AMD Radeon 890M vs AMD Radeon R7 M380 Comparison

The AMD Radeon R7 M380 and the AMD Radeon 890M represent two very different eras of mobile graphics. The R7 M380 is a 2015-era discrete part built for entry-level laptops, while the 890M is a 2024 integrated GPU (IGP) designed for high-performance APUs. The benchmark data shows a stark generational gap, with the 890M delivering nearly four times the compute performance in the shared Geekbench OpenCL test, yet the two GPUs sit at nearly identical overall percentile rankings due to the different benchmark suites used. This analysis walks through the architectural shifts, specific benchmark results, and the practical implications of each part.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The AMD Radeon 890M scores 37,254, which is 75% higher than the AMD Radeon R7 M380’s score of 9,313 in the same test.

Q: How do the two GPUs compare in terms of transistor density?

A: The 890M has a transistor density of 145.9M / mm², which is dramatically higher than the R7 M380’s 12.2M / mm². This reflects the 890M’s 4 nm process node compared to the R7 M380’s 28 nm node.

Q: What is the memory configuration difference?

A: The R7 M380 features 4 GB of dedicated DDR3 memory on a 128-bit bus with 32.00 GB/s bandwidth. The 890M uses System Shared memory, with its bandwidth listed as System Dependent.

Q: Which GPU supports hardware ray tracing?

A: Only the AMD Radeon 890M supports it, featuring 16 dedicated RT cores. The R7 M380 has no RT cores listed.

Q: What is the production status of each GPU?

A: The R7 M380 is listed as End-of-life, while the 890M is listed as Active production status.

Q: What are the DirectX API support levels?

A: The 890M supports DirectX 12 Ultimate (12_2), while the R7 M380 supports DirectX 12 (11_1).

Architecture Differences

The two GPUs are separated by nearly a decade of architectural evolution. The R7 M380 is built on the GCN 1.0 architecture using the Tropo chip, manufactured on a 28 nm process at TSMC. It packs 1,500 million transistors into a 123 mm² die, resulting in a transistor density of 12.2M / mm². In contrast, the 890M uses the RDNA 3.5 architecture with the Strix Point chip, built on a 4 nm process at the same foundry. This newer chip contains 34,000 million transistors on a 233 mm² die, achieving a density of 145.9M / mm² — a massive leap in integration efficiency.

The compute core layout differs substantially. The R7 M380 has 640 shading units, 40 texture mapping units (TMUs), and 16 raster operation pipelines (ROPs). The 890M nearly doubles the shader count to 1,024, and doubles both TMUs to 64 and ROPs to 32. The 890M also introduces 16 RT cores, a feature entirely absent from the older GCN part. Clock speeds also tell a story: the R7 M380 runs at a 900 MHz base and 915 MHz boost, while the 890M has a low 400 MHz base but boosts up to 2,900 MHz, reflecting its efficient 4 nm design.

Memory architecture is fundamentally different. The R7 M380 uses 4 GB of dedicated DDR3 with a 128-bit bus, yielding 32.00 GB/s of bandwidth. The 890M relies on System Shared memory, meaning its bandwidth is System Dependent. The bus interface also differs, with the R7 M380 using PCIe 3.0 x16 and the 890M using PCIe 4.0 x8. The 890M’s power profile is notably efficient at 15 W TDP, while the R7 M380’s TDP is not listed.

Where Each One Wins

The benchmark data shows a clear split in strengths. The 890M wins heavily in raw compute and modern API support. In the Geekbench OpenCL test, it achieves 37,254 points, a 75% lead over the R7 M380. This advantage is backed by its much higher pixel rate of 92.80 GPixel/s and texture rate of 185.6 GTexel/s, compared to the R7 M380’s 14.64 GPixel/s and 36.60 GTexel/s respectively. The 890M also supports DirectX 12 Ultimate and Vulkan 1.4, making it suitable for modern titles that require these features.

The R7 M380, despite being end-of-life, holds its own in one specific regard: its dedicated memory. The 4 GB of DDR3 on a 128-bit bus provides a fixed 32.00 GB/s bandwidth, which can be more predictable than system-shared memory that depends on the host platform. Its PCIe 3.0 x16 interface is also wider than the 890M’s PCIe 4.0 x8, though the newer standard offers higher per-lane throughput. In terms of the shared benchmark, the R7 M380’s score of 9,313 places it within 1% of the AMD Radeon Vega 8, showing it remained competitive with early integrated graphics from its era.

