The Power of Dual-GPU: Not Just for Gamers

In today’s ever-evolving world of technology, the concept of multi-GPU for gaming, such as SLI and CrossFire, has become obsolete. However, the potential of dual-GPU remains valuable in other applications, as demonstrated by the FluidX3D demo. Under the ProjectPhysX brand, developer Dr. Moritz Lehmann showcased a configuration that combined an Intel A770 with an Nvidia Titan Xp, a system that is nearing its seventh anniversary. Although this combination may seem peculiar, the results show that the two GPUs together create a potent partnership.

The multi-GPU demo turned out to be surprisingly straightforward. Dr. Lehmann utilized Acer’s Predator A770 16GB and Nvidia’s Titan Xp, with each GPU simulating and rendering half of the simulation. While DX12 and Vulkan are the most popular APIs for multi-GPU operations, FluidX3D runs on OpenCL, a development by Khronos Group, the same creators of Vulkan.

While no precise performance data was provided, Dr. Moritz claims that the dual-GPU setup was 70% more efficient than each GPU individually, which makes sense as the A770 and Titan Xp deliver similar results in FluidX3D. The simulation took one hour and 13 minutes, with the rendering lasting approximately 14 minutes. This implies that each card individually would require around two hours for simulation alone.

As humorous as this combination may seem, it had its rational justification. As emphasized by the developer, combining extremely powerful GPUs with significantly weaker ones makes little sense. In the case of FluidX3D, it is beneficial for the GPUs to have similar memory capacity and bandwidth. The A770 boasts 16GB memory and 560GB/s bandwidth, while the Titan Xp offers 12GB memory and 548GB/s bandwidth. Thus, this combination proves to be a sensible choice.

It’s hard to believe that the gaming industry has abandoned multi-GPU technology, considering the 70% increase in performance showcased during the FluidX3D demo. Commentators during the demo were quick to point out that DX12 and Vulkan have substantial support for this technology, and GPU-to-GPU connection is more reliable than ever, thanks to the latest PCIe versions being exceptionally fast.

In response to these comments, Dr. Moritz presented his analysis and highlighted several issues related to multi-GPU configurations in gaming. The primary concern revolves around the cost associated with developing multi-GPU solutions for games. In the past, this burden fell on Nvidia and AMD, but with the advent of DX12 and Vulkan, the responsibility has shifted to game developers. These APIs offer powerful multi-GPU capabilities but require manual optimization to be effective. However, game developers do not reap the benefits of implementing multi-GPU support, as it has always been a niche feature, even among PC enthusiasts.

Rather than investing in multi-GPU technologies, the industry has been heading in a different direction to achieve greater performance. The focus now lies more on single GPU configurations, creating flagship models that have grown so large that “you can’t even fit one of them in a regular computer case, let alone two,” as Dr. Moritz states. Considering that many GPUs, like the RTX 4090 Founders Edition, occupy three slots, one can easily agree. Currently, multi-GPU technology is thriving in data centers, supercomputers, and AI-focused systems, where the implementation cost is justified.

FAQ:

1. What was the use of multi-GPU technology in the presented demo?
Multi-GPU technology was utilized for simulating and rendering simulations in the FluidX3D program.

2. What graphics cards were used in the dual-GPU setup?
The demo employed the Acer Predator A770 16GB and the Nvidia Titan Xp graphics cards.

3. What API was used in the multi-GPU demo?
The demo utilized the OpenCL API, which is developed by the Khronos Group.

4. What was the estimated performance increase using the dual-GPU setup?
According to the demo’s creator, the dual-GPU setup was 70% more efficient than each GPU individually.

5. What were the main issues associated with multi-GPU configurations in gaming?
The primary issues were the high costs associated with developing multi-GPU solutions for games and the lack of profitability for game developers when implementing multi-GPU support.

6. What are the current trends in the industry regarding GPU performance?
The industry is currently focusing on single GPU configurations, creating larger flagship models instead of employing multi-GPU technology.

Definitions:
– Multi-GPU: A technology that allows the simultaneous use of multiple graphics cards to enhance performance.
– SLI: Scalable Link Interface, a technology developed by Nvidia that enables the connection of two or more graphics cards to improve performance.
– CrossFire: A technology developed by AMD that allows the combination of multiple graphics cards to enhance performance.
– GPU: Graphics Processing Unit, a unit responsible for processing graphics.
– Demo: A demonstration or presentation.
– API: Application Programming Interface, a set of protocols and tools for building software applications.
– DX12: DirectX 12, an appropriate API for application programming in Windows systems.
– Vulkan: An API designed for efficient cross-platform work.
– OpenCL: Open Computing Language, an open standard for parallel programming that can run on various devices.
– Khronos Group: An organization responsible for standardizing multimedia technologies.
– PCIe: Peripheral Component Interconnect Express, a computer bus standard primarily used in computers.
– Dual-GPU: A computer configuration with two graphics cards.
– Supercomputers: High-performance computers capable of executing complex calculations.

Sources:
– Nvidia: https://www.nvidia.com/
– AMD: https://www.amd.com/
– Khronos Group – OpenCL: https://www.khronos.org/opencl/

The source of the article is from the blog myshopsguide.com