RDNA 2 (Released November 18, 2020) is AMD’s breakthrough GPU microarchitecture, powering the Radeon RX 6000 series and next-gen consoles (PlayStation 5, Xbox Series X/S, Steam Deck).
It marked AMD’s return to high-end GPU competitiveness with significant innovations in efficiency, caching, and feature support 128. Below is a detailed analysis of its key aspects:
⚙️ 1. Architectural Innovations
Compute Units (CUs) and Workgroup Processors (WGPs)
- Enhanced CU Design: Each RDNA 2 CU contains 64 shader cores, with up to 80 CUs in flagship dies (e.g., Navi 21). This doubled the CU count from RDNA 1’s max of 40 CUs 16.
- WGP Organization: Two CUs form a WGP, sharing 32 KB L0 cache (16 KB vector + 16 KB scalar). WGPs operate in two modes:
- WGP Mode: Unified 128 KB Local Data Store (LDS).
- CU Mode: Split 64 KB LDS halves for finer task handling 5.
- Efficiency Gains: RDNA 2 achieved 54% higher performance-per-watt vs. RDNA 1, with 21% attributed to Infinity Cache and 30% to clock-speed optimizations 16.
Infinity Cache
- Purpose: A global L3 cache (16–128 MB) designed to reduce reliance on GDDR6 memory bandwidth. It allowed AMD to use a narrower 256-bit bus instead of power-hungry 512-bit designs 168.
- Performance:
- 128 MB cache (Navi 21) delivered 56% hit rates at 4K, reducing latency by 34% vs. RDNA 1 6.
- Peak bandwidth of 1,986.6 GB/s, enabling effective bandwidth 2.17× higher than raw GDDR6 specs 18.
- Lower-tier GPUs (e.g., Navi 23: 32 MB; Navi 22: 96 MB) scaled cache size for cost efficiency 8.
🚀 2. Key Technologies
Ray Tracing
- Ray Accelerators: Dedicated units per CU handling 4 ray/box intersections or 1 ray/triangle intersection per clock. Accessed via DirectX Raytracing (DXR) or VulkanRT 18.
- Performance: Initially lagged behind Nvidia’s Ampere but gained up to 40% improvement via driver updates in 2023 18.
Clock Speeds and Power Efficiency
- Frequency Boost: RDNA 2 GPUs hit >2.1 GHz (e.g., RX 6800 XT: 2,250 MHz boost), a 30% increase over RDNA 1 at same power 168.
- Optimizations: Fine-grain clock gating, redesigned data paths, and CPU-GPU collaboration on TSMC’s 7 nm process 6.
Media Engine
- VCN 3.x Block: Added AV1 decoding (up to 8K) but omitted encoding until RDNA 3. Lower-end dies (Navi 24) lack media encoders entirely 18.
💻 3. Product Lineup and Specifications
RDNA 2 GPUs spanned four dies, targeting desktop, mobile, and workstation markets:
Desktop GPUs (Select Models)
| Model | GPU Die | CUs / Cores | Infinity Cache | Memory | TDP | Launch MSRP |
|---|---|---|---|---|---|---|
| RX 6950 XT | Navi 21 | 80 / 5120 | 128 MB | 16 GB GDDR6 | 335 W | $1,099 |
| RX 6900 XT | Navi 21 | 80 / 5120 | 128 MB | 16 GB GDDR6 | 300 W | $999 |
| RX 6800 XT | Navi 21 | 72 / 4608 | 128 MB | 16 GB GDDR6 | 300 W | $649 |
| RX 6700 XT | Navi 22 | 40 / 2560 | 96 MB | 12 GB GDDR6 | 230 W | $479 |
| RX 6600 XT | Navi 23 | 32 / 2048 | 32 MB | 8 GB GDDR6 | 160 W | $379 |
| RX 6500 XT | Navi 24 | 16 / 1024 | 16 MB | 4 GB GDDR6 | 107 W | $199 |
*Post-2023 price cuts reduced MSRPs by 5–30% (e.g., RX 6900 XT: $699) 9.*
Console Integration
- PlayStation 5 / Xbox Series X|S: Custom RDNA 2 APUs with 36 CUs (PS5) and 52 CUs (Xbox Series X), optimized for console-specific memory and cooling 18.
⚖️ 4. Competitive Analysis
- vs. Nvidia Ampere:
- Strengths: Higher rasterization performance at 1440p (thanks to Infinity Cache), lower VRAM latency, and aggressive pricing 68.
- Weaknesses: Ray tracing and AI (e.g., DLSS) lagged behind Ampere 8.
- Efficiency: Matched Nvidia in power efficiency despite using GDDR6 (vs. power-hungry GDDR6X) 8.
🌍 5. Impact and Legacy
- Market Position: Restored AMD’s competitiveness in high-end GPUs after a decade of trailing Nvidia 68.
- Developer Ecosystem: Supported DirectX 12 Ultimate (ray tracing, mesh shaders) and AMD’s FidelityFX tools, enhancing game optimization 2.
- Successor: RDNA 3 (2022) built on this foundation with chiplet design and >50% perf-per-watt gains