AMD Steam Deck OLED GPU vs Intel Arc B390 Comparison

Head-to-Head Benchmarks

The recorded data contains no direct head-to-head benchmark results between the AMD Steam Deck OLED GPU and the Intel Arc B390. The database shows only a single benchmark entry for the Intel Arc B390: a 3DMark Steel Nomad DX12 score of 1482 points. This places the Intel part at the 9th percentile when compared against all GPUs in the database, a low standing that indicates it trails the vast majority of recorded graphics processors.

For context, the Intel Arc B390's nearest rivals in the database are all older NVIDIA mobile and OEM parts. It sits 1.3% ahead of the GeForce GT 520MX (which scored 1463), 1.5% ahead of the GeForce 800M (1460), 2.5% ahead of the GeForce GT 625 OEM (1446), and 2.7% ahead of the GeForce GT 710 (1443). These are narrow margins, and the Arc B390's score of 1482 is only 39 points above the lowest-scoring rival listed. The data indicates that the Arc B390 is not a high-performance part in absolute terms; it clusters with entry-level GPUs from well over a decade ago.

The AMD Steam Deck OLED GPU has no benchmark scores recorded in the database, and it also has no nearest rivals listed. Its average benchmark score is zero, and its percentile ranking is 50, which is a mid-pack placement. Without measured benchmark data, the database cannot confirm how the AMD part performs in any specific workload. The comparison between these two GPUs must therefore rely on architectural specifications, clock behavior, memory configuration, and the single recorded Intel benchmark, rather than direct score-to-score comparisons.

What can be stated from the available numbers is that the Intel Arc B390 delivers 7.680 TFLOPS of FP32 compute, while the AMD Steam Deck OLED GPU delivers 1.638 TFLOPS. That is a 4.69x difference in raw shader throughput, a substantial gap. In FP16, the Intel part reaches 15.36 TFLOPS versus 3.277 TFLOPS for the AMD part. Pixel throughput favors Intel at 60.00 GPixel/s versus 25.60 GPixel/s, and texture throughput favors Intel at 120.0 GTexel/s versus 51.20 GTexel/s. However, the AMD part is designed for a 15 W envelope, while the Intel part is rated at 80 W, and those figures must be considered when interpreting the performance gap.

Architecture Differences

The architecture split is fundamental. The AMD Steam Deck OLED GPU uses the RDNA 2.0 architecture and is built on TSMC's 6 nm process. The chip is named Sephiroth and is classified in the database as a Console GPU (Valve). The Intel Arc B390 uses the Xe3-LPG architecture and is built on Intel's 3 nm process, with the chip named Panther Lake and classified as Arc Graphics-M (Panther Lake). The process node difference is significant: 6 nm versus 3 nm, and the foundries differ as well, TSMC for the AMD part and Intel for the Intel part.

The AMD chip contains 2,400 million transistors on a die size of 131 mm², producing a transistor density of 18.3 million transistors per square millimeter. The Intel chip's transistor count and die size are listed as unknown in the database, so no density figure can be computed or compared.

Shader resources differ sharply. The AMD part has 512 shading units, 32 texture mapping units, 16 raster operation units, and 8 ray tracing cores. The Intel part has 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. In every one of these counts, the Intel part has more hardware: three times the shading units, 1.5 times the TMUs, 1.5 times the ROPs, and 1.5 times the RT cores.

Clock behavior is also very different. The AMD GPU has a base clock of 1000 MHz and a boost clock of 1600 MHz. The Intel GPU has a base clock of 300 MHz and a boost clock of 2500 MHz. The Intel part's base clock is far lower, but its boost clock is 900 MHz higher than the AMD part's boost. The Intel part's boost clock is 56.25% higher than the AMD part's boost clock, which helps explain its higher throughput rates despite the lower base operating point.

Memory architecture is a major divergence. The AMD Steam Deck OLED GPU uses 16 GB of LPDDR5 memory on a 128-bit bus, with memory clocks of 1375 MHz and 11 Gbps effective, yielding a bandwidth of 176.0 GB/s. The Intel Arc B390 uses system shared memory, with the bus width, type, and bandwidth all listed as system dependent. This means the Intel part's memory performance is not fixed; it depends entirely on the host platform's memory subsystem. The AMD part has a dedicated, fixed memory configuration, while the Intel part has no dedicated memory of its own.

The Intel Arc B390 is classified as an IGP (integrated graphics processor) with no power connectors and a slot width of "IGP". Its display outputs are listed as portable device dependent. The AMD Steam Deck OLED GPU has a display output of one USB Type-C port. The physical dimensions of the AMD part are recorded as 298 mm in length, 117 mm in height, and 49 mm in width, while the Intel part has no dimensions recorded.

Both parts support DirectX 12 Ultimate (12_2) and OpenGL 4.6. The AMD part supports Vulkan 1.3, while the Intel part supports Vulkan 1.4, a newer version of the API.

Where Each One Wins

Based on the recorded specifications, the Intel Arc B390 wins decisively in raw compute throughput. Its FP32 output of 7.680 TFLOPS is more than quadruple the AMD part's 1.638 TFLOPS. Its FP16 output of 15.36 TFLOPS is similarly dominant against the AMD part's 3.277 TFLOPS. The Intel part also wins in pixel fill rate, 60.00 GPixel/s versus 25.60 GPixel/s, and in texture fill rate, 120.0 GTexel/s versus 51.20 GTexel/s. The Intel part has more shading units, more TMUs, more ROPs, and more RT cores. Its boost clock of 2500 MHz is substantially higher than the AMD part's 1600 MHz boost.

The AMD Steam Deck OLED GPU wins in power efficiency based on the recorded data. It is rated at 15 W TDP, while the Intel Arc B390 is rated at 80 W TDP. The AMD part delivers its 1.638 TFLOPS within that 15 W envelope, while the Intel part requires 80 W for its 7.680 TFLOPS. In terms of performance per watt, the AMD part outputs 0.1092 TFLOPS per watt, while the Intel part outputs 0.0960 TFLOPS per watt. The AMD part also has a fixed memory configuration with 16 GB of dedicated LPDDR5 and 176.0 GB/s of bandwidth, which is a guaranteed resource. The Intel part depends on system shared memory, which is not a fixed quantity.

The AMD part also wins on physical integration specifics for a handheld form factor. It has a recorded physical footprint, while the Intel part has no dimensions listed. The AMD part uses a single USB Type-C display output, while the Intel part's display outputs are dependent on the portable device. The release dates differ: the AMD part was released on 2023-11-08, while the Intel part was released on 2026-01-26. Both are listed as active in production.

In the single recorded benchmark, the Intel Arc B390 scores 1482 points in 3DMark Steel Nomad DX12. The AMD part has no recorded benchmark score, so the database cannot verify any win for the AMD part in a measured workload. The 50th percentile ranking for the AMD part is a database placement, not a measured score, and should not be interpreted as a performance result.

The Verdict

The data supports a clear split verdict. For raw compute performance, the Intel Arc B390 is the stronger part on every measured specification: FP32, FP16, pixel rate, texture rate, shader count, TMUs, ROPs, and RT cores. Its single recorded benchmark score of 1482 in 3DMark Steel Nomad DX12 places it at the 9th percentile overall, however, which indicates that despite its architectural advantages over the AMD part, it is still a low-performing GPU relative to the full database. Its nearest rivals are the GeForce GT 520MX, GeForce 800M, GeForce GT 625 OEM, and GeForce GT 710, all of which are entry-level parts. The margins over those rivals are small: 1.3%, 1.5%, 2.5%, and 2.7%, respectively.

For the AMD Steam Deck OLED GPU, the database has no measured benchmark data. The only quantitative facts are its specifications: 512 shading units, 16 GB LPDDR5, 176.0 GB/s bandwidth, 1.638 TFLOPS FP32, and a 15 W TDP. Its 50th percentile ranking is a database position, not a test result. Users seeking a part with verified benchmark performance in the database would select the Intel Arc B390, because it has a recorded score. Users prioritizing a fixed memory pool and a very low power envelope would select the AMD part, because its 15 W TDP is the lowest power figure recorded for either part.

