AMD Radeon R5 M240 vs AMD Radeon R7 M370 Comparison

The AMD Radeon R5 M240 and AMD Radeon R7 M370 are both end-of-life mobile graphics solutions from AMD's Gem System generation, built on the same GCN 1.0 architecture. The data shows a closely matched pair in raw compute benchmarks, yet the specification sheets reveal significant differences in memory subsystem and execution resources that dictate where each part holds an advantage. The R7 M370 edges ahead in the only shared benchmark, but the R5 M240 counters with higher clock rates and a superior pixel fill rate.

Head-to-Head Benchmarks

The sole direct comparison available is the Geekbench OpenCL test, where the AMD Radeon R7 M370 scores 7063 against the AMD Radeon R5 M240's 6975. That difference translates to a 1.2% deficit for the R5 M240, a margin small enough to be considered within run-to-run variance. The R7 M370 therefore claims the only head-to-head win (1 win vs 0 for the R5 M240), but the benchmark result indicates the two are nearly identical in raw compute throughput.

Placing these scores in context with their nearest rivals clarifies the picture. The R5 M240's 6975 OpenCL score sits 0.4% above the NVIDIA GeForce GTX 675M's 6946 and 2.1% above the AMD FirePro M5100's 6830. However, it trails the NVIDIA GeForce GTX 680M (7023) by 0.7% and the NVIDIA T600 (7035) by 0.8%. The R7 M370's 7063 OpenCL result is 1% higher than the FirePro M5100 and 2.3% higher than the AMD Radeon R7 M460's 6612, but it falls 1% short of the FirePro M5100's average score when considering its overall average benchmark score of 6764, and it is 2.6% behind the GTX 675M.

This clustering within a few percentage points of each other across multiple rival GPUs underscores how competitive this performance tier is. The R7 M370 also has a Geekbench Vulkan score of 6465, which is lower than its OpenCL result, suggesting the architecture's compute performance is more consistently exposed through OpenCL. For the R5 M240, the percentile ranking of 39 versus the R7 M370's 38 indicates that despite the R7's higher raw score, the R5 M240 sits in a marginally better position relative to all other GPUs in the database—a quirk driven by the distribution of scores around these two parts.

Where Each One Wins

The R7 M370 wins decisively in memory-intensive workloads. It pairs a 128-bit memory bus with 2 GB of GDDR5 memory running at 900 MHz (3.6 Gbps effective), yielding a memory bandwidth of 57.60 GB/s. The R5 M240, by contrast, uses a 64-bit bus with 1 GB of DDR3 memory at the same 900 MHz clock (1800 Mbps effective), producing only 14.40 GB/s of bandwidth. That is a 4x difference in favor of the R7 M370, which will manifest in any task that streams large amounts of texture data or geometry, such as higher-resolution gaming or compute kernels with large working sets.

The R5 M240 fights back in clock-sensitive and pixel-throughput-bound scenarios. Its base clock of 1000 MHz and boost clock of 1030 MHz outpace the R7 M370's 875 MHz base and 960 MHz boost. The higher clocks drive a pixel rate of 8.240 GPixel/s for the R5 M240 versus 7.680 GPixel/s for the R7 M370—a 7.3% advantage for the smaller part. This means the R5 M240 can fill a framebuffer with color and depth data faster, which helps in low-resolution, fill-rate-limited situations where memory bandwidth is less of a constraint.

In texture-heavy tasks, the R7 M370 takes the lead. Its 24 texture mapping units (TMUs) and 23.04 GTexel/s texture rate outstrip the R5 M240's 20 TMUs and 20.60 GTexel/s. The R7 M370 also holds a compute advantage with 384 shading units versus 320, producing 737.3 GFLOPS of FP32 performance against 659.2 GFLOPS for the R5 M240—an 11.8% higher theoretical compute ceiling.

Architecture Differences

Both GPUs are built on GCN 1.0 architecture and manufactured by TSMC on a 28 nm process node, with identical transistor densities of 12.3M per mm². The underlying design philosophy is the same: a unified shader array with vector processors optimized for parallel throughput. The difference lies in the physical implementation. The R5 M240 uses the Jet chip, which packs 690 million transistors into a 56 mm² die. The R7 M370 uses the Litho chip, a larger 77 mm² die containing 950 million transistors—a 37.7% increase in transistor count and a 37.5% increase in die area.

