Next Generation Rendering
Total Chaos finished a few weeks ago but the big announcements made by Chaos Group at their conference in Sofia, Bulgaria, are still settling.
Announcements that got the crowds excited included the pending release of Corona 2 for 3ds Max and V-Ray Cloud, but the standout topic across the event was the launch of V-Ray Next. Here we explore this major release from Chaos Group, what’s new, and what it could mean for artists all over the globe.
The YellowDog team were there to see the official release of V-Ray Next for 3ds Max at Total Chaos; Maya and other integrations are expected later in the year.
There is no question that this is a major release for Chaos Group: their biggest for a long long time. V-Ray 3 has been out for 4 years and in that time, there have been 6 ‘major’ releases. Every service pack has had major additions and for many software companies, these frequent lightweight updates would have stood on their own feet as a release.
V-Ray Next was the perfect opportunity for Chaos Group, as the man who puts the V in V-Ray (Vlado) puts it in this podcast, to clean the hairs from the code and create a V-Ray that was truly ‘next generation’. At its very heart, this release marks a huge step forward in the use of GPU and AI for scene intelligence. It is as ‘smart’ as it is fast.
Scene intelligence / AI
Adaptive Dome Lights (ADL) and light samples. ADL gives more accurate image-based lighting and it is much faster: the need for light portals is no more. There is no need to adjust light samples. The results are faster to get correct by a factor of 2x according to some reports. These adaptive dome lights and scene intelligence with automatic light sample adjustment will bring huge benefits to interior visualisers in particular.
Smart Camera Settings. The V-Ray Physical camera now has Automatic Exposure and Automatic White Balance. If we were to sum it up, it is like Instagram has come to rendering – getting the right exposure first time is more likely now than it has ever been.
Smart Denoising. In V-Ray 3, Chaos introduced denoising but it was never a real-time solution. Machine learning of data from light passes in a thousand noisy renders and clean renders enabled NVIDIA to develop their OptiX AI-accelerated denoiser. This has now been added to the V-Ray pipeline and a render can be ‘denoised’ in real time. This feature will be great for early look dev and tests and is sure to speed up workflows. As the Optix AI-accelerated denoiser is better for single frames, this will have the greatest benefit for architectural visualisers and designers.
There are some fundamental changes under the hood to the code in V-Ray Next and some cool smart features for the CPU engine but the headline grabbing progression with this release is the GPU improvement.
V-Ray Next GPU totally replaces V-Ray RT.
V-Ray Next GPU claims to be up to 2X faster than RT but more importantly, due to the architecture change and dedicated UI for GPU, the offset of features in favour of speed is no longer such a cause for disgruntlement amongst GPU users as Chief Creative Officer of Chaos Group, Lon Grohs explains:
“We went from a GPU mega kernel to a multi-kernel architecture…
…Now that we have done that we have a 2 x speed increase but we are also able to introduce new features that won’t slow down the code”.
V-Ray Next is a super cool name for a super cool release.
Based on what we’ve seen so far at YellowDog, we think artists will love it.
To talk to an award winning Chaos Group render partner and test how you can speed up your V-Ray renders, get in touch with the YellowDog team today.