NVIDIA Advises Developers: Optimize Game Performance by Localizing CPU Threads

nVidia Advises Developers: Optimize Game Performance by Localizing CPU Threads

NVIDIA, the leading graphics processing unit (GPU) manufacturer, has recently provided valuable advice to game developers on how to optimize their game code for improved performance. The key recommendation is to focus on reducing the number of CPU threads utilized in order to maximize efficiency and minimize complexity.

While modern desktop PCs can support up to 32 threads, it is important to note that not all threads are created equal. A physical thread, also known as a core, differs from a logical thread generated through hyperthreading. Additionally, Intel’s hybrid processors feature P-Cores that are significantly more powerful than the accompanying low-power E-Cores.

Interestingly, many CPU-bound games actually experience a decline in performance when the core count exceeds a certain threshold. This means that the benefits of additional threading parallelism are outweighed by the associated overhead. By limiting the code to fewer CPU threads, game developers can achieve optimal performance levels and reduce complexity.

In fact, high-end desktop systems with more than eight physical cores have shown performance gains of up to 15% when the thread count of their worker pools is reduced to be lower than the core count of the CPU. However, it is important to acknowledge that the reasons behind these performance fluctuations are complex and varied. While one game may experience a 10% performance drop, another may see a 10% gain on the same system. This highlights the challenge of providing a universal solution across all titles and systems.

NVIDIA further suggests that high-end CPUs with more than eight cores, such as the Ryzen 9 7950X and Core i9-13900K, can render certain games faster by utilizing only half of their thread counts. This is particularly true for titles that are highly sensitive to memory/cache Latency. By limiting the threads to a single core cluster (CCD or P-cores), critical game data can be kept within the same space, avoiding the need to assign a “far-away” core to the same resource.

Another factor to consider is the boost clock of high-end CPUs with 16 or more cores. These CPUs have both a single-core boost clock and a multi-core boost clock. As more cores are loaded, the boost clock tends to decrease to manage power consumption. Consequently, reducing the number of threads can lead to improved boost clocks. However, it is important to note that this approach may not be effective for games that are compute-bound rather than memory/cache-bound. The outcome will vary depending on the complexity of the workload.

Several hardware and software factors contribute to the performance impact of thread count reduction. Higher-core-count CPUs often have lower CPU speeds, and reducing the number of threads can allow active cores to boost their frequency. Decreasing the thread count can also alleviate pressure on the memory subsystem, reducing latency and enhancing CPU cache efficiency. Software resource contention, caused by locks and atomics accessed by multiple threads concurrently, can be mitigated by reducing thread count. Additionally, operating system scheduling issues, such as high context switch numbers and core parking algorithms, can be addressed by optimizing thread allocation.

Hyper-threading, or Simultaneous Multi-threading (SMT), can sometimes lead to performance degradation as two threads compete for resources. This becomes particularly problematic when one of the threads is critical for latency-sensitive tasks. By scheduling a second thread, resources available to the primary thread are reduced, resulting in slower performance for the critical thread. Game developers should carefully consider these implications when optimizing their code.

The introduction of hybrid cores further complicates code optimization. To avoid unnecessary stutters or lags, it is recommended to allocate most game threads to P-cores while pushing system and OS threads to E-cores. Limiting thread usage to less than the physical core count ensures that game threads primarily utilize P-cores, maximizing performance.

In conclusion, NVIDIA’s advice to game developers emphasizes the importance of optimizing CPU thread usage for enhanced game performance. By reducing the number of threads and considering factors such as memory/cache latency, hardware resource contention, software resource contention, OS scheduling, power management, and the impact of hyper-threading, developers can unlock the full potential of high-end CPUs and deliver an optimized gaming experience to players.

Background Information

---

About Intel: Intel Corporation, a global technology leader, is for its semiconductor innovations that power computing and communication devices worldwide. As a pioneer in microprocessor technology, Intel has left an indelible mark on the evolution of computing with its processors that drive everything from PCs to data centers and beyond. With a history of advancements, Intel's relentless pursuit of innovation continues to shape the digital landscape, offering solutions that empower businesses and individuals to achieve new levels of productivity and connectivity.