For users prioritizing modern feature sets, hardware ray tracing, and raw compute throughput, the 890M is the clear winner. The R7 M380’s niche would be legacy applications or systems requiring a discrete GPU with dedicated VRAM, but the data shows it cannot match the 890M in any benchmark test where both have results.

Specification Differences

| Specification | AMD Radeon R7 M380 | AMD Radeon 890M |

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

| Architecture | GCN 1.0 | RDNA 3.5 |

| Process Node | 28 nm | 4 nm |

| Transistors | 1,500 million | 34,000 million |

| Die Size | 123 mm² | 233 mm² |

| Transistor Density | 12.2M / mm² | 145.9M / mm² |

| Base Clock | 900 MHz | 400 MHz |

| Boost Clock | 915 MHz | 2900 MHz |

| Memory | 4 GB DDR3 | System Shared |

| Memory Bus | 128 bit | System Shared |

| Memory Bandwidth | 32.00 GB/s | System Dependent |

| Shading Units | 640 | 1024 |

| TMUs | 40 | 64 |

| ROPs | 16 | 32 |

| RT Cores | None | 16 |

| Pixel Rate | 14.64 GPixel/s | 92.80 GPixel/s |

| Texture Rate | 36.60 GTexel/s | 185.6 GTexel/s |

| FP32 | 1,171.2 GFLOPS | 5.939 TFLOPS |

| FP16 | N/A | 5.939 TFLOPS (1:1) |

| TDP | N/A | 15 W |

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

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

| Vulkan | 1.2.170 | 1.4 |

| Production Status | End-of-life | Active |

| Release Date | 2015-05-04 | 2024-07-14 |

Head-to-Head Benchmarks

The only shared benchmark between the two GPUs is the Geekbench OpenCL test, and the results are decisive. The AMD Radeon 890M scores 37,254, while the AMD Radeon R7 M380 scores 9,313. This represents a 75% delta in favor of the 890M, meaning the newer GPU delivers approximately 4.0 times the compute performance of the older part. This is the single head-to-head data point available, and it aligns with the raw specification differences.

Looking at the broader benchmark results for the 890M, it also achieves 40,808 in Geekbench Vulkan and 8,076 in Passmark G3D, though the R7 M380 has no comparable results for these tests. The 890M’s Passmark GPU Compute score of 4,157 further demonstrates its compute focus. The R7 M380’s only benchmark result is the Geekbench OpenCL score of 9,313, which puts it in the 46th percentile of all GPUs. The 890M’s average benchmark score is 9,210, putting it in the 45th percentile — a nearly identical ranking despite the massive difference in raw performance.

The nearest rivals for each GPU show interesting convergence. The R7 M380’s closest competitor is the NVIDIA GeForce GTX 850M at 9,302 (0.1% delta), while the 890M’s closest is the AMD Radeon Vega 8 at 9,221 (-0.1% delta). This suggests that in terms of average benchmark scores, both GPUs sit in a similar performance class, but the 890M’s much higher peak scores in specific tests like Geekbench OpenCL skew its average downward due to lower scores in other tests like Passmark DirectX 9 (97) and DirectX 11 (73).

The Verdict

The data presents a straightforward conclusion: the AMD Radeon 890M is the superior GPU by nearly every measurable metric. Its Geekbench OpenCL score of 37,254 dwarfs the R7 M380’s 9,313, and its architectural advantages — 4 nm process, RDNA 3.5, 16 RT cores, and 2,900 MHz boost clock — make it the clear choice for any modern workload. The 890M supports DirectX 12 Ultimate, which is essential for current gaming titles, and its 15 W TDP makes it highly efficient for mobile devices.

The R7 M380, while end-of-life, still has a place in legacy systems. Its dedicated 4 GB of DDR3 memory provides fixed bandwidth that can be advantageous in scenarios where system memory is slow or shared with the CPU. Its PCIe 3.0 x16 interface is also wider than the 890M’s x8, though the newer standard’s higher data rates likely negate this advantage. For users running older software that relies on GCN architecture quirks, the R7 M380 remains a functional option.

For new purchases, the 890M is the only logical choice. Its 75% lead in the shared benchmark, combined with modern API support and hardware ray tracing, ensures it will handle a wider range of tasks. The R7 M380’s 46th percentile ranking mirrors the 890M’s 45th, but this is misleading given the disparity in test coverage. The 890M’s higher peak scores in compute-heavy workloads, such as its 40,808 Geekbench Vulkan result, demonstrate a capability far beyond the R7 M380’s reach. Users seeking a GPU for modern gaming, creative work, or general compute should choose the 890M; those maintaining older systems may find the R7 M380 sufficient for basic tasks, but the data shows no scenario where it outperforms the newer part.