OctaneRender® 1.025 b2.55b test release [obsolete]

A forum where development builds are posted for testing by the community.
Forum rules
NOTE: The software in this forum is not %100 reliable, they are development builds and are meant for testing by experienced octane users. If you are a new octane user, we recommend to use the current stable release from the 'Commercial Product News & Releases' forum.
Post Reply
User avatar
roeland
OctaneRender Team
Posts: 1823
Joined: Wed Mar 09, 2011 10:09 pm

Hi,

The OctaneRender Team is please to announce the betas 2.55b test release.

NOTE: Due to many internal changes, you need to resave your default scene when you first run either beta2.55 or later. When it's started, Click on the top left File menu, then 'Save as Default'. If you do not do this, you might have a different tonemapping response curve or other incorrect defaults.

Fixes since beta 2.55
  • Mac OS X and Linux versions load meshes correctly.
  • Fixed rendering errors when an invalid type of node is attached to the normal map input pin.
  • Resolved black patches on glossy materials.
  • Medium nodes stay attached when swapping between diffuse and specular material
  • After swapping a node in the node inspector the component for that node no longer dissapears if it has the same size as the old.
  • Crash if a macro input node is connected to the diffuse transmission pin.

New features since beta 2.52

There have been a lot of changes since beta 2.52, these are the most important ones:
  • We have made a few large scale changes in how geometry is stored in the engine, which are a part of our in progress rewrite of OctaneRender's geometry system to allow for instancing and object transform editing, coming in a next release soon.
    These are invisible to the end-user, but could produce errors or artifacts on geometry and changes in render speed and memory use for a given amount of geometry.
    Although we have done numerous tests, we specifically like to see what the community thinks and if the speed is better or worse (this could be dependent of the type of scene/geometry),
    the differences in stored geometry size on the GPU, and possible artifacts on geometry rendered.
  • A new node medium node type that can be connected to transmitting materials, being the diffuse material (for diffuse transmission) and the specular material.
    There are 2 medium types provided in this release:
    • absorption: this is a medium with absorption using beer's law.
      This was present in earlier releases as 2 parameters on the specular material, but these are now moved into a medium type node.
      NOTE: We have changed the way the absorption node handles the colour to be absorbed.
      It now works in a substractive way (as do most other unbiased engines), so you need to configure the inverted colour instead to get an absorption colour desired.
      Note that this is not backwards compatible, any OCS projects saved with OctaneRender 2.52 or earlier with absorption configured on a specular material using the old system,
      will lose the absorption colour, and it will have to be reconfigured with the new medium node system instead.

      Absorption (and SSS) now also allows the use of 3D procedural textures to define the absorption for true volumetric 3D procedural media.
    • scattering: This is a medium with single-scattering SSS, and also absorption.
      You can use it to create true unbiased SSS (although not multiple-scattering yet), using various parameters, including scattering texture, emission texture and various other parameters.
      Note that this is single-scattering SSS, not multiple-scattering. It's much faster than the latter, and much more practical, although it does not allow a few things such as volumetric caustics.
      We are planning to add a multiple-scattering SSS medium node in the near future, but we think the single-scattering model is much better for all practical scenes.
    Both absorption and SSS medium nodes should work with all integrators, directlighting, pathtracing and PMC.
    NOTE: To use a medium for SSS or absorption on a diffuse transmission material, you must first connect a value or colour to your diffuse material's transmission input pin, to turn it into a diffuse transmitter. See this tutorial for more information.
  • IOR (Index of reflection) input on the glossy material.
    OctaneRender 2.52 and before provided a slider called 'specular' to define the specular reflection amount on a glossy material.
    We have now added a new parameter called 'index' which allows you to use an IOR (index of reflection) value instead, to accurately define the glossyness of a material from a reference measured value.

    When the value is 0, the glossy material behaves exactly like before and you load existing scenes and make glossy materials using the familiar specular colour method. To keep backwards
    compatibility, this is the default value.

    If you want to use the IOR system to define specular reflection instead, simply set the specular slider to 1.0 (maximum) and then set a value for the index parameter (usually something between 1.0 and 3.0)
NOTE: The material ball shadowing is slightly wrong, this will be rectified in the next release...
As this release does not include instancing support yet (which was meant for 2.6), we decided to version it as 2.55, being somewhere in the middle :)


Driver Requirements

This release was built with CUDA 4.0, which means that you should install the latest graphics driver (http://www.nvidia.com/page/drivers.html).
For Linux and Mac OS you also need to install the CUDA Toolkit 4.0 (http://developer.nvidia.com/cuda-toolkit-40).
For Mac OS you also need the latest CUDA 4.0 driver (http://www.nvidia.com/object/mac-driver-archive.html).


Bug fixes & Minor new features
  • Corruption of macro input nodes that are linked to the wrong input pin after deleting a macro input node.
  • Embedding of HDR images in OCM files.
  • Crash with bool inputs for macro nodes.
  • Fixed the problem of saving the same frame twice with turntable / daylight animation.

Known issues
  • On Linux the PMC kernel doesn't work on pre-Fermi GPUs (i.e compute model 1.1 - 1.3) - we couldn't find the problem yet, which seems to be a compiler bug.

