So, this is the new series. I have plans for the series, but they are not set in stone. So that's what we'll do. Beneath you can see a list of features I plan to plan to have after this series. The closer to the top - the better chance of me actually going for it. However, till about December you can heavily influence my choice about it, because I might have forgotten something, or misinterpreted audience's desire to have something implemented. So, under this comment, or just by commenting you can add your inputs on what do you expect from a parkour series. Or, like other people's comments. In December I'll look into the comments and see if there were some interesting options I haven't considered, but people want. 1. Ledge grab and climb (done) 2. Wall jump and or double jump (done) 3. Beam walk (done) 4. Ledge traversal and more ledge actions 5. Start/stop animations for locomotion 6. IK for legs 7. Slides like in apex/deadlock 8. Toggle-able crouch just to have it 9. Rope walking with a true rope/cable 10. Wall running 11. Raw wall climbing zelda style 12. Climbing w/points uncharted style 13. IK for arms 14. Procedural animations when close to a wall 15. Ladders 16. Hookshot or similar techs (but idk if it's a good idea for 3d).
I think this is a great idea and look forward to this series no matter what you choose to do with it. I would really like to see the fundamentals covered (which perhaps you do in your Character Controller Done Right playlist) and then expand upon that with examples. I think procedural movement is important for a good parkour system, so IK for arms and legs is important. I would also like to see movement over short obstacles such as a fence, table, or car hood. You have sliding as an option. Would you consider covering skiing and snowboarding/skateboarding?
@@user-rx3xl7zn1u Yeah, first character controller playlist is a progress from first 10 lines the engine suggests to you to the 1k lines scalable codebase, with fundamentals and thought process. Yup, fences and short vaults are a nice idea, will add it!) Don't think I'll add skiing and snowboarding, but I'll probably make a witcher3-ish mechanic that forcibly makes character slide from too steep hills, and from there I think snowboarding is just to add the board.
I'd like to suggest, somewhere on the list, having multiple actions being doable on a wall or slide for example. Say you can climb with manual control, and slide, but you can also use a sword during those actions. Would there be separate climb, slash_while_climb, heal_while_climb states? how would transitions work?
@@natanmaia3575 Yeah, those will be the separate world. For combat we'll need to do "mapping doubledip", to ask both weapon and area awareness on their opinions and decide what to do. And those actions are certainly separate states as they will brobably have different backend animations, different input vector processing (won't rotate on the ledge for example) etc. Plus there probably will be unique transitions, like, heal-on-land staggers if hit, but heal-on-ledge clumsy-drops with a long punishing stun, awkward hands and legs shakes and vilgelm's scream.
Without opening the video, I already knew that the video would be awesome! Keep it up my man! You're an absolute inspiration for a lot of us aspiring games devs and enthusiasts!
Awesome video on theory! I have no idea how Godot specifically works but it is really nice to learn about theory on how some things are implemented in games.
Well, melee - be my guest, the channel already has a bunch of stuff for melees, just adapt them for shooter. But guns and FPS I don't think will happen. Firstly, I don't really like shooters and I don't really like first person view. Maaaybe in like 4+ years if I'll be alive, but run out of ideas for 3d person content.
Well, I plan to adapt every single one of these systems for multiplayer, as multiplayer is a planned series for the future :D. Sometimes late winter - early spring 2025 I believe. Long story short, the main things for great multiplayer is to choose your synchronized entity and to learn how to work with time making it "editable" in runtime for your game. And with our well encapsulated state abstraction with already manageable parameter of progress, half of the work is done actually.
@@PointDown Yeah I think the biggest challenge I have right now is the usage of the animation tracks for a lot of the controls in some of your other videos. I'm not really sure how to get that synched over a network yet. I'm gonna see if I can get something put together with what you've taught so far though!
@coltr1 Well, that's not an issue, but a blessing. I don't use animations for controls, I use animations to store data, *but I never play them*. This means, no strange time management artifacts will occur when. Iet's say, we reconcile with the server to substract 0.05 seconds of root motion. In my system I just calculate that motion from backend animation, substract it, and then substract 0.05 seconds of progress from the state. The operation is a bit more complex when we need to reconcile a transition in the past, but it's all ultimately doable, exactly because my systemis not bound by animations by any means, my systemis just math that can calculate it's state in the future and in the past, if you provide the delta time.
@ Yeah you have a great system here, I just need to think more about how to implement multiplayer and learn more about Godot. Love your work, I’m glad that we got people like yourself exploring real technical implementations. If you’re ever up to it I would love to chat some time. Either way, great videos as always!
@@coltr1Thanks!) Sure thing we can talk, I try to answer in discord whenever I'm free, or you can write me an email for more oldfashioned ways. The links are in the channel's description.
