Out-of-Core Geometry Support – powered by Octane Scene AI

Octane 4 can render fast out-of-core geometry from CPU memory, with minimal speed loss, even across multiple GPUs – a first for an unbiased GPU renderer. All meshes and textures can, if desired, be stored in CPU memory while Octane is rendering at fully interactive speeds with no viewport latency. Scene AI models visibility of surfaces in real time to dynamically move geometry in and out of core for maximum performance.

核心外几何支持–由Octane Scene AI提供支持

Octane 4可以从CPU内存中快速渲染内核外几何图形，即使在多个GPU上，速度损失也最小，这是无偏GPU渲染器的首次。如果需要，所有网格和纹理都可以存储在CPU内存中，而Octane则以完全交互的速度渲染而没有视口延迟。 Scene AI实时建模表面的可见性，以动态地将几何体移入和移出岩心，以实现最佳性能。

Spectral AI Denoiser / (go to the subject)

Octane 4 introduces groundbreaking machine learning technology to interactively denoise beauty and AOV passes on-the-fly, both in the viewport and for final frame production rendering. Octane’s state of the art Spectral AI Denoiser operates on internal perceptual models of material, spectral irradiance and scene data deep in the engine. Domain specific AI denoisers help with glass, refractions, SSS, depth of field and motion blur.

频谱AI去噪器/（转到主题）

Octane 4引入了开创性的机器学习技术，以交互方式使美感降噪，并且AOV在视口中以及最终的帧制作渲染中即时通过。Octane最先进的Spectral AI Denoiser可以根据引擎内部深处的材质，光谱辐照度和场景数据的内部感知模型进行操作。特定领域的AI降噪器可帮助处理玻璃，折射，SSS，景深和运动模糊。

Volumetric AI Denoiser / (go to the subject)

Octane’s Volumetric AI Denoiser removes noise in volumes (and volume passes) that previously required tens of thousands of samples to achieve noise free results.

体积AI去噪器/（转到主题）

Octane的体积AI Denoiser消除了以前需要成千上万个采样才能实现无噪波结果的体积噪波（和体积通过）。

AI Light / (go to the subject)

AI Light further complements the AI Denoiser; it is fully unbiased, and tracks emissive points live, in real-time with up to 6-10x speed improvements for scenes with many small light sources such as multi-point or spot lights.

AI Light /（转到主题）

AI Light进一步补充了AI Denoiser；它完全无偏，并且实时跟踪发射点，对于使用许多小光源（例如多点或聚光灯）的场景，实时提高了6-10倍的速度。

Light Linking and Light Exclusion / (go to the subject)

Light linking and light exclusion in Octane 4 helps artists granularly manage scenes with complex lighting without creating multiple render passes. With light IDs set in emitter nodes, artists can now include and exclude lights by material in object layer nodes, and also globally in the kernel settings, allowing control over the direct influence of emitters on objects

灯光链接和灯光排除/（转到主题）

Octane 4中的灯光链接和灯光排除功能可帮助艺术家精细地管理具有复杂照明的场景，而无需创建多个渲染通道。通过在发射器节点中设置灯光ID，艺术家现在可以在对象层节点中以及材质的全局设置中包括和排除光源中的灯光，从而可以控制发射器对对象的直接影响。

Planetary Environment System / (go to the subject)

The new planetary environment in Octane 4 is based on the Nishita sky model, and simulates the scattering of sunlight in the upper atmosphere and space. It expands physics and lighting conditions to support atmospheric shaders and planet shaders with unbiased path-traced accuracy.

行星环境系统/（转到主题）

Octane 4中的新行星环境基于Nishita天空模型，并模拟了高层大气和太空中阳光的散射。它扩展了物理和光照条件，以无偏路径精确度支持大气着色器和行星着色器。

Universal Material / (go to the subject)

With Universal Materials in Octane 4, the process of combining different materials is dramatically simplified, enabling artists to use a single ‘uber’ material to achieve the complexity of what previously required multiple types of material mixes – such as diffuse, specular, glossy and metallic. Universal Material incorporates multiple BXDF lobes including: Specular Transmission, Diffuse BRDF, Specular Reflection, Coating, and Sheen.

通用材质/（转到主题）

借助Octane 4中的通用材质，可以大大简化组合不同材质的过程，从而使艺术家能够使用一种单一的“超级”材质来实现以前需要多种类型的材质混合的复杂性，例如漫射，镜面，光泽和金属。通用材质包含多个BXDF瓣，包括：镜面透射，漫反射BRDF，镜面反射，镀膜和光泽。

Native Substance PBR Texture Integration / (go to the subject)

OctaneRender 4 now provides Substance PBR texture integration in DCC tools such as C4D and 3ds Max.

原生物质PBR纹理整合/（转到主题）

OctaneRender 4现在在DCC工具（例如C4D和3ds Max）中提供了Substance PBR纹理集成。

UDIM Textures

For production texture workflows, OctaneRender 4 supports UDIM textures through a new image tile texture node enabling artists to set up custom tile grids.

UDIM纹理

对于生产纹理工作流程，OctaneRender 4通过新的图像图块纹理节点支持UDIM纹理，从而使美术师可以设置自定义图块网格。

8:1 GPU Texture Compression / (go to the subject)

Dynamically compressed textures in Octane 4 can use 1/3 – 1/8 the VRAM of normal RGBA, grayscale and float textures. Compressed textures incur zero speed loss, and are nearly identical in visual quality to uncompressed textures.

8：1 GPU纹理压缩/（转到主题）

Octane 4中的动态压缩纹理可以使用普通RGBA，灰度和浮点纹理的VRAM的1/3 – 1/8。压缩纹理导致零速度损失，并且视觉质量与未压缩纹理几乎相同。

IOR texture input for Materials / (go to the subject)

Material IOR in the base layer of Universal Materials can be controlled procedurally with Texture nodes and OSL shaders connected to a new 1/IOR texture input pin.

材质/的IOR纹理输入（转到主题）

可以通过连接到新的1 / IOR纹理输入引脚的Texture节点和OSL着色器，以程序方式控制Universal Materials基础层中的Material IOR。