AMD Radeon R5 M335 vs AMD Radeon R7 M260 Comparison

The AMD Radeon R5 M335 and AMD Radeon R7 M260 are both end-of-life mobile graphics solutions from AMD’s Gem System generation, but they target different performance profiles. The data shows a split decision across the two available benchmark suites: the R5 M335 wins the OpenCL test by a substantial 28% margin, while the R7 M260 counters with a 10% victory in the Vulkan test. This creates a nuanced picture where the "better" GPU depends entirely on the workload and API in question.

Head-to-Head Benchmarks

The most striking result in the data is the R5 M335’s dominance in the Geekbench OpenCL test. It scores 4,745 points against the R7 M260’s 3,708 points, a 28% advantage. This is a decisive win for the older GCN 1.0 architecture in a compute-heavy, general-purpose workload. The R5 M335’s score places it in the 28th percentile of all GPUs, while the R7 M260 sits at the 26th percentile, confirming that the OpenCL result is not an outlier but reflects a real performance gap in this specific test.

However, the tables turn completely in the Geekbench Vulkan test. Here, the R7 M260 scores 5,289 points, outpacing the R5 M335’s 4,758 points by 10%. This result is particularly interesting because Vulkan is a lower-level API that often favors architectures with more modern feature sets. The R7 M260’s GCN 3.0 architecture and higher base clock of 940 MHz (versus the R5 M335’s unspecified base clock) likely contribute to this advantage. The R7 M260’s Vulkan score is its stronger benchmark, while the R5 M335’s two scores are remarkably consistent at 4,745 and 4,758.

Looking at the average benchmark scores, the R5 M335 holds a slight edge at 4,752 points versus the R7 M260’s 4,499 points. This 5.6% difference in the overall average is smaller than the individual OpenCL delta suggests, because the R7 M260’s strong Vulkan performance pulls its average up. The nearest rival data provides additional context: the R5 M335 is most closely matched with the AMD Radeon R8 M445DX (0.5% ahead) and the AMD Radeon R5 M255 (0.7% behind), while the R7 M260 sits near the AMD FirePro W4190M (0.1% behind) and Intel HD Graphics P530 (1.3% behind). Neither GPU is a class leader; both are firmly in the entry-level mobile segment.

Architecture Differences

The two GPUs come from different architectural generations despite sharing the same 28 nm TSMC process node. The R5 M335 uses the "Exo" chip based on GCN 1.0 architecture, while the R7 M260 uses the "Topaz" chip with GCN 3.0 architecture. This generational leap is significant: GCN 3.0 introduced improvements in geometry processing and power management that GCN 1.0 lacks.

The transistor counts tell a dramatic story of silicon complexity. The R7 M260 packs 1,550 million transistors on a 125 mm² die, while the R5 M335 contains just 690 million transistors on a 56 mm² die. This means the R7 M260 has more than double the transistor count and a die that is over twice the size. Interestingly, the transistor density is nearly identical at 12.4M / mm² for the R7 M260 and 12.3M / mm² for the R5 M335, confirming that the size difference comes from more hardware, not denser packaging.

The compute resources differ accordingly. The R7 M260 features 384 shading units, 24 texture mapping units (TMUs), and 8 raster operation units (ROPs). The R5 M335 counters with 320 shading units, 20 TMUs, and 8 ROPs. This gives the R7 M260 a 20% advantage in shading units and TMUs, which explains its higher texture fill rate of 23.52 GTexel/s versus the R5 M335’s 20.60 GTexel/s. However, the R5 M335 achieves a higher pixel rate of 8.240 GPixel/s compared to the R7 M260’s 7.840 GPixel/s, despite having the same number of ROPs.

The FP32 compute performance reflects the raw shader advantage: the R7 M260 delivers 752.6 GFLOPS, while the R5 M335 manages 659.2 GFLOPS. The R7 M260 also supports FP16 at the same 752.6 GFLOPS rate, while the R5 M335 has no listed FP16 capability. The R7 M260’s base clock of 940 MHz and boost clock of 980 MHz provide additional context, though the R5 M335’s clocks are not specified in the data.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R5 M335 has an average benchmark score of 4,752, which is 5.6% higher than the R7 M260’s 4,499 points.

Q: Does the R7 M260 win any benchmark tests?

