AMD Radeon R5 M230 vs NVIDIA GeForce GTX 970M Comparison

# The Verdict

The NVIDIA GeForce GTX 970M and AMD Radeon R5 M230 occupy entirely different tiers of mobile graphics performance, despite both holding the same 27th percentile ranking among all GPUs. The data is unambiguous: the GTX 970M delivers a 313.9% higher Geekbench OpenCL score than the R5 M230, making it the clear choice for any workload that leverages GPU compute or 3D rendering. The R5 M230, with its 320 shading units and 16.00 GB/s memory bandwidth, is fundamentally a basic IGP-class part, while the GTX 970M operates as a discrete-class mobile GPU with 1280 shading units and 120.3 GB/s bandwidth. For users who need playable frame rates in modern titles or GPU-accelerated productivity, the GTX 970M is the only rational selection from this pairing. Conversely, the R5 M230 is suitable only for lightweight, legacy applications where its lower power footprint and integrated form factor are acceptable trade-offs. The nearest rival data reinforces this gap: the GTX 970M's average benchmark score of 4628 sits within 1.1% of the R5 M230's 4577, but that aggregate figure masks the massive delta in the only head-to-head benchmark available. In that single comparison, the GTX 970M produces 18946 OpenCL points versus 4577 for the R5 M230 — a 4.1x advantage. There are no wins for the R5 M230 in the head-to-head data. The verdict is straightforward: pick the GTX 970M for any serious graphics work, and pick the R5 M230 only if the platform requires an integrated solution with minimal performance expectations.

FAQ

Q: How much faster is the GTX 970M than the R5 M230 in the only benchmark where both are tested?

A: In Geekbench OpenCL, the GTX 970M scores 18946 versus 4577 for the R5 M230, a delta of 313.9% in favor of the NVIDIA part.

Q: Which GPU has better memory bandwidth, and what is the exact difference?

A: The GTX 970M offers 120.3 GB/s of bandwidth using a 192-bit bus with 6 GB of GDDR5, while the R5 M230 provides only 16.00 GB/s over a 64-bit bus with 2 GB of DDR3. That is a 7.5x bandwidth advantage for the GTX 970M.

Q: Do both GPUs support DirectX 12?

A: Yes, but at different feature levels. The GTX 970M supports DirectX 12 (12_1), while the R5 M230 supports DirectX 12 (11_1). The NVIDIA part also lists Vulkan 1.4 support versus Vulkan 1.2.170 for the AMD part.

Q: Which GPU has a higher transistor count and die size?

A: The GTX 970M uses the GM204 chip with 5,200 million transistors on a 398 mm² die, whereas the R5 M230 uses the Jet chip with 690 million transistors on a 56 mm² die. This reflects a 7.5x transistor count difference and a 7.1x die size difference.

Q: What are the pixel and texture fill rates for each GPU?

A: The GTX 970M achieves 49.82 GPixel/s and 83.04 GTexel/s. The R5 M230 achieves 4.880 GPixel/s and 12.20 GTexel/s. The GTX 970M is over 10x faster in pixel throughput and nearly 7x faster in texture throughput.

Q: Which GPU has a higher average benchmark score across all tests?

A: The GTX 970M averages 4628 across all benchmark entries, while the R5 M230 averages 4577. The difference is 1.1% in favor of the GTX 970M, though this aggregate is skewed by the R5 M230 having only one benchmark result.

Architecture Differences

The GTX 970M is built on NVIDIA's Maxwell 2.0 architecture using the GM204 chip, fabricated on a 28 nm process at TSMC. The R5 M230 uses AMD's GCN 1.0 architecture with the Jet chip, also on TSMC's 28 nm node. This is where the similarities end. The GTX 970M packs 5,200 million transistors into a 398 mm² die, yielding a transistor density of 13.1M / mm². The R5 M230 has 690 million transistors on a 56 mm² die, with a density of 12.3M / mm². While densities are comparable, the absolute scale is not: the GTX 970M has roughly 7.5x more transistors.

The compute resources differ dramatically. The GTX 970M features 1280 shading units, 80 texture mapping units (TMUs), and 48 raster operation units (ROPs). The R5 M230 has 320 shading units, 20 TMUs, and 8 ROPs — a 4x reduction in shaders, 4x in TMUs, and 6x in ROPs. Neither GPU includes ray tracing cores or tensor cores, as both predate those technologies.

