos for design decisions narrations, but if you'll want to look at the code - to check the branch from that last controller's video as it have some minor problems fixed.
@@PointDown kek. this just happened to pop into my feed while I was making a fighting game and trying to decide how to implement a state machine, so needless to say I'm happy.
I just want to say that i hope you do well on this platform. You are exactly what people need right now. I'm done with quick and dirty tutorials that convince you that you can make a full game made of bandaids. It makes it so difficult to actually make a game. It makes you feel like it's impossible to make the game you want. You feel it's only possible to make the simplest game you can think of, nothing actually good.
No one shows actual architecture. So thank you.
Having said all that, i feel like for someone that doesn't know that much about programming (even tho i had programming for two years in school) it's hard to understand what you're saying.
When you start talking about simulation and layers and backend, stuff like that, it's really easy to get a bit lost. I only get the most basic idea of what you mean (honestly the hammer thing really helped lol)
As you said, the people that know about these things aren't watching this video. It's the ones that have no idea that are here.
So, can you recommend SIMPLE and SHORT books or videos to get the idea?
I'm personally very lazy and not very smart so it would really help. Or better yet, make a video. The advantage of you making it is that you can control what you talk about, therefore giving a much better understanding of what you're saying.
Also, personally, i'm very visual, and so are many others. I think it's a good idea to show how things in different scripts are related. Even if you just said something 2 seconds ago, it's hard for people to keep it in mind at the same time you're explaining something else. And then understand that well enough to remember it for the next thing you're talking about.
Check out godotneers for a good way of showing things visually (i've only watched his data model video)
Having said all that, this video did have a lot of visual aides
Man i went for a long while lol, sorry about that
Thank you again
First, as extremely talkative ("writative", lmao) person, I'll never feel bad about reading a long letter-comment. The more the better and don't be ashamed to do that, I find great pleasure in talking in general. Thank you for your support.
Secondly, you are unbelievably on point :D. So, right now I'm releasing the first video on enemies subject, and I made it waaay slower as an experiment with format. Also somewhere in Sunday I'll do a voting on how people liked the slower pace. The whole enemies series will be slower no matter the voting outcome, for experiment purity. But make sure to put your vote, I'll happily bounce back to galloping tempo if people will say they liked that more ^__^.
Considering simple and short books, sadly, nothing comes to mind right now. But, if nothing sudden will happen, I have a plan to make a series about gamedev and programming for people with zero programming experience. and I want to make it reeealy far-coming, like, from "what is reality, what is an abstraction and how do we describe one with the other?" word-river. Somewhere mid-autumn probably.
About the tempo: It does seem like a hard thing to balance.
How do you make a well explained, information dense video, that doesn't alienate the people that are there to see it in the first place?
Slowing it down and being more detailed is probably a good thing to try.
But make it too slow, and people get the feeling like the video is very empty and they click off.
You'll only keep the people that are actually following the tutorial and people watching for entertainment (true weirdoes lol)
I guess both slow and fast are valid for different reasons
The series you mentioned sounds amazing. It sounds like exactly what i need. I feel like it would be a good base for whatever you do in the future.
This is damn insightful, I love the way you differentiate between tools and frameworks. I've been wondering, as a new Godot user and a long time user of Vue, what a framework might look like to encapsulate the complexities I've run into. I'm excited to see what other content you have, thank you for sharing!!
Tutorials only isolating and solving a single issue is really an understatement. I feel like 90% of devs run from tutorial to tutorial and never really dig their heels into fully understand concepts in their entirety forever chasing a dream idea.
I'm looking forward to this tutorial,. I'm new to Godot and getting my brain back into programming in general.
YOU know we are using State Machine Diagrams before this Machanism
wow great job, looking forward to ur controller!
I wish I could understand English better, sounds like you really know what you're talking about. Also, I might have missed/missinterpreted some parts of the video but I wonder will you make a video on animation blending? Not about blending trees but on how to implement a custom animation blending system (something to smoothly blend between idle/walk/run/dash animations depending on speed, maybe an implementation of randomized/partually procedural animations (altho I'd imagine if the system is flexible enough it would'nt be hard to make an extension for it to do all of that))
Maybe I will, but not in the nearest future, and it won't be some "superior custom animation system from the ground up". Godot's animation stuff is actually pretty good and capable for animation purposes, the whole video was as essay on the attempts to shove your game logic inside your animations. Considering your question about smooth walk/run/sprint, the answer you seek is not animation blending, but animation speed manipulation. Let's say you have your run animation default as 4 m/s speed and your sprint animation as 7 m/s. If you want to have a representation for about 5 m/s movement, you can't just "blend" run and sprint with some coefficient. Because to blend animations is essentially to sum a bunch of 3d-vectors with said coefficient and animation of "70% run, 30% sprint" will be a gibberish. What industry does instead is it speeds up the character and animation at exactly the same rate to get the feet touching floor synchronized with character speed. So you have your run animated corresponding to 4 m/s, then to simulate 5 m/s movement you just need to play your run animation 25% faster. Then there will be some level at about 5.6 m/s, where 40% accelerated run animation gives up and you start to play 30% slowed sprint instead. If you are aiming to replicate said effect with Godot, check the AnimationPlayer.play method, it actually has 3 optional arguments, one of those is playback speed exactly for such needs. Or you can "scale up" all the animations of the player with its field AnimationPlayer.speed_scale. See more information here: docs.godotengine.org/en/stable/classes/class_animationplayer.html#class-animationplayer-property-speed-scale.
I see, I remember trying to make something in unity 3-4 years ago and a lot of "beginner" tutorials usually pointed straight to the animation controller, which is a pretty powerfull tool but working around it is a huge mistake which back then ended up with me drawing pentagrams in state graph and trying to use engine only libraries to access current animation state which was a disaster. MVC might require a bit more work to get started but makes things a lot smoother in the long term and even if something goes wrong it'll be a lot easier to fix because you know how your code works. Hope more people see your tutorials!
русский акцент? information was so useful
Русский акцент. Thanks!) :D
The animation tree is just not consistent. I would rather blend and map out every cycle of my anims with the Animation Player. Thanks for this vid!
Yup. I also not so long ago faced a necessity to split my character into torso and legs to be able to add those sexy "torso only" actions like reloads etc. After bashing my head into blend tree for several hours and seeing how awful it is, I went into "fine, I'll do it myself" mode and threw together a system of two simple animation players masked by a simple Node on top directing and syncing them :D.
@@PointDown I've been looking through the docs whenever I get time. Haven't really dug into it yet but seems way more lenient and manageable. Do you have a video on blending the torso and legs manually without anim_tree? I'll take a look when I got free time.
@@pewpeee Yes, I do!) In fact, I have two of them:
Main one: th-cam.com/video/Fsa2wxyQvzM/w-d-xo.html
And a sequel: th-cam.com/video/huwW1JZt0BI/w-d-xo.html
I am also finishing the edit of the last update with a lot of bugfixes for several controller's systems including that double-animator's design. It will be out tomorrow, so, what I'd recommend is to watch those videos for design decisions narrations, but if you'll want to look at the code - to check the branch from that last controller's video as it have some minor problems fixed.
@@PointDown Wow, Thank you! I'll definitely check those out later. Gave you a sub earlier! Interesting stuff!
I love you so much can I please have your children
@@OmeletteGirl That escalated quickly :D
@@PointDown kek. this just happened to pop into my feed while I was making a fighting game and trying to decide how to implement a state machine, so needless to say I'm happy.