Jump to content

Steadicam

Members
  • Content count

    5
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

Community Reputation

0 Neutral

About Steadicam

  • Rank
    Peon

Personal Information

  • Name
    Steadicam
  • Location
    Steadicam
  1. I was attempting to build a type of link system using the law of cosines and discovered that I need to use the "spherical law of cosines" instead for calculating this "spherical triangle" as seen below. I'm already pushing my limited math-in-Houdini skills and was hoping someone familiar with the concept could help me approach it in a good Houdini way to obtain the data/measurements needed to complete the solve for each angle in A B & C below. In my use-case the simple image below represents a neutral position for three points. Please note that the side measurements shown are NOT in radians ( simply distance from points) which would be a requirement to solve. •Each point (A,B,C) is a fixed distance from the center of a sphere. Essentially creating a spherical triangle. •Each point only rotates about one axis between it and the center of the sphere creating angles •The A point is in a fixed position on the sphere and does not move. Consider it a kind of base point •The C point travels, with a limited range along the surface of the sphere(essentially rotating it's position from the center if the sphere). •The distance between A & B (AB) and B & C (BC) are both fixed lengths and function as arms. •The B point essentially functions as an elbow and always stays the same distance between both A & C respectively as the C point moves. Essentially I need C to be driven by XYZ rotations from the center of the sphere and for B to "chase it" like a hinged elbow would "chase" a hand within the constraints of it's distance between A & C. Using the Spherical Law of Cosines I should be able to measure the new distance of CA to find all the angles of A, B & C (in Radians?). https://en.wikipedia.org/wiki/Spherical_law_of_cosines Essentially a simple inverse kinematic that avoids using the Houdini bone system. Since A is always a fixed point I should the be able to simply compare the neutral angle of A to the new one and use that difference to drive the animation necessary to rotate both A and B to position the the BC and AB sides/arms to align with the C position as it travels along the suffice of the sphere. I've included the scene below for reference but it's only a representation and the only potentially useful item is my measuring widget which unfortunately cannot do radians at this point. Thanks for any help or pointers. Michael Law_of_cosines.hip
  2. Spherical law of cosines help

    Accidentally put this in scripting. Apologies to list moderators. Will repost in general if that is alright.
  3. Spherical law of cosines help

    I was attempting to build a type of link system using the law of cosines and discovered that I need to use the "spherical law of cosines" instead for calculating this "spherical triangle" as seen below. I was already at the limit of my limited math-in-Houdini skills and was really hoping someone could help me approach this the best way to obtain the data/measurements needed. The simple image below represents a neutral position for three points. •Each point (A,B,C) is a fixed distance from the center of a sphere. Essentially creating a spherical triangle. •Each point only rotates about one axis between it and the center of the sphere creating angles •The A point is in a fixed position on the sphere and does not move. Consider it a kind of base point •The C point travels, with a limited range along the surface of the sphere(essentially rotating it's position from the center if the sphere). •The distance between A & B (AB) and B & C (BC) are both fixed lengths and function as arms. •The B point essentially functions as an elbow and always stays the same distance between both A & C respectively as the C point moves. Essentially I need C to be driven by XYZ rotations from the center of the sphere and for B to "chase it" like an elbow would "chase" a hand but within the constraints of it's single rotation and distances between A & C. Using the Law of cosines (or in this case the Spherical Law of Cosines) I should be able to measure the new distance of CA to find all the angles of A, B & C (in Radians?). https://en.wikipedia.org/wiki/Spherical_law_of_cosines Essentially a simple inverse kinematic that avoids using the Houdini bone system. Since A is always a fixed point I should the be able to simply compare the neutral angle of A to the new one and use that difference to drive the animation necessary to rotate both A and B to position the the BC and AB sides/arms to once again align with the new C position. I've included the scene below for reference but it's only a representation and the only potentially useful item is my measuring widget which unfortunately can't do radians at this point. Thanks for any help or pointers. Michael Law_of_cosines.hip
  4. I have two points on a spherical surface. Each point should only rotate along it s own Y axis (which is simply an axis along the from the center of the sphere to the point itself). A good analogy example would be two flagpoles standing on a small planet and I want each flag to rotate in the direction of the other flagpole. Not pointed at the other flag but simply calculate the y rotation of each flagpole to most orient itself in the direction of the other. I'd also like to stay out of VEX for this. I did create a clunky solution which utilized a boolean intersection operation based on some geometric shapes I created based on the point position but it was really convoluted and not a very mathmatical solution. Any thoughts on the simplest approach? I feel like it should be really simple but for some reason I'm really struggling with this. Thanks!
  5. I need to revolve a curve around an axis between two points. I might be missing something obvious but I can't figure out how to do this in obj/modeling. the Revolve tool would otherwise be perfect if it wasn't requiring the "direction" value to function. I can't figure a way to align that automatically with a point's XYZ from the origin. Any suggestions?
×