Architecturally, Maxwell 2.0 was designed for higher instruction-level parallelism and improved power efficiency per clock compared to earlier NVIDIA designs, while GCN 1.0 emphasized throughput-oriented compute with a unified shader design. The GTX 970M's memory subsystem uses a 192-bit interface with GDDR5, while the R5 M230 uses a 64-bit interface with DDR3. The GTX 970M also supports a wider API feature set, including DirectX 12 (12_1) and Vulkan 1.4, versus DirectX 12 (11_1) and Vulkan 1.2.170 for the R5 M230. The GTX 970M is packaged as an MXM Module with an MXM-B (3.0) bus interface, while the R5 M230 is an IGP (integrated graphics processor) using PCIe 3.0 x8. Both are end-of-life products, with the GTX 970M releasing in October 2014 and the R5 M230 in January 2014.

Specification Differences

The two GPUs differ across every major specification field. Process node and foundry are identical (28 nm, TSMC), but transistor count is 5,200 million for the GTX 970M versus 690 million for the R5 M230. Die size is 398 mm² versus 56 mm². The GTX 970M has a base clock of 924 MHz and boost clock of 1038 MHz; the R5 M230 lists no base or boost clock. Memory clocks are 1253 MHz (5 Gbps effective) for the GTX 970M versus 1000 MHz (2 Gbps effective) for the R5 M230. Memory capacity is 6 GB GDDR5 versus 2 GB DDR3. Bus width is 192-bit versus 64-bit. Bandwidth is 120.3 GB/s versus 16.00 GB/s. Shading units: 1280 versus 320. TMUs: 80 versus 20. ROPs: 48 versus 8. Pixel rate is 49.82 GPixel/s versus 4.880 GPixel/s. Texture rate is 83.04 GTexel/s versus 12.20 GTexel/s. FP32 compute is 2.657 TFLOPS versus 390.4 GFLOPS. Neither has a listed TDP. Slot width is MXM Module versus IGP. Power connectors are "None" for the GTX 970M and not listed for the R5 M230. Bus interface is MXM-B (3.0) versus PCIe 3.0 x8. Display outputs are portable-device-dependent for both. DirectX support is 12 (12_1) versus 12 (11_1), OpenGL is 4.6 for both, Vulkan is 1.4 versus 1.2.170. Production status is end-of-life for both. Release dates are October 2014 versus January 2014. Predecessors are GeForce 800M versus Solar System; successors are GeForce 10 Mobile versus Polaris Mobile. Neither has a launch MSRP.

Head-to-Head Benchmarks

The head-to-head data contains exactly one benchmark: Geekbench OpenCL. The GTX 970M scores 18946, while the R5 M230 scores 4577. The delta is 313.9% in favor of the GTX 970M, which wins the only direct comparison. There are zero wins for the R5 M230.

Beyond the single head-to-head test, the broader benchmark suites paint a consistent picture. The GTX 970M has results across ten tests, including 3DMark Steel Nomad DX12 (472), Geekbench Vulkan (18292), Passmark DirectX 10 (28), DirectX 11 (42), DirectX 12 (24), DirectX 9 (99), G2D (381), G3D (5704), and GPU Compute (2289). The R5 M230 has only one benchmark result beyond the OpenCL test — none are listed besides Geekbench OpenCL. This means the R5 M230's entire performance profile rests on a single compute workload, while the GTX 970M demonstrates capability across multiple DirectX versions and general-purpose compute.

The nearest rival data puts the GTX 970M at an average score of 4628, within 0.2% of the NVIDIA Quadro M3000M (4621) and -0.6% of both the AMD Radeon R5 M320 (4657) and AMD Radeon RX 9060 XT 16 GB (4657). The R5 M230 averages 4577, which is 0.2% behind the AMD Radeon RX 560 (4569), 0.4% ahead of the Intel HD Graphics P530 (4560), and -1.1% versus the GTX 970M. The percentile ranking for both is 27th among all GPUs, but that parity is misleading given the 313.9% OpenCL delta. The GTX 970M's FP32 throughput of 2.657 TFLOPS is 6.8x the R5 M230's 390.4 GFLOPS. Pixel fill rate is 10.2x higher, and texture fill rate is 6.8x higher. Memory bandwidth is 7.5x higher.

The largest single win for the GTX 970M is the OpenCL compute benchmark, where it achieves 4.1x the score of the R5 M230. The GTX 970M also holds advantages in every architectural metric: more shading units, more TMUs, more ROPs, more memory, wider bus, and higher clocks. The only category where the R5 M230 is not categorically behind is transistor density, where it trails by only 0.8M / mm² (12.3 versus 13.1). That negligible density difference does not translate into any performance benefit. The R5 M230's smaller die and lower transistor count suggest it was designed for minimal power and cost, not performance. The GTX 970M's larger die and memory subsystem are built for sustained gaming and compute workloads. Every benchmark that exists in the FACT PACK confirms this hierarchy.