AMD Radeon 890M vs AMD Radeon R9 M375 Comparison

AMD Radeon R9 M375 and AMD Radeon 890M represent two very different eras of AMD mobile graphics, separated by nearly a decade of architectural evolution. The R9 M375 is a 2015-era discrete part built on GCN 1.0, while the 890M is a 2024 integrated GPU based on RDNA 3.5. The benchmark data shows a decisive performance gap, but the story is more nuanced than a simple win/loss tally. This analysis walks through where each GPU holds an advantage, what the numbers mean in context, and who should consider which part based strictly on the provided specifications and test results.

Where Each One Wins

The Radeon R9 M375 wins in exactly zero of the head-to-head benchmark comparisons available. The data shows the 890M taking both shared tests outright. However, the R9 M375 does hold a slight edge in one specific metric: its average benchmark score across all recorded tests is 10070, which is higher than the 890M’s 9210 average. This is a curious inversion — the older discrete GPU averages higher across its limited test set, while the newer integrated part dominates the specific shared workloads. The R9 M375’s percentile ranking (48th) is also marginally better than the 890M’s (45th), meaning it sits slightly higher relative to all other GPUs in the database.

The 890M wins decisively in raw compute and graphics throughput. Its shading unit count (1024 vs 640) and texture mapping units (64 vs 40) are substantially higher. The pixel rate tells a similar story: 92.80 GPixel/s versus 16.24 GPixel/s. The 890M also has a massive clock advantage, boosting to 2900 MHz compared to the R9 M375’s 1015 MHz boost. For FP32 compute, the 890M delivers 5.939 TFLOPS against the R9 M375’s 1,299.2 GFLOPS. In practical terms, this means the 890M is built for modern workloads — its 16 ray tracing cores and DirectX 12 Ultimate support are entirely absent from the older part.

The Verdict

The data points to a clear split in use cases. For anyone running modern DirectX 12 Ultimate titles, ray-traced content, or Vulkan 1.4 applications, the Radeon 890M is the only viable option. Its Geekbench Vulkan score of 40808 crushes the R9 M375’s 9682, a delta of 76.3%. Similarly, OpenCL compute workloads favor the 890M by 71.9%, with scores of 37254 versus 10457. The 890M is also the only part with a production status of “Active,” meaning it’s currently available in systems, while the R9 M375 is marked “End-of-life.”

The R9 M375 is not without merit, but its strengths are relative and narrow. Its average benchmark score of 10070 edges out the 890M’s 9210, and its nearest rivals in the database — the NVIDIA Quadro K5100M (10043, +0.3% delta) and AMD Radeon Pro 5300M (10013, +0.6%) — are all within a 1.1% margin. This suggests the R9 M375 sits in a stable performance tier for legacy 1080p gaming or older API workloads. However, its 2 GB DDR3 memory with 28.80 GB/s bandwidth is a severe bottleneck for anything modern. The 890M uses system-shared memory, which is dependent on the host platform, but its RDNA 3.5 architecture is designed to handle far more demanding tasks.

Head-to-Head Benchmarks

The two shared benchmarks provide the clearest picture of the performance gap. In Geekbench OpenCL, the 890M scores 37254 against the R9 M375’s 10457. This 71.9% delta is massive and reflects the fundamental architectural differences: the 890M’s 1024 shading units, 64 TMUs, and 32 ROPs are all double or nearly double the R9 M375’s counts. The clock speed difference compounds this — the 890M boosts to 2900 MHz, nearly three times the R9 M375’s 1015 MHz. The result is a compute workload that the older part simply cannot keep pace with.

Geekbench Vulkan shows an even wider margin. The 890M posts 40808, while the R9 M375 manages 9682, a 76.3% deficit for the older GPU. Vulkan is a modern low-level API, and the R9 M375’s support stops at version 1.2.170, whereas the 890M supports Vulkan 1.4. The 890M’s 16 ray tracing cores and RDNA 3.5 architecture are clearly better suited to the API’s advanced features. Interestingly, the R9 M375’s nearest rivals in the database — the GTX 870M (9959, +1.1%) and GTX 950A (10273, -2%) — are within a narrow band, indicating that its Vulkan score of 9682 is on par with older discrete parts, but that band is nowhere near the 890M’s performance.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R9 M375 has an average benchmark score of 10070, which is higher than the AMD Radeon 890M’s 9210. However, the 890M wins both shared head-to-head tests.

Q: Does the Radeon 890M support ray tracing?

A: Yes, the Radeon 890M includes 16 ray tracing cores, a feature entirely absent from the R9 M375. This is part of its RDNA 3.5 architecture.

Q: What is the biggest performance difference between the two?

A: In Geekbench Vulkan, the 890M scores 40808 versus 9682 for the R9 M375, a 76.3% difference. The OpenCL gap is 71.9%, with scores of 37254 and 10457 respectively.

Q: Which GPU is currently in production?

A: The AMD Radeon 890M is marked as “Active” in production status, while the AMD Radeon R9 M375 is listed as “End-of-life.”

Q: How do the memory configurations differ?

A: The R9 M375 has 2 GB of dedicated DDR3 memory with a 128-bit bus and 28.80 GB/s bandwidth. The 890M uses system-shared memory, with size, type, bus width, and bandwidth all listed as dependent on the system.

Q: Which part has better API support?

A: The Radeon 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the R9 M375 supports DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6.

Architecture Differences

The architectural gap between these two GPUs is generational. The R9 M375 uses the Tropo chip based on GCN 1.0, built on a 28 nm process at TSMC. It packs 1,500 million transistors on a 123 mm² die, yielding a transistor density of 12.2 million per mm². The 890M uses the Strix Point chip with RDNA 3.5, fabricated on a 4 nm process, also at TSMC. It contains 34,000 million transistors on a 233 mm² die, for a density of 145.9 million per mm² — a 12x increase in density.

The 890M’s RDNA 3.5 architecture brings features that GCN 1.0 cannot match. It has 16 dedicated ray tracing cores, while the R9 M375 has none. The 890M also supports FP16 compute at a 1:1 ratio with FP32 (both 5.939 TFLOPS), whereas the R9 M375 has no listed FP16 capability. The 890M is an integrated GPU (IGP) with a 15 W TDP and no power connectors, while the R9 M375 is a discrete part with no TDP listed. The 890M’s slot width is listed as “IGP,” and its display outputs are “Portable Device Dependent,” reflecting its mobile integrated nature.

Specification Differences

The specification table shows several key differences beyond architecture. The 890M has more of everything that matters for compute: 1024 shading units versus 640, 64 TMUs versus 40, and 32 ROPs versus 16. Clock speeds favor the 890M dramatically, with a 2900 MHz boost versus the R9 M375’s 1015 MHz. The 890M’s pixel rate is 92.80 GPixel/s versus 16.24 GPixel/s, and its texture rate is 185.6 GTexel/s versus 40.60 GTexel/s. FP32 performance is 5.939 TFLOPS versus 1,299.2 GFLOPS.

Memory is another clear differentiator. The R9 M375 uses 2 GB of DDR3 on a 128-bit bus, with 28.80 GB/s bandwidth. The 890M uses system-shared memory, with all memory specifications listed as “System Shared” or “System Dependent.” The bus interface also differs: the 890M uses PCIe 4.0 x8, while the R9 M375 uses PCIe 3.0 x16. The 890M has a 15 W TDP and requires no power connectors; the R9 M375 has no TDP or power connector data. Finally, the 890M supports DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the R9 M375 is limited to DirectX 12 (11_1) and Vulkan 1.2.170. The release dates tell the story: the R9 M375 launched in May 2015, and the 890M arrived in July 2024.