AMD Radeon 820M vs AMD Steam Deck OLED GPU Comparison

Head-to-Head Benchmarks

The recorded data for both the AMD Radeon 820M and the AMD Steam Deck OLED GPU shows no benchmark entries in the database. Neither part has a measured average benchmark score, and the head-to-head comparison table is empty. As a result, there are no exact performance deltas, percentile shifts, or win counts to report from our measurements.

The absence of scores does not erase the architectural evidence. The Radeon 820M is built on RDNA 3.5 with a 4 nm process, while the Steam Deck OLED GPU uses RDNA 2.0 on a 6 nm node. The Steam Deck OLED GPU carries 512 shading units, 32 texture mapping units, and 16 render output units. The Radeon 820M has 128 shading units, 8 TMUs, and 4 ROPs. That is a 4x difference in shading units, a 4x difference in TMUs, and a 4x difference in ROPs, all in favor of the Steam Deck OLED GPU.

Clock speeds tell a different story. The Radeon 820M boosts to 2800 MHz, whereas the Steam Deck OLED GPU boosts to 1600 MHz. The Radeon 820M also has a higher base clock at 400 MHz versus 1000 MHz for the Steam Deck part, meaning the Steam Deck GPU starts higher but tops out lower. When clock rate is combined with unit counts, the peak FP32 throughput for the Steam Deck OLED GPU is 1.638 TFLOPS, while the Radeon 820M reaches 716.8 GFLOPS. The Steam Deck OLED GPU leads by roughly a factor of 2.3 in raw single-precision compute.

Pixel rate confirms the same direction. The Steam Deck OLED GPU outputs 25.60 GPixel/s, and the Radeon 820M manages 11.20 GPixel/s. Texture rate shows 51.20 GTexel/s for the Steam Deck part versus 22.40 GTexel/s for the Radeon 820M. Both pixel and texture throughput are about 2.3x higher on the Steam Deck OLED GPU.

The Radeon 820M counters with a much higher boost clock, 2800 MHz versus 1600 MHz, and a newer architecture generation. RDNA 3.5 over RDNA 2.0 brings architectural improvements, though the database does not quantify those gains for this specific pair. The Radeon 820M also supports a newer Vulkan version, 1.4 versus 1.3, and a higher memory clock specification at 1375 MHz for the Steam Deck part versus "System Shared" for the Radeon.

FAQ

Q: Which GPU has more shading units?

A: The AMD Steam Deck OLED GPU has 512 shading units, while the AMD Radeon 820M has 128 shading units.

Q: What is the peak FP32 performance difference?

A: The Steam Deck OLED GPU delivers 1.638 TFLOPS FP32, and the Radeon 820M delivers 716.8 GFLOPS FP32. The Steam Deck part has roughly 2.3x higher FP32 throughput.

Q: Do the two GPUs use the same architecture?

A: No. The Radeon 820M uses RDNA 3.5, and the Steam Deck OLED GPU uses RDNA 2.0.

Q: What are the boost clock speeds?

A: The Radeon 820M boosts to 2800 MHz. The Steam Deck OLED GPU boosts to 1600 MHz.

Q: How much memory does each GPU use?

A: The Steam Deck OLED GPU has 16 GB of LPDDR5 memory on a 128-bit bus with 176.0 GB/s bandwidth. The Radeon 820M uses system-shared memory with system-dependent bandwidth.

Q: Which GPU has ray tracing cores?

A: Both have ray tracing cores. The Radeon 820M has 2, and the Steam Deck OLED GPU has 8.

Architecture Differences

The Radeon 820M is built on the RDNA 3.5 architecture and belongs to the Navi III IGP generation for Strix Point Mobile. It uses a 4 nm TSMC process node. The chip is named Krackan Point 2. The Steam Deck OLED GPU is built on RDNA 2.0 and belongs to the Console GPU generation from Valve. It uses a 6 nm TSMC process node and the chip is named Sephiroth.

Transistor data is only available for the Steam Deck OLED GPU, which has 2,400 million transistors on a 131 mm² die, giving a transistor density of 18.3M per mm². The Radeon 820M has unknown transistor count and die size. The Radeon 820M has 128 shading units, 8 TMUs, 4 ROPs, and 2 ray tracing cores. The Steam Deck OLED GPU has 512 shading units, 32 TMUs, 16 ROPs, and 8 ray tracing cores.

