IK with constraints now specify angular constraints relative to the rest position of the bone (specified by the Rest Angle parameters, multiplied by the pre-transform). This new method is more flexible than the previous method of specifying absolute x and y angle constraints.

PolyKnit will branch if the user selects the state with the shift-key held down.

PolyKnit will branch if the user selects the state with the shift-key held down.

PolyKnit will branch if the user selects the state with the shift-key held down.

PolyKnit will branch if the user selects the state with the shift-key held down.

PolyKnit will branch if the user selects the state with the shift-key held down.

PolyKnit state can now detect meta edges and non-meta edges, and provides visual feedback to the user as to which is which. Meta edges are shown in thicker lines. The user has the option to force a non-meta edge by clicking with the middle mouse button on the next point.

PolyKnit state can now detect meta edges and non-meta edges, and provides visual feedback to the user as to which is which. Meta edges are shown in thicker lines. The user has the option to force a non-meta edge by clicking with the middle mouse button on the next point.

PolyKnit state can now detect meta edges and non-meta edges, and provides visual feedback to the user as to which is which. Meta edges are shown in thicker lines. The user has the option to force a non-meta edge by clicking with the middle mouse button on the next point.

PolyKnit state can now detect meta edges and non-meta edges, and provides visual feedback to the user as to which is which. Meta edges are shown in thicker lines. The user has the option to force a non-meta edge by clicking with the middle mouse button on the next point.

PolyKnit state can now detect meta edges and non-meta edges, and provides visual feedback to the user as to which is which. Meta edges are shown in thicker lines. The user has the option to force a non-meta edge by clicking with the middle mouse button on the next point.

The dsreload command will now scan for new operators. This means that when you create a new scripted operator (VEX, Sub-Network), you don't have to re-start Houdini for the change to become active.

The dsreload command will now scan for new operators. This means that when you create a new scripted operator (VEX, Sub-Network), you don't have to re-start Houdini for the change to become active.

The dsreload command will now scan for new operators. This means that when you create a new scripted operator (VEX, Sub-Network), you don't have to re-start Houdini for the change to become active.

The dsreload command will now scan for new operators. This means that when you create a new scripted operator (VEX, Sub-Network), you don't have to re-start Houdini for the change to become active.

The dsreload command will now scan for new operators. This means that when you create a new scripted operator (VEX, Sub-Network), you don't have to re-start Houdini for the change to become active.

VMantra handles compositing of the Af variable differently now. Previously, the compositing was flawed in that opaque/partially opaque surfaces which had the Af variable set to 0 (i.e. a matte surface) would not composite properly in the rendering process. Now, the Of variable controls how opaque the Af variable is. That is, with a surface shader: surface partial_matte() { Cf = 0; Of = .5; Af = 0; } the surface shaded will occlude 50% of the background color and also 50% of the background alpha. Previously, the Af of the background would have shown through 100%.

VMantra handles compositing of the Af variable differently now. Previously, the compositing was flawed in that opaque/partially opaque surfaces which had the Af variable set to 0 (i.e. a matte surface) would not composite properly in the rendering process. Now, the Of variable controls how opaque the Af variable is. That is, with a surface shader: surface partial_matte() { Cf = 0; Of = .5; Af = 0; } the surface shaded will occlude 50% of the background color and also 50% of the background alpha. Previously, the Af of the background would have shown through 100%.

VMantra handles compositing of the Af variable differently now. Previously, the compositing was flawed in that opaque/partially opaque surfaces which had the Af variable set to 0 (i.e. a matte surface) would not composite properly in the rendering process. Now, the Of variable controls how opaque the Af variable is. That is, with a surface shader: surface partial_matte() { Cf = 0; Of = .5; Af = 0; } the surface shaded will occlude 50% of the background color and also 50% of the background alpha. Previously, the Af of the background would have shown through 100%.

VMantra handles compositing of the Af variable differently now. Previously, the compositing was flawed in that opaque/partially opaque surfaces which had the Af variable set to 0 (i.e. a matte surface) would not composite properly in the rendering process. Now, the Of variable controls how opaque the Af variable is. That is, with a surface shader: surface partial_matte() { Cf = 0; Of = .5; Af = 0; } the surface shaded will occlude 50% of the background color and also 50% of the background alpha. Previously, the Af of the background would have shown through 100%.

VMantra handles compositing of the Af variable differently now. Previously, the compositing was flawed in that opaque/partially opaque surfaces which had the Af variable set to 0 (i.e. a matte surface) would not composite properly in the rendering process. Now, the Of variable controls how opaque the Af variable is. That is, with a surface shader: surface partial_matte() { Cf = 0; Of = .5; Af = 0; } the surface shaded will occlude 50% of the background color and also 50% of the background alpha. Previously, the Af of the background would have shown through 100%.

If a scale handle is oriented to the world axes, it will now draw less of itself in an ortho viewport. This will help expose the parts of the handle that make sense for that orthographic projection. The handle will be easier to pick and manipulate and will cover less screen space.

If a scale handle is oriented to the world axes, it will now draw less of itself in an ortho viewport. This will help expose the parts of the handle that make sense for that orthographic projection. The handle will be easier to pick and manipulate and will cover less screen space.

If a scale handle is oriented to the world axes, it will now draw less of itself in an ortho viewport. This will help expose the parts of the handle that make sense for that orthographic projection. The handle will be easier to pick and manipulate and will cover less screen space.

If a scale handle is oriented to the world axes, it will now draw less of itself in an ortho viewport. This will help expose the parts of the handle that make sense for that orthographic projection. The handle will be easier to pick and manipulate and will cover less screen space.