Recent HtoA 5.2.1 made it possible to actually use Arnold6 GPU in Houdini, i tool it on a test ride with particles. (5 million particles). 

the setup is quite simple, pure diffuse shading only AA samples. I've chosen a darker frame to test the sample quality. 

CPU 2 minutes on 6 core Xeon :

GPU 1 minute (Nvidia Quadro RTX5000),  same sample count the CPU render above but more little more noise.

GPU 1:30 minute, increased the sample to get a noise free render.

the Arnold GPU render it getting slowly faster the Arnold CPU in some cases, there is still long road ahead to for the GPU renderer speed to catch up with competition.

birth of a massive star. for my little astrophysics exploration, I've created a star simulation using openCL in Houdini. it's based on nbody physics model, only driven by gravity. using nvidia quadro rtx 5000.

the colours are not quite right, I am trying to integrate kelvin colour temperature.  the current colour is based on density and speed.  

In this simulation, I assumed  48% amount of negative gravity to fill the mystery of dark matter. I used 1 million nbodies for the simulation.  for the steps, I am planning to add finer particle streams to get more details.

older scene.
Houdini openGL Rop  4 sec a frame in 4k and DOF. Arnold and Mantra took both 2-3 minutes to render it in 4k.

3delight is back from the dead. It looks great so far, currently, it can't beat Renderman. plus its beta, I had issues with portrait camera format.

the image below is with 3delight, clock in 4:20 min

test with previs Scene but with Blender realtime Eevee render. 2 million polygons. 10-20 fps on RTX 5000. (live viewport below)

Houdini openGL ROP :

Realtime engines are on the rise! I will keep an eye on this!

I’ve tested RenderMan with spline rendering. it’s much faster compared with baked Geometry. the spline primitives are very good in Renderman.

I’ve stopped the test with mantra after 10 minutes, it is super slow even with optimized render settings.

the Renderman images rendered for 3:20 on a 6 core Xeon
2.7ghz with pxr pathtracer. 4k resolution

geometry:

splines primitives:

rendering splines directly saves a lot disk space and meshing time. interactive rendering was super responsive.

Octane 2019.2 renderings is very fast, I’ve used the geo to render it. 2 minutes, as except its lot faster the CPU renderings. the spline rendering in octane does features custom width yet.

Octane gives physical correct shading out of the box. Renderman the shading is extreme flexible with NPR renderings, which is hard to get in Octane. the IPR responsibility was a little slow with shader tweaks.

i am testing the different integrators of Renderman 23 in Houdini. the limit was 2 minutes render time on a 6 core Xeon, one HDRI dome light. Alembic geo 1.7 GB file.

direct lighting vs Pathtracer (below)

pxr unified vs pxr unified bidirectional (below)

pxr bi directional path guiding vs pxr vcm (below)

the shader settings for pxr unified integrator need to be adjusted to match the look. it took 30 seconds for the first pixel. the interactive responds time is excellent . the clamped highlights comes the buggy Houdini 18.0.287, below some scene with Houdini 18.0.237

Renderman 23 CPU vs Arnold 6 GPU (below)

the reflection differences come from normals, Arnold calculates it’s on normals, Renderman took the normals from the Geometry. i’ve tried with Karma but i had issues and crashes with it.

2 minutes render time, left CPU and right GPU with high poly mesh 1.9GB data. (6 core Xeon and Quadro RTX5000)

in this cache, the GPU was faster and more stable. its response time on shaders changes not too bad. only sampling changes took 3-4 seconds to respond

.Octane was crashing on this scene while loading the geometry. Blender Cycles needed 4 minutes to load the alembics.

the CPU is much more efficient in cleaning up the final noise than the GPU. the render result is line up on both, but sampling is quite different. it would be impossible to mix CPU and GPU renderings.

In the end, the GPU rendering is a little faster the CPU rendering (once you optimized the render setting.) Optix GPU rendering does not support GPU BVH branching only pure AA samples, that’s properly why I am not getting the excepted massive render speedups on the GPU.

Tests in Blender showed CUDA rendering are faster then Optix rendering (with RTX).

I’ve made a simple scene to test the physics of light. for proper light calculation, I used spectral render Indigo and Octane.

Indigo has multiple Engines, standard Spectral Path tracer on CPU or GPU and Bidirectional path tracing with MTL sampling (metropolis Light Transport). Octane has only default Spectral Path tracer on GPU but includes an MTL sampling method. I’v also added Renderman 23 to the test with its unified rendering integrator. It supports bidirectional path, manifold caustics and path guiding on the CPU.  

Another render engine like Arnold or Cycles with regular Path tracing would impractical fro complex light calculation tasks.

The base scene is Sphere and squashed Sphere underneath inside a volume box (uniform VDB).      

The following image is the result of Indigo Renderer with BiDirectional path tracing and MTL. It was by far the fastest rendering.

I've open a Houdini and rendered with RenderMan 22.6.  I used only xrDirectLighting integrator and the render speed was super fast on CPU. I've ditched the Pathertracer for this scene, it gave no visual benefits/improvement in this case. Pure spline rendering, no Polygon meshes.