AMD Radeon 890M vs AMD Radeon R9 M375X Comparison

Head-to-Head Benchmarks

The benchmark data for these two GPUs reveals a decisive performance gap, with the AMD Radeon 890M winning both recorded head-to-head tests outright. The 890M’s Geekbench OpenCL score of 37,254 dwarfs the R9 M375X’s 8,273, representing a 350.3% advantage. This is not a marginal lead; it is a generational leap in raw compute throughput that places the integrated part firmly in a different performance class.

The Vulkan results tell an even more pronounced story. The 890M scores 40,808 against the R9 M375X’s 8,377, a 387.1% delta in favor of the newer chip. Vulkan’s lower overhead and better multi-threaded scaling tend to reward modern architectures, and the data confirms that the 890M’s RDNA 3.5 design extracts far more from the API than the aging GCN 1.0 implementation in the R9 M375X.

Looking at the broader database averages, the 890M posts an average benchmark score of 9,210, while the R9 M375X manages 8,325. That is a 10.6% gap in the aggregate, though the head-to-head tests show a much wider chasm. The discrepancy arises because the R9 M375X’s average is drawn from only two recorded tests (both OpenCL and Vulkan), whereas the 890M has a richer benchmark profile including DirectX and Passmark suites. In the tests where both parts appear, the 890M is consistently 3.5 to 3.9 times faster.

The 890M sits at the 45th percentile of all GPUs in the database, while the R9 M375X ranks at the 43rd percentile. On the surface, those percentiles look close, but the underlying scores tell a different story. The 890M’s nearest rivals include the AMD Radeon Vega 8 (average 9,221, a 0.1% delta), the NVIDIA GeForce GTX 960 (9,273, a 0.7% delta), and the GTX 465 (9,294, a 0.9% delta). The R9 M375X, by contrast, sits near the NVIDIA Quadro K1200 (8,265, a 0.7% delta) and the GTX 675MX (8,427, a 1.2% delta). The 890M is competing with desktop-class discrete GPUs from several generations ago; the R9 M375X is trading blows with entry-level mobile parts.

In the only two benchmarks where both GPUs have recorded scores, the 890M wins both. The margin is so large that no test in the shared set favors the R9 M375X. The wins are not close: the smallest delta is 350.3% in OpenCL, and the largest is 387.1% in Vulkan. For context, a 350% advantage means the 890M delivers roughly 4.5 times the performance of the R9 M375X in OpenCL workloads. In Vulkan, that ratio approaches 4.9 times.

Where Each One Wins

The data shows no benchmark category where the AMD Radeon R9 M375X outperforms the 890M. In the two shared tests, the 890M wins both, and the R9 M375X has zero recorded victories in the head-to-head set. This is a one-sided comparison.

For the 890M, the wins span both compute-oriented APIs. Its OpenCL score of 37,254 indicates strong general-purpose compute capability, useful for tasks like video encoding, physics simulations, and machine learning inference. The Vulkan score of 40,808 points to robust graphics performance under modern APIs, which matters for gaming and real-time rendering workloads. The 890M also has recorded scores in DirectX 12 (Passmark 37), DirectX 11 (73), DirectX 10 (33), DirectX 9 (97), and a Passmark G3D score of 8,076, indicating broad API compatibility.

The R9 M375X has no DirectX or Passmark scores in the database, so its capabilities in those areas cannot be quantified here. Its only recorded results are the two Geekbench tests, both of which are far below the 890M. The R9 M375X’s architecture, GCN 1.0, supports DirectX 12 (11_1) and Vulkan 1.2.170, but the recorded performance data shows it is not competitive in the tests that exist.

In practical terms, the 890M is the clear choice for any workload that leverages OpenCL or Vulkan. The R9 M375X does not win anywhere in the recorded data. If a use case relies on legacy DirectX 9 or 10 paths, the 890M still has recorded scores (97 and 33 respectively in Passmark), while the R9 M375X has none, leaving its legacy performance undocumented.

Architecture Differences

The two GPUs come from entirely different architectural eras. The AMD Radeon 890M is built on RDNA 3.5, the latest graphics architecture from AMD, fabricated on a 4 nm process at TSMC. The R9 M375X uses GCN 1.0, AMD’s first-generation Graphics Core Next design, on a 28 nm process, also from TSMC. The process node difference alone is massive: 4 nm versus 28 nm represents a seven-generation leap in manufacturing technology.

The transistor counts reflect this gap. The 890M packs 34,000 million transistors on a 233 mm² die, yielding a transistor density of 145.9 million per square millimeter. The R9 M375X has just 1,500 million transistors on a 123 mm² die, for a density of 12.2 million per square millimeter. The 890M has more than 22 times the transistor count and nearly 12 times the density, despite being an integrated graphics processor (IGP) rather than a discrete part.