This larger die allows the R7 M370 to implement 384 shading units, 24 TMUs, and 8 ROPs, while the R5 M240 is configured with 320 shading units, 20 TMUs, and 8 ROPs. The ROP count is identical, which explains why the pixel rate advantage goes to the R5 M240 purely on clock speed. The extra shading units and TMUs in the R7 M370 are the primary architectural differentiator beyond the memory controller.

Both parts expose the same API feature set: DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Neither supports hardware ray tracing or tensor cores, as those fields are null in the data. The bus interface is identical at PCIe 3.0 x8, which means neither GPU is likely to be bottlenecked by the host connection in typical mobile workloads. The memory controller design, however, is a stark contrast: the R7 M370's 128-bit GDDR5 interface versus the R5 M240's 64-bit DDR3 interface is the single most consequential architectural divergence between the two.

Specification Differences

The two GPUs differ across several specification fields, and those differences define their respective performance profiles. Clock speeds are the first divergence: the R5 M240 runs at 1000 MHz base and 1030 MHz boost, while the R7 M370 is clocked lower at 875 MHz base and 960 MHz boost. Memory clocks are expressed differently in the data—both run at 900 MHz, but the R5 M240's effective data rate is 1800 Mbps while the R7 M370's is 3.6 Gbps, reflecting the different memory types.

Memory capacity and type are the most pronounced differences. The R5 M240 offers 1024 MB of DDR3 on a 64-bit bus, yielding 14.40 GB/s of bandwidth. The R7 M370 doubles capacity to 2 GB of GDDR5 on a 128-bit bus, quadrupling bandwidth to 57.60 GB/s. The execution resources also differ: the R7 M370 has 384 shading units and 24 TMUs versus the R5 M240's 320 shading units and 20 TMUs. ROPs are equal at 8.

Derived throughput figures reflect these core counts. The R5 M240 achieves 8.240 GPixel/s pixel rate and 20.60 GTexel/s texture rate, while the R7 M370 manages 7.680 GPixel/s and 23.04 GTexel/s. Floating-point performance favors the R7 M370 at 737.3 GFLOPS versus 659.2 GFLOPS. The production status is end-of-life for both, but the release dates differ: the R5 M240 launched on September 17, 2014, and the R7 M370 followed on May 4, 2015. Both share the same predecessor (Solar System) and successor (Polaris Mobile), and neither has a launch MSRP listed.

FAQ

Q: Which GPU is faster in the Geekbench OpenCL benchmark?

A: The AMD Radeon R7 M370 scores 7063, which is 1.2% higher than the AMD Radeon R5 M240's 6975.

Q: How does the R5 M240 compare to the NVIDIA GeForce GTX 675M?

A: The R5 M240's OpenCL score of 6975 is 0.4% higher than the GTX 675M's average score of 6946.

Q: What is the memory bandwidth difference between the two GPUs?

A: The R7 M370 has a memory bandwidth of 57.60 GB/s using 2 GB of GDDR5 on a 128-bit bus, while the R5 M240 has 14.40 GB/s using 1 GB of DDR3 on a 64-bit bus.

Q: Does the R5 M240 have any performance advantage over the R7 M370?

A: Yes, the R5 M240 has a higher pixel rate of 8.240 GPixel/s versus 7.680 GPixel/s for the R7 M370, driven by its higher base clock of 1000 MHz and boost clock of 1030 MHz.

Q: Which GPU has more shading units?

A: The R7 M370 has 384 shading units, compared to 320 for the R5 M240, giving it an 11.8% higher FP32 compute throughput of 737.3 GFLOPS.

Q: Are both GPUs based on the same architecture?

A: Yes, both are built on GCN 1.0 architecture using a 28 nm TSMC process, but the R5 M240 uses the Jet chip (690 million transistors, 56 mm²) while the R7 M370 uses the Litho chip (950 million transistors, 77 mm²).