I do respect your approach, but I feel like this is outdated since AAA Studios nowadays use Motion Matching. Recently Unreal Engine published Motion Matching and how they made it accessible for every Platform (PC, Consoles, Mobile). I find it very nice to have something like this and I think this could be a very nice foundation to advance into Motion Matching. Theoretically it's possible to study how Epic Games implemented Motion Matching since you can access the Source Code. It's a very interesting Topic and good luck for this project!
Thanks!) Well, "AAA" is there mainly for a spicy title. I don't personally think the approach is outdated, because let's be real, Godot still is, and probably will be for at least 5 more years, an engine for indies and compact software enthusiasts that do what they do mainly as a hobby or as a pre-financing prototype. It means that even if truly created, the mighty motion mapping technology will be a fifth wheel for most of the users, because it's a machine that is powered by having 150 animations for locomotions, not including combat animations. And those who I know and me myself, well, are happy if we have an animation per action, and that's not always the case and we forced to use ugly proxies. However, "the spirit" of motion matching, i.e. actual trajectory:animation mapping will be implemented in some form. I want at the end of the series to have at least a couple of vaults, a fence and a simple step up. And a system that is "modern" enough for me to process stepping up, sliding a butt over the fence and vaulting, while the player just presses W button forward. In essence, it will be a proto-motionmatching and I'm looking forward to steal some techniques.
TPS animations are recorded for a neutral position, like "looks forward, gun parallel to the ground". Then they are played, but also, torso is rotated in a way in accommodates to your aiming direction. First, probably watch this vid: ruclips.net/video/xpoPfUKI9tw/видео.html I am not a fan of shooters and don't think there will be tutorials on them soon, but during the parkour series we'll use this tech for our IK, from there just adapt it to rotate torso.
@@firerunnew Ахах Ну, спасибо!) Но это не экспромт, запись, которую я потом чищу руками, удаляю затупы, бэ, мэ, шумные вдохи и тд. Ну, и читаю, разумеется, скрипт, который написан перед этим в тексте. Но прогресс заметный, от части канал англоязычным сделал ради практики, можешь почекать первые видосы, там гораздо хуже все, а всего пол года прошло).
So, this is the new series. I have plans for the series, but they are not set in stone. So that's what we'll do. Beneath you can see a list of features I plan to plan to have after this series. The closer to the top - the better chance of me actually going for it. However, till about December you can heavily influence my choice about it, because I might have forgotten something, or misinterpreted audience's desire to have something implemented. So, under this comment, or just by commenting you can add your inputs on what do you expect from a parkour series. Or, like other people's comments. In December I'll look into the comments and see if there were some interesting options I haven't considered, but people want.
1. Ledge grab and climb (done)
2. Wall jump and or double jump (done)
3. Beam walk (done)
4. Ledge traversal and more ledge actions
5. Start/stop animations for locomotion
6. IK for legs
7. Slides like in apex/deadlock
8. Toggle-able crouch just to have it
9. Rope walking with a true rope/cable
10. Wall running
11. Raw wall climbing zelda style
12. Climbing w/points uncharted style
13. IK for arms
14. Procedural animations when close to a wall
15. Ladders
16. Hookshot or similar techs (but idk if it's a good idea for 3d).
I think this is a great idea and look forward to this series no matter what you choose to do with it.
I would really like to see the fundamentals covered (which perhaps you do in your Character Controller Done Right playlist) and then expand upon that with examples. I think procedural movement is important for a good parkour system, so IK for arms and legs is important. I would also like to see movement over short obstacles such as a fence, table, or car hood.
You have sliding as an option. Would you consider covering skiing and snowboarding/skateboarding?
@@user-rx3xl7zn1u Yeah, first character controller playlist is a progress from first 10 lines the engine suggests to you to the 1k lines scalable codebase, with fundamentals and thought process.
Yup, fences and short vaults are a nice idea, will add it!)
Don't think I'll add skiing and snowboarding, but I'll probably make a witcher3-ish mechanic that forcibly makes character slide from too steep hills, and from there I think snowboarding is just to add the board.
I'd like to suggest, somewhere on the list, having multiple actions being doable on a wall or slide for example. Say you can climb with manual control, and slide, but you can also use a sword during those actions. Would there be separate climb, slash_while_climb, heal_while_climb states? how would transitions work?
@@natanmaia3575 Yeah, those will be the separate world. For combat we'll need to do "mapping doubledip", to ask both weapon and area awareness on their opinions and decide what to do. And those actions are certainly separate states as they will brobably have different backend animations, different input vector processing (won't rotate on the ledge for example) etc. Plus there probably will be unique transitions, like, heal-on-land staggers if hit, but heal-on-ledge clumsy-drops with a long punishing stun, awkward hands and legs shakes and vilgelm's scream.
I dont know if its part of one of the points of the list, but im trying to point my gun of my 3rd person character into the direction im looking at
Without opening the video, I already knew that the video would be awesome! Keep it up my man! You're an absolute inspiration for a lot of us aspiring games devs and enthusiasts!
