Thank you for creating this tutorial. You've really helped as I've been struggling for a while getting my enemy NPC character to look at various other random NPC targets before engaging them. I've now got the enemy NPC not only looking at the targeted NPC but it also rotates the body as well giving it a more life like motion. All of this is done via control rig and I am really happy with how it's turned out :)
Amazing tutorial. How would I modify this for a top down game? I've seen/played a few games where you aim at enemies who are on a different elevation to you and vice versa
Sir are you planning to do a walkthrough of the new Animation Sample Project? I've been trying hard to understand and hopefully detach it to be used for my own project, but I'm starting to get overwhelmed trying to understand the whole system 😅
this is super cool! I don't totally understand the theory, though -- why do we need the ztrs here? don't the attr just follow the lookat and the bones follow the attr?
one more question: I thought that parent/rotation constraints were "permanent" things -- you establish the constraint and then they will stay forever. why do you keep re-establishing the constraint during the forward pass vs. once in the construction?
It's just layering. The you can have as many nodes as you need above and they might all serve different purposes...The first node is the parent that has the transforms then the next node is a zero'd out transform that can be used for local offsets and maybe one for constraints etc etc
SO I thought the way you did. But so far that seems to be the thinking from other DCC packages, where as here its evaluated every frame cuz it's state based. I'm pretty sure I"ve gone round and round on this a couple of times though
It seems that the control rig variables are not accessible from the AnimBP if you use a modular control rig. Although you can create variables, and only variables, in the Modular Control Rig interface I can't find a way to connect these back to the Rig Modules. However you can access animation channels/controls from the AnimBP just fine, so you could replace the variables with animation channels inside your rig modules instead though.
I don't think I have even tried this yet. Largely cuz so far the rigs I've worked on primary purpose is runtime and very specific. If I get back into the modular stuff and find a way to do this I will let you know
hi, i am trying to figure out how animation work, is Control Rig in UE5 the same as animated in Blender or Maya ? Is Control Rig a replacement for those 3D software ?
Hi. Thanks for the tutorial. please i have a question if you can assist. i have a long puncho that goes from the shoulders to the feet. i want to add some cloth effect to it. which is better a cloth sim or Rbans ( how many bones and are cross constraints achieved with linear limits) for a game or a mix. the main concern is that the cloth passes through itself when the arms are in the rest position. And another if you dont mind. i want to animate the puncho can i switch the skeleton before the animation because the mesh gets deformed wierdly in cinema 4d and i cant use morp targets with cloth sim.. PLEASE any assistance will be great. THANKS
That question sort of rolled back on itself. Generally you can use cloth or you can use physics, it depends on the quality bar. You can also use both and in certain animations disable cloth and go with whatever you animated in your DCC.
What would be needed to be able to use the lookat to "complement" the animation? As of right now it kind of hijacks the transforms completely, so if there was some breathing animation or something its lost?
I think breathing would need to be additive or the rig would need to be setup to handle it with offset bones for example. I think there is also a version where the input pose is taken into consideration as part of the solve. I’ll have to see if I can mock something up.
So rather than using the initial transform of the bone you would use the current from the input pose. Let me think about it and I’ll try some stuff out.
@@livinfreestyle6727 Yeah I guess what I was conveying (badly) might be that a great deal of the upper body animation is no longer perceived because the aim constraint takes over. So I guess it would be a matter of finding the delta between the constrained orientation and the animated orientation and kind of using it as additive. Otherwise yeah I guess you could add some joints in the hierarchy that their sole purpose is to receive the IK constraint. But it effectively doubles the joint count then for the spine and neck. Im starting on my Control Rig journey so it's all quite interesting. :D Anyways! thanks for the reply!
@@ElectricBlobz another thing you could do depending on your game camera is try limiting it to just the twist axis and you might find you get More of your idle. Either way I’ll try some experiments and get back to you all
I`m trying to accomplish a goal for my core game loop: i need to find a way to put random 3d objects in another object (imagine cookies in a cup), and i need to calculate how much percent of the cup is filled. Objects and cups can be different forms and sizes. I`m i bit lost, any tips on where to start and is it possible at all :)
I guess it depends on the complexity of the objects but one thing you could do is if you know the volume of the container, just divide it by the volume of the objects. You could also set it up as a table...this container can hold this many objects. If it doesn't have to be perfect you could also simulate the objects to drop in and the ones that won't fit just wont fit
@@livinfreestyle6727 Thanks. I was also thinking about that. But there is not build-in funcionality in unreal to get the volume of different and complex objects, right? Because for example in AutoCad you can easily see it in the object properties
Please, never stop making tutorials, I've learned from you more than anyone else in this platform, thank you for your work
This definitely made my day. I will try to keep making tutorials but pretty soon you all will know everything!
