Tyler Bay
tbay312
About Me
Tyler Bay is the owner of cgforge.com - which is an online Houdini school that offers high quality courses, resources, and individual mentorships to help you achieve your Houdini goals.
EXPERTISE
CG Supervisor
INDUSTRY
Education
Houdini Skills
Availability
Not Specified
My Badges
Houdini Certified Instructor
2025
My Tutorials
Recent Forum Posts
Applying Normal Maps (texture) in Material X... Sept. 21, 2026, 9:55 p.m.
From what I can tell, normal maps are not implemented properly in general because the conversion from tangent --> object spaces causes the signal to become dependent on the mesh resolution. This is true for regular 'ol texture maps and hex tiled maps.
Additionally, SideFX did not keep the parameters consistent with the Hextiled Triplanar and the Hextiled Normal Map nodes, so it can become difficult to line up the two signals properly if you intend on introducing random rotations with the hextiles.
Additionally, SideFX did not keep the parameters consistent with the Hextiled Triplanar and the Hextiled Normal Map nodes, so it can become difficult to line up the two signals properly if you intend on introducing random rotations with the hextiles.
Question Regarding Pyro July 23, 2026, 6:38 p.m.
Good to know! Thank you for pointing that out Tamte.
Question Regarding Pyro July 23, 2026, 1:40 p.m.
tamtetbay312
From my understanding, the dense pyro solver was also the primary solver that used OpenCL. So, with this depreciation, which solver is going to be primary option to use with GPU acceleration? Sparse isn't compatible at the moment. And, from what I can tell, that leaves the minimal solver and COPs as the other options.
Pyro Solver SOP, while having 3 solve modes: Sparse, Dense (OCL on/off), Minimal, all of them are handled by Pyro Solver (Sparse) DOP
in other words Smoke Object (Sparse) and Smoke/Pyro Solver (Sparse) DOPs have Enable Sparse Solving checkbox, which when it's Off will be able to use OCL as it will solve in dense mode
so while it's a bit different from the old dense Pyro Solver DOP and some useful features may be missing, it's understandable that keeping 2 very similar dense solvers may seem redundant
Ahh I see. So based on that, the dense mode + OpenCL is not being removed because it's just integrated into the Sparse solver?