The 890M is part of the "Strix Point" chip, belonging to the Navi III IGP generation. It integrates 1,024 shading units, 64 texture mapping units, 32 ROPs, and 16 ray tracing cores. The R9 M375X, based on the "Tropo" chip from the Gem System (R9 M300) generation, has 640 shading units, 40 TMUs, and 16 ROPs, with no ray tracing cores. The 890M also has a dedicated ray tracing block, a feature entirely absent from the GCN 1.0 design.

Clock speeds tell a nuanced story. The R9 M375X runs at a base clock of 925 MHz and boosts to 1015 MHz, which are high for its era. The 890M has a much lower base clock of 400 MHz but a far higher boost clock of 2900 MHz. The 890M’s boost clock is nearly three times the R9 M375X’s boost, which helps explain its massive performance lead despite the lower base frequency.

Memory architecture differs fundamentally. The 890M uses system-shared memory, with its size, type, bus width, and bandwidth all dependent on the host system. The R9 M375X has dedicated 2 GB of GDDR5 on a 128-bit bus, delivering 72.00 GB/s of bandwidth. The 890M’s memory bandwidth is listed as "System Dependent," meaning it varies with the platform’s RAM configuration, but its compute throughput advantage compensates for any bandwidth limitations.

The API support also reflects the generational gap. The 890M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The R9 M375X supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The 890M’s higher DirectX feature level and newer Vulkan version enable features like mesh shaders and variable rate shading, which the R9 M375X cannot handle.

Specification Differences

The two GPUs differ across nearly every measurable specification. The process node is 4 nm for the 890M versus 28 nm for the R9 M375X. Transistor count is 34,000 million versus 1,500 million. Die size is 233 mm² versus 123 mm². Transistor density is 145.9 million per square millimeter versus 12.2 million.

The shading unit count is 1,024 versus 640, a 60% advantage for the 890M. Texture mapping units are 64 versus 40, and ROPs are 32 versus 16. The 890M has 16 ray tracing cores; the R9 M375X has none. Pixel rate is 92.80 GPixel/s for the 890M versus 16.24 GPixel/s for the R9 M375X. Texture rate is 185.6 GTexel/s versus 40.60 GTexel/s. FP32 performance is 5.939 TFLOPS versus 1,299.2 GFLOPS (approximately 1.30 TFLOPS), a 4.6 times advantage for the 890M. The 890M supports FP16 at a 1:1 ratio with 5.939 TFLOPS; the R9 M375X has no recorded FP16 capability.

Clock speeds differ: base is 400 MHz versus 925 MHz, boost is 2900 MHz versus 1015 MHz. The 890M has no dedicated memory, using system-shared resources, while the R9 M375X has 2 GB of GDDR5 on a 128-bit bus with 72.00 GB/s bandwidth. The 890M’s TDP is 15 W; the R9 M375X has no recorded TDP. The 890M is an IGP with no power connectors and a PCIe 4.0 x8 interface; the R9 M375X uses PCIe 3.0 x16.

The 890M is currently in production, released on July 14, 2024, with a successor of Navi II IGP. The R9 M375X is end-of-life, released on May 4, 2015, with a successor of Polaris Mobile. The 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4; the R9 M375X supports DirectX 12 (11_1) and Vulkan 1.2.170.

FAQ

Q: How much faster is the AMD Radeon 890M than the R9 M375X in OpenCL?

A: The 890M scores 37,254 in Geekbench OpenCL versus 8,273 for the R9 M375X, a 350.3% advantage.

Q: Which GPU has better Vulkan performance?

A: The 890M wins decisively with a score of 40,808 against the R9 M375X’s 8,377, a 387.1% delta.

Q: Does the R9 M375X win any recorded benchmark?

A: No. In the head-to-head benchmark set, the 890M wins both tests, and the R9 M375X has zero recorded victories.

Q: What are the architectural generations of these two GPUs?

A: The 890M uses RDNA 3.5 on a 4 nm process, while the R9 M375X uses GCN 1.0 on a 28 nm process.

Q: How do their shading unit counts compare?

A: The 890M has 1,024 shading units, while the R9 M375X has 640, a 60% difference in the 890M’s favor.

Q: What is the average benchmark score for each GPU?

A: The 890M averages 9,210 across all its recorded benchmarks, while the R9 M375X averages 8,325. The 890M ranks at the 45th percentile of all GPUs, and the R9 M375X at the 43rd percentile.