The Radeon 820M's FP16 throughput is 716.8 GFLOPS with a 1:1 ratio to FP32. The Steam Deck OLED GPU's FP16 throughput is 3.277 TFLOPS with a 2:1 ratio, meaning it doubles FP16 rate relative to FP32. The Radeon 820M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Steam Deck OLED GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.3.

The Radeon 820M is an integrated graphics processor with a 15 W TDP and no power connectors. It uses a PCIe 4.0 x8 bus interface and its display outputs are portable-device dependent. The Steam Deck OLED GPU also has a 15 W TDP, but its bus interface is not listed. Its display output is a single USB Type-C port. The Steam Deck GPU has physical dimensions of 298 mm in length, 117 mm in height, and 49 mm in width. The Radeon 820M has no listed physical dimensions because it is an IGP.

The Radeon 820M was released on 2025-02-28 and has a predecessor of Navi II IGP. The Steam Deck OLED GPU was released on 2023-11-08 and has no listed predecessor or successor. Both parts are listed as Active in production status.

The Verdict

The data shows two very different designs aimed at different roles. The Steam Deck OLED GPU has a massive lead in raw resource counts: 4x shading units, 4x TMUs, 4x ROPs, and 4x ray tracing cores. It also has dedicated 16 GB of LPDDR5 memory with 176.0 GB/s bandwidth, while the Radeon 820M relies on system-shared memory with system-dependent bandwidth. The Steam Deck part achieves 1.638 TFLOPS FP32 versus 716.8 GFLOPS for the Radeon 820M, a clear compute advantage.

The Radeon 820M counters with a higher boost clock of 2800 MHz versus 1600 MHz, a newer RDNA 3.5 architecture over RDNA 2.0, and a smaller 4 nm process node versus 6 nm. It also supports the newer Vulkan 1.4 API. However, the database records no benchmark scores to show whether the newer architecture and higher clocks close the gap in real workloads. The measured pixel rate and texture rate both favor the Steam Deck OLED GPU by roughly 2.3x.

For a system that needs an integrated GPU inside a portable device, the Radeon 820M offers a modern architecture on a leading-edge node with a high boost clock and system-shared memory flexibility. For a fixed console-style platform with dedicated memory and substantially more execution resources, the Steam Deck OLED GPU is the stronger part on paper. The absence of benchmark data means the verdict rests on the listed specifications: the Steam Deck OLED GPU is the higher-throughput processor, and the Radeon 820M is the more advanced, more compact integrated solution.

Specification Differences

| Specification | AMD Radeon 820M | AMD Steam Deck OLED GPU |

|---|---|---|

| Architecture | RDNA 3.5 | RDNA 2.0 |

| Process Node | 4 nm | 6 nm |

| Chip | Krackan Point 2 | Sephiroth |

| Transistors | Unknown | 2,400 million |

| Die Size | Unknown | 131 mm² |

| Base Clock | 400 MHz | 1000 MHz |

| Boost Clock | 2800 MHz | 1600 MHz |

| Memory Size | System Shared | 16 GB |

| Memory Type | System Shared | LPDDR5 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 176.0 GB/s |

| Shading Units | 128 | 512 |

| TMUs | 8 | 32 |

| ROPs | 4 | 16 |

| Ray Tracing Cores | 2 | 8 |

| Pixel Rate | 11.20 GPixel/s | 25.60 GPixel/s |

| Texture Rate | 22.40 GTexel/s | 51.20 GTexel/s |

| FP32 Performance | 716.8 GFLOPS | 1.638 TFLOPS |

| FP16 Performance | 716.8 GFLOPS (1:1) | 3.277 TFLOPS (2:1) |

| TDP | 15 W | 15 W |

| Bus Interface | PCIe 4.0 x8 | Not listed |

| Display Outputs | Portable Device Dependent | 1x USB Type-C |

| Vulkan Version | 1.4 | 1.3 |

| Release Date | 2025-02-28 | 2023-11-08 |

| Dimensions | Not listed | 298 mm x 117 mm x 49 mm |