Thank you for creating this tutorial. You've really helped as I've been struggling for a while getting my enemy NPC character to look at various other random NPC targets before engaging them. I've now got the enemy NPC not only looking at the targeted NPC but it also rotates the body as well giving it a more life like motion. All of this is done via control rig and I am really happy with how it's turned out :)
Awesome!!! 🎉
Yes plz don't stop you are making future for young game developera ❤
👍
Amazing tutorial. How would I modify this for a top down game? I've seen/played a few games where you aim at enemies who are on a different elevation to you and vice versa
Just have it use the mouse position for the look at instead of a made up vector from character rotation
Sir are you planning to do a walkthrough of the new Animation Sample Project? I've been trying hard to understand and hopefully detach it to be used for my own project, but I'm starting to get overwhelmed trying to understand the whole system 😅
we did the Motion Match anim turn in place last week, if that's helpful
this is super cool! I don't totally understand the theory, though -- why do we need the ztrs here? don't the attr just follow the lookat and the bones follow the attr?
one more question: I thought that parent/rotation constraints were "permanent" things -- you establish the constraint and then they will stay forever. why do you keep re-establishing the constraint during the forward pass vs. once in the construction?
It's just layering. The you can have as many nodes as you need above and they might all serve different purposes...The first node is the parent that has the transforms then the next node is a zero'd out transform that can be used for local offsets and maybe one for constraints etc etc
SO I thought the way you did. But so far that seems to be the thinking from other DCC packages, where as here its evaluated every frame cuz it's state based. I'm pretty sure I"ve gone round and round on this a couple of times though
@@livinfreestyle6727 fascinating. thanks for the reply and the really helpful videos!!!
It seems that the control rig variables are not accessible from the AnimBP if you use a modular control rig. Although you can create variables, and only variables, in the Modular Control Rig interface I can't find a way to connect these back to the Rig Modules. However you can access animation channels/controls from the AnimBP just fine, so you could replace the variables with animation channels inside your rig modules instead though.
I don't think I have even tried this yet. Largely cuz so far the rigs I've worked on primary purpose is runtime and very specific. If I get back into the modular stuff and find a way to do this I will let you know
hi, i am trying to figure out how animation work, is Control Rig in UE5 the same as animated in Blender or Maya ? Is Control Rig a replacement for those 3D software ?
I think this vid might help:
ua-cam.com/users/liven6BUie-56Ss?feature=share
Hi. Thanks for the tutorial. please i have a question if you can assist. i have a long puncho that goes from the shoulders to the feet. i want to add some cloth effect to it. which is better a cloth sim or Rbans ( how many bones and are cross constraints achieved with linear limits) for a game or a mix. the main concern is that the cloth passes through itself when the arms are in the rest position. And another if you dont mind. i want to animate the puncho can i switch the skeleton before the animation because the mesh gets deformed wierdly in cinema 4d and i cant use morp targets with cloth sim.. PLEASE any assistance will be great. THANKS
That question sort of rolled back on itself. Generally you can use cloth or you can use physics, it depends on the quality bar. You can also use both and in certain animations disable cloth and go with whatever you animated in your DCC.
@@livinfreestyle6727 thanks for the reply. Do you have a discord. Or any you can recommend
@@ajibolalaleye769 We do, link is in the video comments section
What would be needed to be able to use the lookat to "complement" the animation? As of right now it kind of hijacks the transforms completely, so if there was some breathing animation or something its lost?
I think breathing would need to be additive or the rig would need to be setup to handle it with offset bones for example. I think there is also a version where the input pose is taken into consideration as part of the solve. I’ll have to see if I can mock something up.
So rather than using the initial transform of the bone you would use the current from the input pose. Let me think about it and I’ll try some stuff out.
@@livinfreestyle6727 Yeah I guess what I was conveying (badly) might be that a great deal of the upper body animation is no longer perceived because the aim constraint takes over. So I guess it would be a matter of finding the delta between the constrained orientation and the animated orientation and kind of using it as additive. Otherwise yeah I guess you could add some joints in the hierarchy that their sole purpose is to receive the IK constraint. But it effectively doubles the joint count then for the spine and neck. Im starting on my Control Rig journey so it's all quite interesting. :D Anyways! thanks for the reply!
@@ElectricBlobz another thing you could do depending on your game camera is try limiting it to just the twist axis and you might find you get More of your idle. Either way I’ll try some experiments and get back to you all
I have a question: how to spawn a vector 2d value type controller with maybe specific primary axis order? If it possible
Ok, I figured it out by myself :)
Nice self solve!
I`m trying to accomplish a goal for my core game loop: i need to find a way to put random 3d objects in another object (imagine cookies in a cup), and i need to calculate how much percent of the cup is filled. Objects and cups can be different forms and sizes. I`m i bit lost, any tips on where to start and is it possible at all :)
I guess it depends on the complexity of the objects but one thing you could do is if you know the volume of the container, just divide it by the volume of the objects. You could also set it up as a table...this container can hold this many objects. If it doesn't have to be perfect you could also simulate the objects to drop in and the ones that won't fit just wont fit
@@livinfreestyle6727 Thanks. I was also thinking about that. But there is not build-in funcionality in unreal to get the volume of different and complex objects, right? Because for example in AutoCad you can easily see it in the object properties