The rise of GPU-based rendering in the VFX and CG space is growing ever more prominent.
As an alternative to rendering from the CPU, working with your GPU presents a very tempting option. But is it really worth the switch from the familiar realm of harnessing the power of your CPU?
It’s an important decision, whether you’re a sole creative or part of a larger team. With so much to consider, digesting a little expert insight on the topic can be especially valuable.
The team over at Redshift have spent years developing their GPU-based renderer; so while it’s clear that they favor the approach over its CPU cousin, there’s no denying they’re specialists in its intricacies.
We turned to Redshift to argue the case for GPU rendering. Panos Zompolas is the founder and CTO of Redshift Rendering Technologies, and here he’s highlighted five reasons he believes GPU is the way forward in 2016.
1) Power at speed
In the competitive world of VFX, success can be about more than quality of content alone. Responding rapidly to client demands or avoiding rendering bottlenecks when iterating in the creative process is vital. And there, argues Zompolas, GPU rendering comes into its own.
“GPUs paired with a technology like Redshift’s can render images significantly faster compared to CPU renderers,” he explains.
“Faster rendering means lower hardware costs, especially considering that a single GPU equals the rendering performance of five-to-20 high-end CPUs on average.
“Faster rendering also means artists are able to quickly iterate on their work and without having to wait for hours to visualize the effect of a geometry, light or shader modification. This has a positive effect on their creativity and, subsequently, on the quality of the final result.”
2) Empowering the artist
Sole trader creatives working from home are a bedrock of the VFX and CG space, and according to Zompolas, adopting GPU rendering can allow them to punch above their weight from both a creative and technology perspective.
“GPU rendering allows individual artists to produce higher-quality work,” he explains. “Hard-to-render effects like raytraced depth of field, motion blur, glossy reflections and global illumination can now be easily employed in 3D scenes – and that includes animations.
“An individual artist can install four high-end GPUs onto a single computer and rival the performance of several tens of high-end CPU-based render farm machines. And you get all of this with lower electricity costs and a fraction of the occupied physical space compared to what an equivalent CPU render farm would require.”
3) Staying relevant through shifting consumption habits
The way consumers digest content has changed, and the industry has had to respond. In a time when streaming and on-demand viewing is muscling out terrestrial broadcast and physical releases, clients’ expectations of VFX studios has significantly shifted. Fortunately, argues Zompolas, it’s a change GPU-based rendering is well placed to serve.
“Larger studios often target high-quality visuals and, in order to achieve that with a CPU renderer, they typically employ their own render farms or external cloud rendering solutions,” he states.
“However, there do exist productions whose size and scope – or deadlines – mean that a large CPU/cloud-based render farm still can’t cope or would simply derail the project’s budget.
“As an example, episodic TV or streaming via Netflix and Amazon content can be very demanding in terms of quality and rendering time – due to the sheer length of a typical series – while needing to fit within strict budgets and deadlines. TV and web ads have similar demands.
“Redshift has proven to be an invaluable tool for larger-scale studios producing such types of content. Several of our clients have managed to bid on – and win – projects that would have otherwise been unfeasible due to rendering/hardware costs alone. That’s even after considering cloud rendering solutions.”
4) Growth without growing pains
As a thriving VFX studio grows in headcount and clout, so to can costs across the board. For teams in that situation, where scaling up comes hand-in-hand with growing pains, Zompolas is confident GPU rendering can ease the grief while providing a springboard for expansion.
“Given that a single GPU can rival the performance of several high-end CPUs, a studio would need considerably fewer computers to achieve the same result as a CPU-based render,” he explains. “That directly translates into significantly lower space, power, cooling, administration and software licensing costs. Computers can be easily upgraded by adding more GPUs in them or by swapping old GPUs for newer or higher-end ones.”
5) Tomorrow’s rendering today
The Redshift team, along with many companies currently investing in GPU rendering, are confident that the approach is better future-proofed than the traditional CPU approach, and that, says Zompolas, makes jumping on board in 2016 particularly smart.
“This year we’ll witness the first GPUs made with the latest, state-of-the-art 14nm fabrication technology,” he enthuses. “This will undoubtedly mean a considerable jump in computational performance and a reduction in power consumption.
“Additionally, the new GPUs will contain even larger amounts of memory, which means that larger scenes will be rendered even easier by a GPU renderer. Such technological advances will offer great performance and usability benefits to GPU renderers like Redshift.”
by Chris McMahon