Leaked: NVIDIA GeForce RTX 50 “GB202” GPU to Boast 192 SMs and 512-bit Memory Bus

GB202 to feature 192 Streaming Multiprocessors

The latest rumors from trusted nVidia leaker, Kopite7kimi, shed light on the upcoming Blackwell flagship gaming GPU. According to Kopite, the GB100 Blackwell data-center GPU will employ 10 TPCs (Texture Processing Clusters) within each of its 8 SMs (Streaming Multiprocessors). This suggests a potential increase in the total core count compared to GH100.

But what’s even more interesting is the hinted configuration of the GB202 GPU, which is said to feature 12 SMs with 8 TPCs each, resulting in a total of 192 SMs. It’s important to note that while the GB1XX GPU family is designed for data centers, the GB2XX series is specifically intended for gaming and potentially workstation applications, promising an enhanced gaming experience from NVIDIA.

Unfortunately, details about the design of Tensor and RT Cores remain undisclosed at this point. However, the 192 SM setup in the GB202 GPU translates to a significant increase of 24,576 CUDA cores compared to the current flagship RTX 40 series. This represents a noteworthy 1/3 increase in core count.

Kopite also reaffirms earlier rumors about the inclusion of a 512-bit memory bus in the GB202 GPU. It’s worth noting that this configuration may not be utilized in the RTX 5090. If NVIDIA adopts GDDR7 memory technology for the RTX 50 series, it would result in a considerable boost in memory bandwidth.

While NVIDIA’s roadmap initially indicated a 2025 launch for the successor to Ada Lovelace, there are rumors suggesting it may arrive sooner. The exact configuration of the RTX 5090 GPU remains unknown at this stage, and there may still be variants under consideration. However, given Kopite’s track record and confidence, it seems NVIDIA is finalizing the specifications for its next-generation flagship gaming GPU.

Source: kopite7kimi

Background Information

---

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 cutting-edge solutions that power everything from gaming and creative workloads to enterprise applications. Renowned 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 prowess, 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

---

Ada Lovelace: NVIDIA Ada Lovelace architecture is a new GPU architecture that delivers incredible performance for gaming, creating, professional graphics, AI, and compute. It features third-generation RT Cores, fourth-generation Tensor cores, and next-generation CUDA cores that enable revolutionary performance for ray tracing and AI-based neural graphics. It also supports new rendering features such as DLSS 3.0, RTX Remix, shader execution reordering, displaced micro-meshes, opacity micromaps, and Nvidia Reflex43.

---

GDDR7: GDDR7 (Graphics Double Data Rate 7) is the seventh generation of graphics double data rate (GDDR) memory. It is a type of dynamic random-access memory (DRAM) that is specifically designed for use in graphics cards. GDDR7 memory offers a number of advantages over previous generations of GDDR memory. GDDR7 is a significant improvement over previous generations of GDDR memory. It offers faster speeds up to 32 gigabits per second (Gbps) per pin, lower power consumption, and improved error correction. This makes it ideal for use in high-performance graphics cards and other applications that require high bandwidth and low latency.

---

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.

---

RT Cores: RT Cores are a type of processor core technology that is designed to provide real-time computing capabilities. This technology is used in a variety of applications in the computer industry, such as embedded systems, robotics, and industrial automation. RT Cores are designed to provide fast and reliable computing performance, allowing for the execution of complex tasks in a short amount of time. They are also capable of handling multiple tasks simultaneously, making them ideal for applications that require high levels of multitasking. RT Cores are also used in the development of autonomous vehicles, as they are able to process data quickly and accurately. In addition, RT Cores are used in the development of artificial intelligence, as they are able to process large amounts of data quickly and accurately.