Відео

Unfortunately that is a limitation of the python code completion. I have not tried other LSPs but the default python setup in VSCode will only give completion when referencing libraries directly.

@@brundlethwaite Thanks for the reply, in case you find something on this please do let me know

Ideally this will be running on the Frimble engine but I do not know If it will ever get that far. This is a prototype so it may never be released other considerations are. - My interest here is animation and Rigging not programming or releasing a game Unity is the easiest to deal with at the moment. - Unity is easy to use, I know it and it is the most fun to work with at the moment. - As this is a prototype Unity is free. - Unreal would be far more expensive than Unity (even if they had implemented that stupid pricing scheme). - If I release the game It would be for consoles Godot has no support and would end up being more expensive than Unity. - Godot does not have Houdini Engine support. - Flax does not support graphics tablets. - Flax does not have native Houdini engine support. - Cryengine does not have proper UI and requires defunct commercial software. - Cryengine does not have native Houdini engine support. - Programming in Unreal is horrible and is it weakest area. - For game logic I do not want to use blueprints as they and not efficient and have worse performance that C++. - I do not want to use C++ I would prefer to work with C, Odin or C#. - Animation is Unreal's weakest area Unity is far easier to work with here. - Animation import and Export it Unreal's weakest area. - Unreal's coordinate... - Unreal's syntax..... ... ... This is a little brief, I start to give some more reasons but I do not want this to become a long list and I do not have the time to go into even the basic reasons as each of them require context and explanation to make sense.

@@brundlethwaite no way! Unreals animation is weaker than unity’s? I spent sometime with Unity and hated C++ until I started using it and now I look at c# as a jumbled mess lol. I do like Unity’s DOTS I almost switch back to it because of that alone.

"no way! Unreals animation is weaker than unity’s?" What relevance does this have? I did not make a feature comparison? This is not a reason or advice on choosing. And it is not relevant to anybody by me, for this specific project. " I spent sometime with Unity and hated C++ until I started using it and now I look at c# as a jumbled mess lol." What relevance does this have? This is not a programming language comparison? These are really bullet points as they applies to this specific project. I am not going to spend days (or for this topic months and years) of my life trying to explain the specifics. You asked me why I am working with Unity you did not ask me why someone else should.

@@brundlethwaite From your first reply "- Animation is Unreal's weakest area Unity is far easier to work with here." I was replying to this. Maybe you worded this wrong? From your first reply "- I do not want to use C++ I would prefer to work with C, Odin or C#. " I was replying to this point. I used to prefer c# now I look at it different after using C++ My question initially was to see whats stopping you from converting to UE. I presented some slight counter arguments to push you towards it. Since your content + Unreal would be beneficial to me. No disrespect to you or unity was intended

"My question initially was to see whats stopping you from converting to UE." You did not state that. You said "What are the reasons you continue with Unity over Godot or Unreal Engine?" which is a completely different question. I am not converting from anything I currently have projects in Unreal, Unigen, Flax, Godot, Cocos2D and Stride and Cryengine. Not all of the projects are fully active but I none are discarded. So if you had asked the first me about converting I would have said that I am using Unreal different projects instead of why I am using Unity in this one. You also chose to ignore the final statement where I said that each of the points was highly contextual. As well as the first statement.

Good to hear. The group and point number issues took me a while to figure out and they can be tricky.

This is limiting the controls not the joints. There is no built in limit to control IK chains (you can make one manually in vex or VOPS). You can force IK chins them to return to an original pose using rest transforms. Using rest transforms will stop you joints from going beyond a straight line as they will always return to the default. I do not think this applies to the IKchains node. As that uses the 2 bone Ik VOP. But if you use the IK Solver VOP you have the rest angles mode.

@@brundlethwaite I`ve switched to apex. I think its to much work to build a costum one as a generalist. so i`ll skip over it for now. THX for the reply !

The character needs to move relative to the camera. So the controller is essentially moving the character in screen space. But the character is not being driven by the controller but is using root motion and is purely driven by the animation. Essentially this is a animation prototype 1st and if I get a good results from that I may turn it into controller prototype. There is also some tool creation that is incidental to the prototypes/

@@brundlethwaite ah, I gotcha. Root motion can be tricky but is well worth it in the long run.

