AMD Radeon R5 M335 vs AMD Radeon R7 M445 Comparison

Where Each One Wins

The recorded benchmark data presents a clear, though narrow, picture. The AMD Radeon R7 M445 wins the only direct head-to-head comparison available, the Geekbench OpenCL test, by a significant margin. It scores 5358 points against the AMD Radeon R5 M335's 4745 points, a delta of 12.9%. This makes the R7 M445 the definitive choice for compute-heavy workloads that leverage OpenCL, such as general-purpose GPU tasks or content creation acceleration.

The AMD Radeon R5 M335, however, is not without its own data point. It has a second recorded benchmark, a Geekbench Vulkan test, where it scores 4758 points. This is the only benchmark in the database where the R5 M335 has a score and the R7 M445 does not. The database does not contain a Vulkan score for the R7 M445, so the R5 M335 holds an unopposed win in this specific API test. This suggests that for applications relying on the Vulkan graphics API, the R5 M335 has a verified performance capability, whereas the R7 M445's performance in this area is unmeasured.

Looking at the broader database percentiles, the R7 M445 sits at the 31st percentile of all GPUs, while the R5 M335 sits at the 28th percentile. This three-point gap reinforces the notion that the R7 M445 is the faster part overall, even if both are positioned in the lower performance tiers of the database. The R5 M335's average benchmark score is 4752, which is lower than the R7 M445's single recorded score of 5358.

In summary, the R7 M445 wins where it matters most in the recorded data: the OpenCL compute test. The R5 M335's only victory is a technical one, being the sole entrant in the Vulkan test. For users prioritizing raw compute performance, the R7 M445 is the clear winner. For those specifically requiring Vulkan API performance, the R5 M335 has the only verified data.

Architecture Differences

The two GPUs, while both from AMD, represent different generations of the Graphics Core Next architecture. The AMD Radeon R7 M445 is built on GCN 3.0, while the AMD Radeon R5 M335 uses the older GCN 1.0. This architectural gap is significant, as GCN 3.0 brought refinements in instruction scheduling and power management over the first-generation design.

The underlying chips also differ. The R7 M445 uses a chip codenamed "Meso," while the R5 M335 uses "Exo." This is reflected in their physical characteristics. The Meso chip is substantially larger, with a die size of 125 mm², and contains 1,550 million transistors. The Exo chip is much smaller, at 56 mm², with 690 million transistors. Interestingly, the transistor density is nearly identical: 12.4 million per mm² for the Meso and 12.3 million per mm² for the Exo. This implies the larger chip is essentially a scaled-up version with more functional units, not a fundamentally denser design.

Both are fabricated on the same 28 nm process at TSMC. The R7 M445 belongs to the "Gem System" generation for the R7 M400 series, while the R5 M335 is from the "Gem System" generation for the R5 M300 series. Both list their predecessor as "Solar System" and successor as "Polaris Mobile."

The feature sets also diverge in terms of API support. The R7 M445 supports DirectX 12 (12_0), a full-featured implementation. The R5 M335 only supports DirectX 12 (11_1), which lacks certain features of the 12_0 feature level. Both support OpenGL 4.6 and Vulkan 1.2.170.

Memory technology is a key differentiator. The R7 M445 is paired with 4 GB of GDDR5 memory, while the R5 M335 uses 2 GB of DDR3. This is a major architectural advantage for the R7 M445, as GDDR5 offers significantly higher bandwidth. The R7 M445's memory runs at 1000 MHz (4 Gbps effective), while the R5 M335's runs at 900 MHz (1800 Mbps effective). The result is a bandwidth of 32.00 GB/s for the R7 M445 versus just 14.40 GB/s for the R5 M335.

Interestingly, despite these differences, the core compute configuration is identical. Both have 320 shading units, 20 texture mapping units (TMUs), and 8 render output units (ROPs). The R5 M335 has a higher theoretical fillrate, with a pixel rate of 8.240 GPixel/s and texture rate of 20.60 GTexel/s, compared to the R7 M445's 7.360 GPixel/s and 18.40 GTexel/s. Similarly, the R5 M335 has a higher FP32 throughput at 659.2 GFLOPS versus 588.8 GFLOPS for the R7 M445. The R7 M445 also has a 1:1 FP16 rate of 588.8 GFLOPS, while the R5 M335 has no recorded FP16 data.

FAQ

Q: Which GPU has higher memory bandwidth?

A: The AMD Radeon R7 M445, with a bandwidth of 32.00 GB/s, more than double the R5 M335's 14.40 GB/s.

Q: What are the core counts for each GPU?

A: Both the AMD Radeon R7 M445 and the AMD Radeon R5 M335 have identical core configurations: 320 shading units, 20 TMUs, and 8 ROPs.

Q: Which GPU supports a higher DirectX version?

A: The AMD Radeon R7 M445 supports DirectX 12 (12_0), while the AMD Radeon R5 M335 supports DirectX 12 (11_1).

Q: What is the difference in their process nodes?

A: There is no difference; both are fabricated on a 28 nm process at TSMC.

Q: Which GPU has a higher average benchmark score?

A: The AMD Radeon R7 M445 has an average benchmark score of 5358, compared to the AMD Radeon R5 M335's average of 4752.

Q: Does the R5 M335 have any benchmark where it outperforms the R7 M445?

A: The R5 M335 has a recorded Geekbench Vulkan score of 4758. The R7 M445 has no recorded Vulkan score, so this is an unopposed data point for the R5 M335.

Specification Differences

The following specifications differ between the two GPUs:

• Architecture: GCN 3.0 (R7 M445) vs GCN 1.0 (R5 M335)
• Chip: Meso (R7 M445) vs Exo (R5 M335)
• Generation: Gem System (R7 M400) (R7 M445) vs Gem System (R5 M300) (R5 M335)
• Transistors: 1,550 million (R7 M445) vs 690 million (R5 M335)
• Die Size: 125 mm² (R7 M445) vs 56 mm² (R5 M335)
• Memory Size: 4 GB (R7 M445) vs 2 GB (R5 M335)
• Memory Type: GDDR5 (R7 M445) vs DDR3 (R5 M335)
• Memory Clock: 1000 MHz, 4 Gbps effective (R7 M445) vs 900 MHz, 1800 Mbps effective (R5 M335)
• