AMD Radeon R9 M290X vs Intel Arc A350M Comparison

The Intel Arc A350M and AMD Radeon R9 M290X represent two very different eras of mobile graphics. The Arc A350M is a modern, power-efficient entry point into Intel’s discrete GPU lineup, built on a state-of-the-art 6 nm process. The R9 M290X is a legacy part from 2014, a high-power MXM module based on AMD’s older 28 nm GCN architecture. The data shows a clear overall winner in raw compute, but the R9 M290X retains a significant advantage in memory bandwidth that matters for specific workloads.

Where Each One Wins

The benchmark results paint a straightforward picture for compute performance. The Intel Arc A350M wins the only head-to-head benchmark available, Geekbench OpenCL, with a decisive 11.4% advantage over the Radeon R9 M290X. This is a clear victory for the newer architecture in general-purpose compute tasks that leverage OpenCL. The Arc A350M’s average benchmark score of 24647 also places it in the 70th percentile of all GPUs, compared to the R9 M290X’s 68th percentile. This indicates the Intel part sits slightly higher in the overall performance hierarchy.

However, the R9 M290X is not without its strengths. Its memory subsystem is substantially more capable. The R9 M290X features a 256-bit memory bus paired with 4 GB of GDDR5 memory, delivering a bandwidth of 153.6 GB/s. In contrast, the Arc A350M uses a 64-bit bus with 4 GB of GDDR6, yielding a much lower 112.0 GB/s. This 41.6 GB/s difference means the AMD card has a theoretical advantage in scenarios that are heavily reliant on memory bandwidth, such as high-resolution texture streaming or certain compute workloads that access large datasets. While the raw shader compute is lower, the R9 M290X’s wider pipe can be a saving grace in specific, bandwidth-bound tasks.

Architecture Differences

The architectural gap between these two is vast. The Arc A350M is built on Intel’s Xe-HPG architecture, specifically the DG2-128 chip, and fabricated on a 6 nm process at TSMC. This allows Intel to pack 7,200 million transistors into a 157 mm² die, achieving a transistor density of 45.9 million per square millimeter. The R9 M290X, codenamed Neptune, uses the older GCN 1.0 architecture on a 28 nm process. It contains only 2,800 million transistors on a much larger 212 mm² die, resulting in a density of just 13.2 million per square millimeter. This process advantage is a primary driver of the Arc A350M’s superior performance-per-watt and clock speeds.

The core configurations differ significantly. The Arc A350M has 768 shading units, 48 texture mapping units (TMUs), and 24 raster operation units (ROPs). It also includes 6 dedicated ray tracing cores, a feature entirely absent from the R9 M290X. This makes the Intel GPU capable of hardware-accelerated ray tracing, a modern feature the older AMD card cannot offer. The R9 M290X, by contrast, has a higher count of 1280 shading units and 80 TMUs, but only 32 ROPs. Despite having more shaders, its lower clock speeds and older architecture result in lower overall throughput.

Clock speeds are another major differentiator. The Arc A350M has a base clock of 1150 MHz and a boost clock of 2200 MHz. The R9 M290X operates at a much lower 850 MHz base and 900 MHz boost. This clock speed advantage, combined with the modern architecture, allows the Intel part to achieve higher pixel and texture rates. The Arc A350M hits 52.80 GPixel/s and 105.6 GTexel/s, while the R9 M290X manages 28.80 GPixel/s and 72.00 GTexel/s. The memory type also differs, with the Arc using GDDR6 at 14 Gbps effective and the R9 using GDDR5 at 4.8 Gbps effective, though the bus width differences temper this clock advantage.

Head-to-Head Benchmarks

The sole head-to-head result in the data is the Geekbench OpenCL test. In this test, the Intel Arc A350M scores 24546, while the AMD Radeon R9 M290X scores 22028. This results in an 11.4% victory for the Intel Arc A350M. This is a substantial margin that reflects the architectural and clock speed advantages discussed above. To put this in perspective, the Arc A350M’s score places it near rivals like the NVIDIA RTX A5000 Mobile (24763, -0.5%) and the AMD Radeon RX 590 (24744, -0.4%). The R9 M290X’s score of 22028 is closer to its nearest rival, the AMD Radeon RX 6600M (23273, 0%).

The Arc A350M also shows a strong result in the Geekbench Vulkan test, scoring 24747. While there is no direct comparison for this test against the R9 M290X, it reinforces the Intel part’s capability in modern graphics APIs. The R9 M290X only has a Geekbench Metal score of 24524, which is a strong result for that specific API but does not translate to the same cross-platform compute performance as the Intel part’s OpenCL and Vulkan scores. The data shows that in the most general compute benchmark available, the newer Intel architecture is clearly superior.

The Verdict

From the data, the Intel Arc A350M is the better overall GPU for modern workloads. It wins the only direct benchmark comparison by a significant 11.4% margin. Its support for DirectX 12 Ultimate (12_2) and Vulkan 1.4, along with hardware ray tracing, makes it a more future-proof option for gaming and modern applications. The higher boost clock and newer 6 nm process give it a clear performance edge in compute tasks. If you are choosing between these two for a system that will run contemporary software, the Arc A350M is the logical choice based on performance and feature set.

The AMD Radeon R9 M290X, however, is not entirely obsolete. Its 256-bit memory bus and 153.6 GB/s of bandwidth remain its key asset. For workloads that are purely memory-bandwidth limited, this card could potentially outperform the Arc A350M, despite its lower compute score. It also has a higher shading unit count, which can sometimes benefit older games or applications that do not scale well with modern architectures. However, these advantages come with a 100 W TDP, which is four times higher than the Arc A350M’s 25 W. For most users, the efficiency and raw compute of the Intel part make it the superior choice. The R9 M290X is a niche option for legacy systems or specific bandwidth-hungry tasks.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The Intel Arc A350M is significantly faster, scoring 24546 compared to the AMD Radeon R9 M290X’s 22028 in the Geekbench OpenCL test, an 11.4% advantage.

Q: Does the AMD Radeon R9 M290X have a higher memory bandwidth?

A: Yes. The R9 M290X has a 256-bit bus with GDDR5 memory, providing 153.6 GB/s of bandwidth. The Intel Arc A350M has a 64-bit bus with GDDR6, providing 112.0 GB/s.

Q: Which GPU supports hardware ray tracing?

A: Only the Intel Arc A350M supports hardware ray tracing, as it includes 6 dedicated ray tracing cores. The AMD Radeon R9 M290X has no such cores.

Q: What are the power consumption differences?

A: The Intel Arc A350M has a TDP of 25 W, while the AMD Radeon R9 M290X has a TDP of 100 W. The Intel part is significantly more power-efficient.

Q: Which GPU has a higher transistor count?

A: The Intel Arc A350M has 7,200 million transistors, which is substantially more than the 2,800 million in the AMD Radeon R9 M290X.

Q: What is the process node for each GPU?

A: The Intel Arc A350M is built on a 6 nm process at TSMC. The AMD Radeon R9 M290X is built on an older 28 nm process, also at TSMC.

Specification Differences

| Specification | Intel Arc A350M | AMD Radeon R9 M290X |

|:--- |:--- |:--- |

| Architecture | Xe-HPG | GCN 1.0 |

| Process Node | 6 nm | 28 nm |

| Transistors | 7,200 million | 2,800 million |

| Die Size | 157 mm² | 212 mm² |

| Base Clock | 1150 MHz | 850 MHz |

| Boost Clock | 2200 MHz | 900 MHz |

| Memory Type | GDDR6 | GDDR5 |

| Memory Bus Width | 64 bit | 256 bit |

| Memory Bandwidth | 112.0 GB/s | 153.6 GB/s |

| Shading Units | 768 | 1280 |

| TMUs | 48 | 80 |

| ROPs | 24 | 32 |

| Ray Tracing Cores | 6 | None |

| Pixel Rate | 52.80 GPixel/s | 28.80 GPixel/s |

| Texture Rate | 105.6 GTexel/s | 72.00 GTexel/s |

| FP32 Performance | 3.379 TFLOPS | 2.304 TFLOPS |

| TDP | 25 W | 100 W |

| Slot Width | IGP | MXM Module |

| Bus Interface | PCIe 4.0 x8 | PCIe 3.0 x16 |

| DirectX Support | 12 Ultimate (12_2) | 12 (11_1) |

| Vulkan Support | 1.4 | 1.2.170 |