There are a series of videos for the prototype for a bone based rigging system on my Patreon. I am working on a full course but editing and recording takes a long time and I have had a lot of interruptions so it is taking a lot longer than I want. Personally I dislike blendshapes as bone based rigs give much better interpolation so will be a bone-based course. Unfortunately it will have to be a paid course. I cannot justify placing it on UA-cam for free as it I too much work to do for no compensation when I have a very limited income.

When you want to extend a rig using the auto-rigging nodes you will need to create scripts like this this is the starting point for doing that.

This is not really a tutorial it is more documentation of me screwing around. I setup the MONOcamera in this video: ua-cam.com/video/l2OLPgTS0tU/v-deo.html Since the camera cannot be accessed by ECS and is a monobehaviour we need to create a hook with a monobehaviour script attached to the camera. This is the script: using System.Collections; using System.Collections.Generic; using UnityEngine; public class MONOCamera : MonoBehaviour { public static UnityEngine.Camera Camera; public void Awake() { Camera = GetComponent<Camera>(); } }

That is really just habit and does not have that much effect here. The "Into this node" option allows you to use the obj/ level transforms into the sop level. If you have an object containing a mesh and you transform it at the object level and then try to bring it into a separate sop you will not have the obj level transforms affecting the geometry so you will only get the local position not the global position.

What solvers are you using that are flipping? I make my transforms manually so that I know they are consistent. So I avoid flipping by always having matrices that do not flip (with the exception of quaternion rotation based spines). This is something that is very difficult to explain without knowing what your scene is doing. But I will try to make a video explaining my approach a cannot give an ETA though.

@@talesofthedeparted Thanks for the reply. I will be looking forward to your video explain your approach. I have been trying to figure this out ever since your deformation videos. I don't thing I am using any solver. Here is a link to my hip file: drive.google.com/file/d/1Hdvh2Vmc3VXx-fSRN4kuU90WfaLgO3Cx/view?usp=drive_link.

Unfortunately no. That is beyond the current scope of my personal projects and I do not have access the equipment needed. I have done some basic setups for facial animation using Computer Vision Libraries and have a facial setup for that but that does not use the MocapSteam node. I do want to get a mocap suit eventially but I cannot afford it at present. For my rigs I am currently using pose space blendshapes driven by expressions for my automatic deformations.

I see, well thank you anyhow! Your videos are immensely helpful for all things KineFx. They helped me get the blendshapes working I just need to figure out how to the load some sort of facial anim to drive them.@@brundlethwaite

These are the same places where I am having the most frustration. I will get to blends. But form what I have got working sliders are not great. At the moment it is easiest to work with on screen controls (which is not something that I like as I favor slider interfaces). Unfortunately there is nothing listed in the documentation but this is something that I am looking for solutions to. As for mocap I would personally stick with KineFX as it so flexible and unless you have extreme slowdown you will be loosing a ton of flexibility. The APEX setup gives you a lot more performance but the mocap approach is far more limited and is not dynamic. I have done some testing but am not yet satisfied with anything. So I am working on these and still have plenty of things to try but there are no real options that I am satisfied with yet.

@@brundlethwaite Thanks, yeah, just willing to use all the Animate state cool things, like constraints and locators, to clean up and adjust mocap data, since constraints in KineFX are pain.. And I feel like it's there just need to figure out how to set it up properly. Generally I've extracted controls (actually the whole skeleton) and retargeted with KineFX, and brought it back with ControlUpdateParms, now need to figure how make it work with all the IKs, and animate on top of it

I will try and I do intent to but It depends on time. I still need to do a lot of videos on FEM for example but I have not got round to them. I also want to do hair and cloth with kineFX rigs for game engines as well as secondary motion but those are a long way in future at the moment.

Awesome looking forward to it, especially the blendshapes@@brundlethwaite

From my limited understanding, it's basically a visual representation of how the apex graph is stored as geometry data. Each node in the APEX graph is represented as a point with attributes. I too found it rather bizarre at first but I think SideFX simply chose to store graph the data as geometry and the visualization is more of a side effect than a purpose. There's probably a better reasoning for storing the data as geometry instead of something like a text document.

Yes I am going to do some APEX videos.

Looking forward to it