Download

Due to this being a Test Release, we are only providing zip/tgz archives, no installers.

Windows 64bit: http://www.refractivesoftware.com/rcdow ... in_x64.zip
Windows 32bit: http://www.refractivesoftware.com/rcdow ... _win32.zip

MacOS X 32bit: http://www.refractivesoftware.com/rcdow ... osx_32.zip
GNU/Linux 64bit: http://www.refractivesoftware.com/rcdow ... ux_x64.tgz

Yours,
The OctaneRender® Team.
Zay
Licensed Customer
Posts: 1123
Joined: Sun Jan 17, 2010 2:53 am

Smoothing still doesn't work for me. I have to add it manually.
This only happens in 2.55 and 2.55b.
Attachments
Untitled.zip
(26.52 KiB) Downloaded 174 times
Win 11 Pro | i5 12600K | 32GB ram | 2x GTX 1080Ti + 3080Ti - studio driver 560.94| Modo/Blender/ZBrush/Daz/Poser
User avatar
pixelrush
Licensed Customer
Posts: 1618
Joined: Mon Jan 11, 2010 7:11 pm
Location: Nelson, New Zealand

Wow I visited here several times this afternoon and didn't notice this was available :roll:
Have to laugh at the numbering scheme - ' 1.025 b2.55b' :lol:
anyway will try it out and see how it is now :geek:
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
User avatar
pixelrush
Licensed Customer
Posts: 1618
Joined: Mon Jan 11, 2010 7:11 pm
Location: Nelson, New Zealand

ok here is a screenie (see 2nd pic) of another 'fail' as per my post for 2.55.
I hadnt noticed if it was like this last time or not cos I assumed in a quick glance it was using the full 2gb but perhaps I was mistaken and it is like this instance where I have filled what was available. weird, should be 1982 not 188mb.
My Quadro 570 reports it has 200mb useable memory without a scene. I wonder if somehow Octane ended up swapping render cards?? very odd....
EDIT: well perhaps not - the scene uses 188mb when it is first loaded see 'correct' 1st pic.

I think this fail comes about when messing with a spec sss having emission and some other setting in combination with pmc and sunlight but will check some more and see if I can narrow it down. :geek:
Attachments
initial vram use
initial vram use
failed vram use
failed vram use
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
User avatar
Refracty
Licensed Customer
Posts: 1599
Joined: Wed Dec 01, 2010 6:42 pm
Location: 3D-Visualisierung Köln
Contact:

I had some strange response problems after updating my Nvidia driver. Try an old Cuda4 driver.
User avatar
pixelrush
Licensed Customer
Posts: 1618
Joined: Mon Jan 11, 2010 7:11 pm
Location: Nelson, New Zealand

Yeah it could be a driver glitch I suppose. 275.36 I have is from June. First cuda 4 is from April.
I hadnt come across this before 2.55/2.55b though and someone else said they had seen it.
Unfortunately I cant reproduce the issue now by playing with settings... :evil:
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
ribrahomedesign
Licensed Customer
Posts: 415
Joined: Fri Mar 05, 2010 9:32 am

hi there.
view issues i came across :
I believe the black renderscreen at the beginning is a driver issue i have ( 285.62)
after about 1/2 a hr rendering lines appear all across the screen .will provide screenshots soon
turbulence bump does not show up in reflection
sunshine does not go trough glass with PMC and Pathtr.
had a view gpu failures (memory bar filled up suddenly)
thanks
Rico
Windows 7 , 64 b / GTX 590 / Archicad 15 , 64 b / Cinema 4D R 13 studio , 64b /Intel(R)Core(TM)Extreme3,20Ghz /always latest Octane.
PeterCGS
Licensed Customer
Posts: 437
Joined: Tue Apr 13, 2010 2:36 pm

I got issues with 2.55 in general that the blackbody emitter is behaving oddly.. I havent done any scientific tests yet but I got this scene that I tried to have a warm emitter (below 3500K). But the warmer it gets the less light it emitts.. I don't get exactly the same issue with build 2.52 (it also emitts less light the warmer it gets) because there is still light enough to feel it. But in 2.55 it is completely gone when moving down to 2500K etc.. Can anyone confirm this?
Win7, i7 980 Hexacore, 24GB RAM, 3xEVGA SC 670 GTX 4GB
User avatar
roeland
OctaneRender Team
Posts: 1823
Joined: Wed Mar 09, 2011 10:09 pm

PeterCGS wrote:I got issues with 2.55 in general that the blackbody emitter is behaving oddly.. I havent done any scientific tests yet but I got this scene that I tried to have a warm emitter (below 3500K). But the warmer it gets the less light it emitts.. I don't get exactly the same issue with build 2.52 (it also emitts less light the warmer it gets) because there is still light enough to feel it. But in 2.55 it is completely gone when moving down to 2500K etc.. Can anyone confirm this?
A blackbody emitter has a "normalize" input, which is probably turned off. If enabled the emitter has a constant brightness independent of the temperature. If turned off, then the brightness will increase with increasing temperature.

What values are you using?

--
Roeland
Post Reply

Return to “Development Build Releases”