Octane rendering performance across multi-RTX 4090s

Known as the world’s fastest unbiased, spectrally correct GPU render engine, Octane uses GPU at the core to run faster in the rendering process. It should come as no surprise that building up a multi-GPU system does make Octane render more quickly. However, to what extent quickly does it make? Let us answer by digging deeper into Octane rendering performance across multi-RTX 4090s.

As far as Octane is concerned, the GPU is by far the most important component of a computer. As of now, Octane Render on Windows only supports a wide range of current NVIDIA GPUs (2012-present) via CUDA. Octane on macOS supports newer AMD GPUs through Metal, even some Intel GPUs. Octane supports CUDA on the Mac, but macOS discontinued official NVIDIA compatibility after High Sierra, so only older NVIDIA cards will work on Macs running High Sierra or earlier. Since Octane performs the best on a GPU that has CUDA-Cores, we will only talk about NVIDIA GPUs in today’s article. Amongst NVIDIA GPUs, the GeForce RTX 4090 is currently the most powerful for Octane rendering. Why is RTX 4090? Let’s explore how Octane utilizes GPU to find the reason. 

The important parts of a GPU for Octane are the actual speed and the VRAM that comes with it. These two specs define how fast Octane runs, and how much in the way of geometry and textures can be loaded into the GPU’s memory at a time.

VRAM is a special type of high-speed RAM integrated into the graphics card that controls how much texture and scene geometry can be loaded into memory at once for Octane to operate on. A scene that fits neatly into VRAM will be significantly faster and more stable than one that does not. RTX 4090 with a decent amount of VRAM (24GB) can meet almost all small to heavy scenes in Octane.

GeForce GPU has good raw performance and RTX 4090 has the best raw performance amongst NVIDIA Geforce GPUs thanks to its newer technology and higher number of cores.

Powered by the ultra-efficient NVIDIA Ada Lovelace architecture, RTX 4090, with 3rd-generation RT Cores, 4th-generation Tensor Cores, an 8th-generation NVIDIA Dual AV1 Encoder, and 24GB of Micron G6X memory capable of reaching 1TB/s bandwidth, brings a massive boost in rendering performance for Octane. It beats other GeForce 40 Series GPUs and GeForce 30 Series GPUs in CUDA cores, RT cores, and Tensor cores, thus rendering performance. 

Let’s take a look at how RTX 4090 is the successor.

OctaneBench 2020.1.5 (Source: Puget Systems)

The RTX 40 Series performs fantastically in Octane renderer. Amongst those, the RTX 4090 dominates the rendering performance with ~37% faster than RTX 4080 and 76% faster than RTX 4070Ti. Not to mention that RTX 4090’s VRAM is the highest of those three in this 40 Series. 

Compared to its predecessors (the RTX 30 Series), RTX 4090 offers a greater improvement in rendering performance. 

OctaneBench 2020.1.5 (Source: Puget Systems)

Taking into the amount of VRAM available, RTX 4090 24GB is the standout performer. It is 90% faster than RTX 3090 for only a 7% price increase. It even outpaces the much more expensive RTX 3090 Ti by 83%. Without a doubt, for those looking for the best-performing GPU for Octane rendering, the RTX 4090 24GB is hands down the most ideal option.

OctaneRender is well-known for its multi-GPU support. While a single RTX 4090 itself is really fast, how fast do multi-RTX 4090 systems perform in Octane? Let us get to the right point of Octane rendering performance across multi-RTX 4090s.

Let’s together look at this OctaneBench from Otoy to get a good sense of how multi-RTX 4090s perform in Octane. 

The graph below shows the OctaneBench’s score on 1, 2, 3, 4, 6, 7, 8, and 12 of the GeForce RTX 4090 24GB VRAM:

OctaneBench 2020.1 on Multi-RTX 4090s (Source: Otoy)

Take a look at it in another way, the relative performance chart. It demonstrates how adding GPUs (RTX 4090) increases Octane rendering performance – shown as a percentage (%) compared to the speed of a 1x RTX 4090:

Octane rendering performance across multi-RTX 4090s

• • • Going from one card to two: we see an impressively perfect gain in rendering performance, an 101% increase. 
    • From two cards to three: there is a little drop-off when it’s just a 73% increase in performance, but it’s still a great improvement. 
    • From three cards to four: the increase is 110%. A 4x RTX 4090 system offers a 4.1x time faster than a single RTX 4090.
    • At 6x RTX 4090s: The gain is perfect again. Compared to a single RTX 4090, 6x RTX 4090s is 6.09x time faster in rendering. What’s awesome scaling! 
    • From six cards to seven, and seven cards to eight: both have a 88% increase in performance. 
    • At the very peak, with 12x RTX 4090s, we notice exactly a 12x increase in performance over a single GPU. It is perfect, or linear, scaling.