  

---

About nVidia: NVIDIA has firmly established itself as a leader in the realm of client computing, continuously pushing the boundaries of innovation in graphics and AI technologies. With a deep commitment to enhancing user experiences, NVIDIA's client computing business focuses on delivering solutions that power everything from gaming and creative workloads to enterprise applications. for its GeForce graphics cards, the company has redefined high-performance gaming, setting industry standards for realistic visuals, fluid frame rates, and immersive experiences. Complementing its gaming expertise, NVIDIA's Quadro and NVIDIA RTX graphics cards cater to professionals in design, content creation, and scientific fields, enabling real-time ray tracing and AI-driven workflows that elevate productivity and creativity to unprecedented heights. By seamlessly integrating graphics, AI, and software, NVIDIA continues to shape the landscape of client computing, fostering innovation and immersive interactions in a rapidly evolving digital world.

  

Technology Explained

---

CPU: The Central Processing Unit (CPU) is the brain of a computer, responsible for executing instructions and performing calculations. It is the most important component of a computer system, as it is responsible for controlling all other components. CPUs are used in a wide range of applications, from desktop computers to mobile devices, gaming consoles, and even supercomputers. CPUs are used to process data, execute instructions, and control the flow of information within a computer system. They are also used to control the input and output of data, as well as to store and retrieve data from memory. CPUs are essential for the functioning of any computer system, and their applications in the computer industry are vast.

---

E-Cores: E-Cores (Efficiency Cores) are a type of technology used in the computer industry to provide a more efficient and reliable way of powering and cooling computer components. They are made up of a combination of copper and aluminum, and are designed to be more efficient than traditional copper cores. E-Cores are used in a variety of applications, such as in CPUs, GPUs, and other computer components. They are also used in servers, laptops, and other electronic devices. The technology is designed to reduce heat and power consumption, while also providing a more reliable and efficient way of powering and cooling computer components.

---

GPU: GPU stands for Graphics Processing Unit and is a specialized type of processor designed to handle graphics-intensive tasks. It is used in the computer industry to render images, videos, and 3D graphics. GPUs are used in gaming consoles, PCs, and mobile devices to provide a smooth and immersive gaming experience. They are also used in the medical field to create 3D models of organs and tissues, and in the automotive industry to create virtual prototypes of cars. GPUs are also used in the field of artificial intelligence to process large amounts of data and create complex models. GPUs are becoming increasingly important in the computer industry as they are able to process large amounts of data quickly and efficiently.

---

Latency: Technology latency is the time it takes for a computer system to respond to a request. It is an important factor in the performance of computer systems, as it affects the speed and efficiency of data processing. In the computer industry, latency is a major factor in the performance of computer networks, storage systems, and other computer systems. Low latency is essential for applications that require fast response times, such as online gaming, streaming media, and real-time data processing. High latency can cause delays in data processing, resulting in slow response times and poor performance. To reduce latency, computer systems use various techniques such as caching, load balancing, and parallel processing. By reducing latency, computer systems can provide faster response times and improved performance.

---

P-Cores: P-Cores (Performance Cores) are a type of processor technology developed by Intel that is designed to improve the performance of computer systems. This technology is based on the concept of multi-core processors, which are processors that contain multiple cores or processing units. P-Cores are designed to increase the speed and efficiency of computer systems by allowing multiple cores to work together in parallel. This technology is used in a variety of applications, including gaming, video editing, and data analysis. P-Cores are also used in servers and other high-performance computing systems. The technology is also used in mobile devices, such as smartphones and tablets, to improve battery life and performance. P-Cores are an important part of the computer industry, as they allow for faster and more efficient computing.

---

SMT: Simultaneous multithreading (SMT) is a technology that allows a CPU core to process two tasks (threads) simultaneously. It is crucial to the swift operation of modern-day CPUs. SMT is AMD’s brand of multithreading, while Hyperthreading is Intel’s