A: Yes, the R7 M260 wins the Geekbench Vulkan test with a score of 5,289, beating the R5 M335’s 4,758 by 10%.

Q: What is the main architectural difference between these two GPUs?

A: The R5 M335 uses GCN 1.0 architecture with an Exo chip, while the R7 M260 uses GCN 3.0 architecture with a Topaz chip. The R7 M260 also has more than double the transistor count (1,550 million versus 690 million).

Q: Do both GPUs have the same memory configuration?

A: Yes, both feature 2 GB of DDR3 memory on a 64-bit bus with 14.40 GB/s bandwidth and 900 MHz memory clock (1800 Mbps effective).

Q: Which GPU has more shading units?

A: The R7 M260 has 384 shading units, while the R5 M335 has 320, giving the R7 M260 a 20% advantage in this specification.

Q: Are these GPUs still in production?

A: No, both are listed as end-of-life products. The R5 M335 was released on 2015-10-20, and the R7 M260 was released on 2014-06-10.

Specification Differences

The following specifications differ between the two GPUs:

• Chip: R5 M335 uses "Exo"; R7 M260 uses "Topaz"
• Architecture: R5 M335 is GCN 1.0; R7 M260 is GCN 3.0
• Generation: R5 M335 is from Gem System (R5 M300); R7 M260 is from Gem System (R7 M200)
• Transistors: R5 M335 has 690 million; R7 M260 has 1,550 million
• Die Size: R5 M335 is 56 mm²; R7 M260 is 125 mm²
• Base Clock: R5 M335 is unspecified; R7 M260 is 940 MHz
• Boost Clock: R5 M335 is unspecified; R7 M260 is 980 MHz
• Shading Units: R5 M335 has 320; R7 M260 has 384
• TMUs: R5 M335 has 20; R7 M260 has 24
• Pixel Rate: R5 M335 is 8.240 GPixel/s; R7 M260 is 7.840 GPixel/s
• Texture Rate: R5 M335 is 20.60 GTexel/s; R7 M260 is 23.52 GTexel/s
• FP32: R5 M335 is 659.2 GFLOPS; R7 M260 is 752.6 GFLOPS
• FP16: R5 M335 is not listed; R7 M260 is 752.6 GFLOPS (1:1)
• DirectX Support: R5 M335 supports 12 (11_1); R7 M260 supports 12 (12_0)
• Display Outputs: R5 M335 is "Portable Device Dependent"; R7 M260 is not specified
• Power Connectors: R5 M335 is "None"; R7 M260 is not specified
• Release Date: R5 M335 is 2015-10-20; R7 M260 is 2014-06-10

The Verdict

The benchmark data presents a clear but divided verdict. For users prioritizing OpenCL compute workloads, the AMD Radeon R5 M335 is the obvious choice, delivering a 28% performance advantage over its rival. Its consistent scores across both tests (4,745 and 4,758) suggest stable, predictable behavior. The R5 M335 also holds the higher average benchmark score at 4,752, making it the safer pick for general-purpose tasks where OpenCL is the primary API.

For Vulkan-based applications, the AMD Radeon R7 M260 is superior, with a 10% lead that demonstrates the benefits of its newer GCN 3.0 architecture and additional shading units. Its 384 shading units and 752.6 GFLOPS of FP32 compute power provide a solid foundation for modern graphics APIs. The R7 M260’s higher texture rate of 23.52 GTexel/s also makes it more capable in texture-heavy scenarios.

The R5 M335’s higher pixel rate of 8.240 GPixel/s versus the R7 M260’s 7.840 GPixel/s is a small but notable advantage in fill-rate-bound scenarios. However, this does not compensate for the R7 M260’s broader compute superiority in most other measured specifications. The R7 M260’s transistor budget of 1,550 million versus 690 million suggests more headroom for complex workloads, even if the benchmark results do not always reflect this in practice.

Ultimately, the data supports a workload-based recommendation. Users running OpenCL-centric applications should select the R5 M335. Users who prioritize Vulkan performance or need the higher texture and compute throughput should select the R7 M260. Neither GPU offers a universal advantage, and the 10% to 28% swings between tests make the choice highly dependent on the software environment. Given the end-of-life status of both products, the decision should be based on the specific applications and APIs in use rather than any general notion of superiority.