The power efficiency calculation favors AMD: 0.1092 TFLOPS per watt versus 0.0960 TFLOPS per watt for Intel. The absolute performance favors Intel by a wide margin: 4.69x in FP32. These two facts define the choice. The Intel Arc B390 is an integrated graphics solution with a high boost clock and a large shader array, but it requires 80 W and relies on system shared memory. The AMD Steam Deck OLED GPU is a console-oriented chip with dedicated memory and a much lower power draw.

The data does not support a single universal winner. The Intel part wins on performance specifications and has the only recorded benchmark score. The AMD part wins on power efficiency and dedicated memory. The release timing also differs: the AMD part released on 2023-11-08, and the Intel part on 2026-01-26. Both are active production parts. The choice depends on whether the priority is raw throughput or power efficiency with fixed memory resources.

FAQ

Q: Which GPU has the higher FP32 compute performance?

A: The Intel Arc B390 has 7.680 TFLOPS FP32, which is 4.69x the AMD Steam Deck OLED GPU's 1.638 TFLOPS.

Q: Does the AMD Steam Deck OLED GPU have any recorded benchmark scores?

A: No. The database lists no benchmark entries for the AMD part, and its average benchmark score is zero.

Q: What is the Intel Arc B390's single recorded benchmark result?

A: It scored 1482 points in 3DMark Steel Nomad DX12, placing it at the 9th percentile among all GPUs.

Q: How does the Intel Arc B390 compare to its nearest rivals in the database?

A: It is 1.3% ahead of the GeForce GT 520MX, 1.5% ahead of the GeForce 800M, 2.5% ahead of the GeForce GT 625 OEM, and 2.7% ahead of the GeForce GT 710.

Q: What memory configuration does each GPU use?

A: The AMD part uses 16 GB of LPDDR5 on a 128-bit bus with 176.0 GB/s bandwidth. The Intel part uses system shared memory with system dependent bandwidth.

Q: What are the TDP ratings for each GPU?

A: The AMD Steam Deck OLED GPU is rated at 15 W, and the Intel Arc B390 is rated at 80 W.

Specification Differences

| Specification | AMD Steam Deck OLED GPU | Intel Arc B390 |

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

| Architecture | RDNA 2.0 | Xe3-LPG |

| Process Node | 6 nm | 3 nm |

| Foundry | TSMC | Intel |

| Transistors | 2,400 million | unknown |

| Die Size | 131 mm² | unknown |

| Transistor Density | 18.3M / mm² | null |

| Base Clock | 1000 MHz | 300 MHz |

| Boost Clock | 1600 MHz | 2500 MHz |

| Memory Size | 16 GB | System Shared |

| Memory Type | LPDDR5 | System Shared |

| Memory Bus Width | 128 bit | System Shared |

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

| Shading Units | 512 | 1536 |

| TMUs | 32 | 48 |

| ROPs | 16 | 24 |

| RT Cores | 8 | 12 |

| Pixel Rate | 25.60 GPixel/s | 60.00 GPixel/s |

| Texture Rate | 51.20 GTexel/s | 120.0 GTexel/s |

| FP32 | 1.638 TFLOPS | 7.680 TFLOPS |

| FP16 | 3.277 TFLOPS (2:1) | 15.36 TFLOPS (2:1) |

| TDP | 15 W | 80 W |

| Slot Width | null | IGP |

| Power Connectors | null | None |

| Bus Interface | null | IGP |

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

| Vulkan Version | 1.3 | 1.4 |

| Release Date | 2023-11-08 | 2026-01-26 |

| Dimensions | 298 mm x 117 mm x 49 mm | null |

| Percentile vs All GPUs | 50 | 9 |

| Average Benchmark Score | 0 | 1482 |

| Nearest Rivals | none | GT 520MX, GeForce 800M, GT 625 OEM, GT 710 |