What makes Octane different from other renderers is that it scales almost perfectly with multi-GPUs. There still exists some little levels of diminishing returns as more cards are added to a single system. Nonetheless, they don’t affect much its powerful rendering performance. It’s hard to find any GPU renderers in the CG industry that can scale perfectly 100%, even more than 100% as OctaneRender. 

Octane is increasingly growing with more and more robust features. Artists all want to make more creative projects with complex scenes and thus need super-powerful hardware resources to meet their rendering demands. Sadly, though you have owned a strong GPU such as RTX 3090 or RTX 4090, there are still frames that take tens of minutes, even hours to render. Not to mention, it is just a single frame in a sequence of frames that make up a creative project. The power of a single GPU is not enough at this time. A multi-GPU system, however, is sure to shine in these circumstances because Octane actually scales almost linearly across multiple GPUs.

iRender has recently tested rendering OctaneStudio for C4D 2022.1 on 2x RTX 3090 and 2x RTX 4090 systems. Let’s see how well these two rigs perform and whether is it worth upgrading to multi-RTX 4090s.

• • Software version: OctaneStudio for C4D 2022.1
  • Size: 1920x1080p
  • Frame rate: 30fps
  • Duration: 60 frames

Render time:

• • 2x RTX 3090: 1h54m29s
  • 2x RTX 409: 53m06s

2x RTX 4090 renders this scene more than twice as fast as the 2x RTX 3090. RTX 4090 with newer architecture and core generation than RTX 3090 (despite its core count doesn’t double that of RTX 3090) is twice (even more) as powerful as RTX 3090. As a result, it’s worth upgrading to this multi-RTX 4090 system.

The fact, however, is that a one or two-GPU system is rather simple, and there are lots of options available in a wide price range out there. You can easily get yourself a system of one or two RTX 3090 or RTX 4090. However, to improve the rendering speed, certainly, you need to upgrade the configuration. If you want to double the rendering speed, it’s required to double the number of GPUs for the system or upgrade to another system that has a GPU that is twice as fast. If you want to double the speed of a single RTX 3090 system, you will need a system of 2x RTX 3090s or 1x RTX 4090. Accordingly, doubling the speed of a 2x RTX 3090 system requires a system of 2x RTX 4090s or 4x RTX 3090s. It’s very challenging, though. The reason is that a three or four-GPU system is more complicated when it requires much more consideration for power, cooling, and physically fitting the whole system into a case. More than four GPU system demands highly specialized and expensive kits. Luckily, iRender exists to help you blow away all these challenges by bringing the power of high-configuration machines to your desk. You do not need to worry about the motherboard, chassis size, cooling, and power supply. We take care of the hardware, the creativity is yours!

iRender provides high-configuration single and multi-GPU servers, specifically 1/2/4/6/8x RTX 4090 & RTX 3090. Powered with two of the most single-core performance CPUs for Octane which are AMD Ryzen™ Threadripper™ PRO 3955WX @ 3.9 – 4.2GHz & AMD Ryzen™ Threadripper™ PRO 5975WX @ 3.6 – 4.5GHz, 256GB RAM and 2T NVMe SSD storage, all our servers can work well with almost all levels of complexity of Octane projects.

iRender gives you, all creative individuals and studios, an affordable answer to unleashing your creativity with the beast RTX 4090 from just 8.2 USD/hour. We are proud to be the only render farm where you can install any software and plugins of any version that serves your project. You will be given complete control over the server(s) and use them as your local computers.

Let’s check out our test videos about Octane rendering performance on our multi-RTX 4090 servers.

We are offering a FREE TRIAL program now. Let’s get it and try our RTX 4090 machines for your Octane.

For further information, please do not hesitate to reach us at [email protected] or mobile: +84915875500. 

iRender – Thank you & Happy Rendering!

Reference sources: Otoy, Puget Systems