OctaneRender™ Standalone 3.08 TEST 4

[justix]

I have a problem with the Safe Frame that do not match the Output render..any clue?

[aoktar]

In Cycles I can see, which miracle is denoiser.
Denoiser is huge speed improvements, no tens % but hundreds %.
Therefore first thing i am looking for in new releases is denoiser ...
Second thing is suport for AMD threadripper and AMD Vega ...

Thanks

always curious why people don't talk about new features rather than what's missing on every new releases. OSL is very big step.

[FrankPooleFloating]

And I believe that they said OSL is the foundation for many of the features to come... Heck, maybe Octane's Denoiser will use OSL some?... Prolly not. I dunno...

Holy crap I almost needed to pinch myself when I saw the windows/fake rooms trick! omfg! Mind blown!

Great DBR! Thanks G, Marcus, Roeland, et al.. You guys rock!

[mirobimbo]

In Cycles I can see, which miracle is denoiser.
Denoiser is huge speed improvements, no tens % but hundreds %.
Therefore first thing i am looking for in new releases is denoiser ...
Second thing is suport for AMD threadripper and AMD Vega ...

Thanks

always curious why people don't talk about new features rather than what's missing on every new releases. OSL is very big step.

OpenCL AMD was announced in march 2015.
https://home.otoy.com/otoy-unveils-octa ... -renderer/

[Goldorak]

In Cycles I can see, which miracle is denoiser.
Denoiser is huge speed improvements, no tens % but hundreds %.
Therefore first thing i am looking for in new releases is denoiser ...
Second thing is suport for AMD threadripper and AMD Vega ...

Thanks

always curious why people don't talk about new features rather than what's missing on every new releases. OSL is very big step.

OpenCL AMD was announced in march 2015.
https://home.otoy.com/otoy-unveils-octa ... -renderer/

That was before Apple deprecated OpenCL (which they decided to replace with Metal), which killed industry wide support for OpenCL (even Khronos is done with it - they’re folding it into Vulkan at some point). With no future for OpenCL, we therefore ported CUDA to AMD and got that and Octane AMD working in March 2016. To this day no one else, including AMD, has built a working cross compiler.

Even then, we had to wait 13 months until July 2017 before AMD could deliver us a stable public windows driver we could ship on. When they did, Octane on AMD was featured at their siggraph event. Now it’s in testing as we work on perf. Just want to point out this work was done.

https://twitter.com/otoy/status/915961890022535168

[abstrax]

Nvidia 635m 2GB intel i7QM 6gb SoDIMM DisplayDriver 385.41

Started logging on 03.11.17 19:23:41

OctaneRender 3.08 TEST 4 (3080005)

CUDA error 209 on device 0: no kernel image is available for execution on the device
-> failed to link file
device 0: failed to load OSL kernels:

Thanks for the report. We will look into it and try to fix it with the next release.

[abstrax]

I found a different behavior from 3.05 and 3.08 and the problem also occurs in 3.07, I placed a mesh emitter inside a specular container (bulb lamp) and it works as expected in 3.05 and another outside but they are both visible since 3.07 as reflection, though the opacity=0, double side unchecked.

Edit : Sorry, the problem occurs only if Transparent emission is activated but if not the illumination is poor compared to 3.05

3.07 and 3.08

Capture2.JPG

3.05

Capture1.JPG

Could you send me the scene or a similar scene showing the problem? We will then investigate. We made some fundamental changes in the emitters to fix some longstanding issues which is probably the reason for the difference, but I would like to make sure that we didn't break anything.

[Jolbertoquini]

Awesome Stuff Guys!

thank you very much for the OSL , I have a quick question https://twitter.com/julesurbach/status/ ... 29?lang=en on This post from Jules it was working the vray flakes? and now when I try I have compilation failed? I have others OSL is working most so far, I'm curious about the flakes?

Cheers,
JO

[FrankPooleFloating]

I was just reading this the other day: http://www.cgchannel.com/2017/11/innobr ... altus-1-5/

innoBright has released Altus 1.9, overhauling the software’s GUI. The new full-screen design, shown above, is intended to provide more visual feedback on the denoising process. ... blah blah blah ... Under the hood, the software now runs on the GPU by default, and via OpenCL only. The old CUDA version is no longer publicly available, with innoBright describing OpenCL as “faster and more reliable”.

And I thought I remembered Goldorak saying something about OpenCL being a turd, and that it was being abandoned etc... Wonder what is up with Altus and OpenCL...

[wallace]

Awesome Stuff Guys!

thank you very much for the OSL , I have a quick question https://twitter.com/julesurbach/status/ ... 29?lang=en on This post from Jules it was working the vray flakes? and now when I try I have compilation failed? I have others OSL is working most so far, I'm curious about the flakes?

Cheers,
JO

Hi,

We don't currently support multiple outputs for OSL texture, so if you remove the alpha output variable from the shader code, then it will compile.

Thanks,
Wallace

Code: Select all

shader
flakes
(
    float flake_scale = 10.0 [[ string description = "Smaller values zoom into the flake map, larger values zoom out" ]],
    float flake_size = 0.5 [[ string description = "Relative size of the flakes" ]],
    float flake_size_variance = 0.0 [[ string description = "0.0 makes all flakes the same size, 1.0 assigns random size between 0 and the given flake size" ]],
    float flake_normal_orientation = 0.0 [[ string description = "Blend between the flake normals (0.0) and the surface normal (1.0)" ]],
     
    output color result = 1.0
)
{
    float safe_flake_size_variance = clamp(flake_size_variance, 0.1, 1.0);
    vector cellCenters[9] = {
        vector( 0.5,  0.5,  0.0),
        vector( 1.5,  0.5,  0.0),
        vector( 1.5,  1.5,  0.0),
        vector( 0.5,  1.5,  0.0),
        vector(-0.5,  1.5,  0.0),
        vector(-0.5,  0.5,  0.0),
        vector(-0.5, -0.5,  0.0),
        vector( 0.5, -0.5,  0.0),
        vector( 1.5, -0.5,  0.0)
    };
     
    point position = vector(u, v, 0.0);
    position = flake_scale * position;
     
    point base = floor(position);
     
    point nearestCell = point(0.0, 0.0, 1.0);
    int nearestCellIndex = -1;
    for(int cellIndex = 0; cellIndex < 9; ++cellIndex)   {
        point cellCenter = base + cellCenters[cellIndex];
         
        vector centerOffset = cellnoise(cellCenter) * 2.0 - 1.0;
        centerOffset[2] *= safe_flake_size_variance;
        centerOffset = normalize(centerOffset);
         
        cellCenter += 0.5 * centerOffset;
        float cellDistance = distance(position, cellCenter);
        if(cellDistance < flake_size && cellCenter[2] < nearestCell[2]) {
            nearestCell = cellCenter;
            nearestCellIndex = cellIndex;
        }
    }
     
    result = color(0.5, 0.5, 1.0);
     
    if (nearestCellIndex != -1) {
        vector randomNormal = cellnoise(base + cellCenters[nearestCellIndex] + vector(0.0, 0.0, 1.5));
        randomNormal = 2.0 * randomNormal - 1.0;
        randomNormal = faceforward(randomNormal, I, randomNormal);
        randomNormal = normalize(mix(randomNormal, vector(0.0, 0.0, 1.0), flake_normal_orientation));
         
        result = color(0.5*randomNormal[0]+0.5, 0.5*randomNormal[1]+0.5, randomNormal[2]);
    }
}