OctaneRender™ for LightWave™ build 0.10.2 released

[juanjgon]

Hello,

This is a new release of the OctaneRender™ for LightWave™ plugin, build 0.10.2 - released for Lightwave 10.x and 11.x Windows x32 and x64.

Download plugin links:
http://render.otoy.com/customerdownload ... .2_x32.zip
http://render.otoy.com/customerdownload ... .2_x64.zip


How to install the plugin:
- Download the install file, you can see that inside it there is the plugin, the octane.dll and octane.dat files and a custom LW menu bar file
- Copy the octane.dll and octane.dat files into you install Lightwave/bin directory
- Add the plugin from Lightwave Layout, select the plugin for your Lightwave version (10.x or 11.x)
- Install the custom menu bar to get access to the Octane features

How to update the plugin:
- It is like install the plugin again (copy the .dll and .dat and add the plugin and menu bar again), but you must be sure that there aren't any octane master plugins running in the Layout before try to add the plugin again. There is a live master plugin needed to start/stop the engine that must be closed to allow plugin reinstall.


*NEW FEATURES IN THE 0.10.2 RELEASE*
- New 1.11 Octane SDK core libs:
- Improved direct light calculation in SSS materials - as a side effect volumetric noises don't work anymore
- Improved accuracy of intersecting volumes with different IORs
- New multi-threaded mesh compilation with better render performance and lower memory usage during compilation (/voxelization)
- Improved daylight system
- Black edges in glossy materials are fixed
- Surfaces with a matte material hiding all objects behind, but are visible indirectly
- Improved combining of noisy multi-GPU results, which resulted in too dark images in the past
- Added architectural glass as option "fake_shadows" in the specular material node
- New spectral parameters in post processing node
- Wireframe rendering using a new channel inside the InfoChannel kernel node
- New PMC kernel option "parallelism" that reduces the number of samples that are investigated in parallel, to make caustics appear earlier at the expense of some performance
- New Dirt Texture node, inside the Octane Procedural nodes.
- New Panoramic Camera node. Support of Spherical and Cylindrical projections.

- New parameters in the Camera Imager node to set a custom Gamma and Response for the IPR window.
- New plugin core to support multiple camera node types
- Add support of Lightwave vertex normal maps.
- Review the interface of render target nodes using sliders, percents, etc.
- IPR now show the total render time.
- New option in the plugin options panel to stop IPR rendering using the kernel samples settings.



*BUGS FIXED IN THE 0.10.2 RELEASE*
- Some IPR functions (like scrub time line or press the Redraw button) disable the IPR "Close" button. Fixed.
- IPR 2D pan function has the Y axis inverted. Fixed.
- Absorption and Scattering medium nodes have a bug in scale value, who must be set to high percent values (from 1000% to 10000% to match Standalone behavior). Fixed to work in real units. Default scale set to 1.0 meter.
- F9/F10 sometimes don't close the IPR window and freeze Layout. Fixed.
- UVMAP node crash LW if plugged into any Octane node input. Fixed.
- Default caustic blur in path tracing kernels is now 0.01
- The default refraction index in glossy material is now 1.3 to match standalone app.



Enjoy it
-Juanjo

[gordonrobb]

Great work thanks. Render a test right now.

Regarding the accuracy of intersection volumes of different IORs. Is there a definitive answer to how they should be modelled in, for example, liquid in glass. Should the liquid intersect the glass, be touching, or should there be a gap? I mean based on how it is calculating it, and wanting the best (most realistic) look, not how easy/difficult it might be to model.

[juanjgon]

Great work thanks. Render a test right now.

Regarding the accuracy of intersection volumes of different IORs. Is there a definitive answer to how they should be modelled in, for example, liquid in glass. Should the liquid intersect the glass, be touching, or should there be a gap? I mean based on how it is calculating it, and wanting the best (most realistic) look, not how easy/difficult it might be to model.

I am not sure about the glass intersections workflow, perhaps we can search in the forum about this issue, plugin must work like standalone or any other plugin.

-Juanjo

[UnCommonGrafx]

Thanks, Juanjo!

Very nice work, overall. Truly appreciating the Octane Renderer.

[UnCommonGrafx]

- IPR now show the total render time.

I see this. Yet, it's not as useful as having the render time show up after an f9 on the viewer. At this time, it shows up as 0.0 time for the render.

[juanjgon]

- IPR now show the total render time.

I see this. Yet, it's not as useful as having the render time show up after an f9 on the viewer. At this time, it shows up as 0.0 time for the render.

Yes, but problem is that I think that LW SDK doesn't have a function add this render time to the image viewer ... I will ask to LW developers anyway.

-Juanjo

[gordonrobb]

Great work thanks. Render a test right now.

Regarding the accuracy of intersection volumes of different IORs. Is there a definitive answer to how they should be modelled in, for example, liquid in glass. Should the liquid intersect the glass, be touching, or should there be a gap? I mean based on how it is calculating it, and wanting the best (most realistic) look, not how easy/difficult it might be to model.

I am not sure about the glass intersections workflow, perhaps we can search in the forum about this issue, plugin must work like standalone or any other plugin.

-Juanjo

All I've found on the forum is questions without answers. Some people saying intersect, some not. I want to know which is recommended, based on an understanding of how it is working. IN the following example the middle one has liquid touching, the left has it intersecting, the right has an air gap. They are all slightly different.

[juanjgon]

For me the middle one is fine. I will try to make a test scene myself.

-Juanjo

[gordonrobb]

For me the middle one is fine. I will try to make a test scene myself.

-Juanjo

I think they all look fine. I just want them to look accurate if you know what i mean. The middle one has a strange line at the top of the liquid.

I may try to model an accurate glass with liquid, then photo graph and try to match reality.

[FrankPooleFloating]

Yes!!!.. thanx so much Juanjo!!... I'm still giddy.. like a little schoolgirl!

Edit: My GPU's thank you too!.. Being able to set max samples for IPR rocks!
We really needed this - and you actually listened to us - and made this happen!!
It is very encouraging to see you quickly add much needed features... Now teach
Otoy this trick! (j/k)

With max at 100 or so - GPU's barely heat up at all - even while setting up complex
scenes for hours... yes!! (not that I worry about temps since w/c.. but electric bill
will prolly be a couple bucks lower)

Thanx again bro!.. going to be heaping much more praise on you - when I get to
play with other stuff - that I have not had time to yet.