Awesome video on theory! I have no idea how Godot specifically works but it is really nice to learn about theory on how some things are implemented in games.
legend
i'm still trying to make a functional 3d bow thanks very much
Can you make system for fps weapon (guns,melee)and stuff?
Well, melee - be my guest, the channel already has a bunch of stuff for melees, just adapt them for shooter. But guns and FPS I don't think will happen. Firstly, I don't really like shooters and I don't really like first person view. Maaaybe in like 4+ years if I'll be alive, but run out of ideas for 3d person content.
I would love to see how you could adapt some of these systems for multiplayer if you ever have the time to do so. Great work as always
Well, I plan to adapt every single one of these systems for multiplayer, as multiplayer is a planned series for the future :D.
Sometimes late winter - early spring 2025 I believe.
Long story short, the main things for great multiplayer is to choose your synchronized entity and to learn how to work with time making it "editable" in runtime for your game. And with our well encapsulated state abstraction with already manageable parameter of progress, half of the work is done actually.
@@PointDown Yeah I think the biggest challenge I have right now is the usage of the animation tracks for a lot of the controls in some of your other videos. I'm not really sure how to get that synched over a network yet. I'm gonna see if I can get something put together with what you've taught so far though!
@coltr1 Well, that's not an issue, but a blessing. I don't use animations for controls, I use animations to store data, *but I never play them*. This means, no strange time management artifacts will occur when. Iet's say, we reconcile with the server to substract 0.05 seconds of root motion. In my system I just calculate that motion from backend animation, substract it, and then substract 0.05 seconds of progress from the state.
The operation is a bit more complex when we need to reconcile a transition in the past, but it's all ultimately doable, exactly because my systemis not bound by animations by any means, my systemis just math that can calculate it's state in the future and in the past, if you provide the delta time.
@ Yeah you have a great system here, I just need to think more about how to implement multiplayer and learn more about Godot. Love your work, I’m glad that we got people like yourself exploring real technical implementations. If you’re ever up to it I would love to chat some time. Either way, great videos as always!
@@coltr1Thanks!)
Sure thing we can talk, I try to answer in discord whenever I'm free, or you can write me an email for more oldfashioned ways. The links are in the channel's description.
I do respect your approach, but I feel like this is outdated since AAA Studios nowadays use Motion Matching. Recently Unreal Engine published Motion Matching and how they made it accessible for every Platform (PC, Consoles, Mobile).
I find it very nice to have something like this and I think this could be a very nice foundation to advance into Motion Matching.
Theoretically it's possible to study how Epic Games implemented Motion Matching since you can access the Source Code. It's a very interesting Topic and good luck for this project!
Thanks!)
Well, "AAA" is there mainly for a spicy title. I don't personally think the approach is outdated, because let's be real, Godot still is, and probably will be for at least 5 more years, an engine for indies and compact software enthusiasts that do what they do mainly as a hobby or as a pre-financing prototype. It means that even if truly created, the mighty motion mapping technology will be a fifth wheel for most of the users, because it's a machine that is powered by having 150 animations for locomotions, not including combat animations. And those who I know and me myself, well, are happy if we have an animation per action, and that's not always the case and we forced to use ugly proxies.
However, "the spirit" of motion matching, i.e. actual trajectory:animation mapping will be implemented in some form. I want at the end of the series to have at least a couple of vaults, a fence and a simple step up. And a system that is "modern" enough for me to process stepping up, sliding a butt over the fence and vaulting, while the player just presses W button forward. In essence, it will be a proto-motionmatching and I'm looking forward to steal some techniques.
Bro I need help can you tell m e how tps animation works ,or male a tutorial in it 😅
TPS animations are recorded for a neutral position, like "looks forward, gun parallel to the ground". Then they are played, but also, torso is rotated in a way in accommodates to your aiming direction.
First, probably watch this vid:
ruclips.net/video/xpoPfUKI9tw/видео.html
I am not a fan of shooters and don't think there will be tutorials on them soon, but during the parkour series we'll use this tech for our IK, from there just adapt it to rotate torso.
@PointDown I am not having problems in shooting it's easy but locomotion is hard for me
Юху! Новый видос! )) кайф, благодарю
Жаль я английский не знаю, и яндекс не переводит
@@S4miGul Я добавлю сабы на английском завтра, сделанные руками. С них потом можешь ютубский автоперевод сделать, терпимо.
Но вообще, повод учить :D
Ты русский? акцент очень похож )
@@firerunnew Да
@@PointDown Ого, но говоришь все ровно смело, мне бы так)
@@firerunnew Ахах
Ну, спасибо!)
Но это не экспромт, запись, которую я потом чищу руками, удаляю затупы, бэ, мэ, шумные вдохи и тд.
Ну, и читаю, разумеется, скрипт, который написан перед этим в тексте.
Но прогресс заметный, от части канал англоязычным сделал ради практики, можешь почекать первые видосы, там гораздо хуже все, а всего пол года прошло).