"Rapid Rig: Modular" - Procedural Auto Rig 2.4.8 for Maya (maya script)

"Rapid Rig: Modular" frees you from the confines of bipedal autorigs! So get creative!

Back to Feature Requests List

Feature Request

Quadruped front limbs and axis align.

Status:added
Date:02/10/2014
Submitted by: eallin eallin
Hi Dustin, I have to ask you for revision of IK structure of front limbs of the quadruped module in Modular rig. No offense and take it just as my opinion but the structure is not good and needs to be more straightforward from point of view of any animator. 1. The clavicula(*_FrontLeg_HipJIK in your solution): make it just FK. It is part of the first IK chain in the limb and it is not good for posing. It should be fully controllable with simple FK handle even in IK mode. There is good to have possibility to aim it to humeral joint(*_FrontLeg_Knee1JIK) with possibility to turn this aim off. But it is just tip and in my opinion it is not neccessary. 2. The humerus(*_FrontLeg_Knee1JIK in your solution) and ulna/radius(*_FrontLeg_Knee2JIK in your solution): the first IK chain of RP handle should be made of humerus and ulna/radius bones. The second one should be placed in just one bone: carpus(*_FrontLeg_AnkleJIK in your solution). *the pole vector handle should be placed BEHIND the elbow of the front limb. In the real world the front limb of any quadruped has the same functionality as the human one so bipedal character with several slight differencies like more freedom of twisting and rotation of felines, etc. ... but its origin is the same for all of mammals. Take it as just a hint but it would be very very helpfull. PS: my collegue gave me very good hint about pole vector aiming. It is good to have placed the pole vector handle to the side of the limb. You can pose it independently on position of the wrist(carpus) so the animator can see it all the time even the limb is crossing body geometry and the IK skeletal chain will not flip ever. At the end: I have to send you good words aswell: good work! I really like Rapid Rif Modular and it is very helpful for us. Thanks a lot, Josef.

Comments on this feature request:

  • eallin

    eallin said almost 13 years ago:

    I forgot about axis aligns: is it possible to align axis of the root node in the Quadruped mode to World Axis? You have the Y axis front one. It would be good to have it as the world Up axis. Thanks again and good luck with your excellent work, Josef.
    Report
  • Dustin Nelson

    Dustin Nelson said over 12 years ago:

    Hi Josef, Thanks for the feedback and sorry for the late response. I want to do a bit more research to make sure that would be a better method than what I currently have. I the meantime, you can create a single FK chain for the clavicle and then attach a leg with a single knee to that joint. For the pole vector, you can move the CV's of the control to where you want it to be. I could probably add a proxy for the positioning of the pole vector target, but I want to make sure it doesn't confuse the user that it is only for where the control would be generated and not where the knee/elbow would aim. Changing the root to be world aligned is pretty simple. You are the second person to request that I the quadruped setup so I will change that. I greatly appreciate your suggestions! User critiques is what make my tools better :) -Dustin
    Report
  • eallin

    eallin said over 12 years ago:

    Hi Dustin, Never mind the time(it is running too fast for people who work too much:)). I hope my feedback will help you to polish your tools. It is far enough good work already. So good luck! Thank you, Josef.
    Report
  • Dustin Nelson

    Dustin Nelson said over 12 years ago:

    I uploaded version 1.1.4 which should hopefully give a much better default quadruped proxy to work with. Let me know what you think. Thanks again! -Dustin
    Report
  • eallin

    eallin said over 12 years ago:

    Hi Dustin, good job! I think "upper humanoid" solution for front limbs is exactly what Rapid Rig Modular needed. One thing for revision: check the *_Spine_MidIKC handle. It has bad axis align with the *_ROOTSecondaryC and *_Spine_TopIKC(these two handles dont match in local space so it is not possible to blend their translate attributes in vector mode). The *_Spine_MidIKCG2 is fed by blendColour node *_Spine_MidIKCG_Inf_Trn_Blend If I could be so much picky(:)) I would like to ask you for last two features: First One: you have custom attribute HeelBall on both front and back IK feet of quadruped(I mean it for biped aswell). Is it possible to add one handle for just rotating of whole skeletal chain of the limb around the Ball? It maybe sounds like nonsense but in my opinion it would be good to give animator possibility to nuantly tweak rotation around the Ball joint. I mean it as subhierarchical solution: the HeelBall custom Attribute is very good concept so leave it as it is and the additive handle can be driven with this custom attribute with following it. In my opinion it is not so difficult to apply it into hierarchy. Second One: is the *_Clavicle_MasterC handle neccessary for clavicular solution? In my opinion it would be enough to have the custom attribute "Parent_Orient" directly on Clavicle handle. Thanks, Josef.
    Report
  • Dustin Nelson

    Dustin Nelson said over 12 years ago:

    Hi Josef, Sorry for the late reply. I thought I had responded but somehow missed it. I have (hopefully) resolved the issue with the Spine_MidIKC. Give it a go in the latest version and let me know if the behaviour is better. In regards to the Ball control, the FootIKC has a Ball Swivel attribute, does that not achieve what you desire? The MasterC on the Clavicle is a standard control for Look-At's and FK Chains. Even if there is just one chain/look-at. It's there for consistency because it holds the common attributes of the module. Due to dependencies in the script to look for that control and how I have structured the scripts, it is not something I prioritized to make a special case for. -Dustin
    Report
  • eallin

    eallin said over 12 years ago:

    Hi Dustin, I just tried the latest version of your solution and the Spine_MiddleIKC is still fed with wrong axis from the *_ROOTSecondaryC handle. The input translation of Spine_MiddleIKC is fed with Y axis from the *_Spine_TopIKC handle BUT with Z axis from the *_ROOTSecondaryC handle so when you move the *_ROOTSecondaryC handle the problem is happenning. Regards, Josef.
    Report
  • Dustin Nelson

    Dustin Nelson said over 12 years ago:

    Hi Josef, sorry for taking so long to get back to you on this. I was on holidays. I uploaded version 1.3.0 which has a fix for the MiddleIKCtrl. Among a bunch of other fixes and icons, one thing you will also notice is I have changed the naming conventions to be more descriptive: C>Ctrl G>Grp J>Jnt This makes it a lot easier for users using custom code to use the rig. Let me know if the latest version fixes the issue for you. I am hoping it will! -Dustin
    Report
  • eallin

    eallin said over 12 years ago:

    Hi Dustin, it looks all works well now. Thanks for your solution. Good work! Josef.
    Report

Post a comment:

Sign in
Forgot your password?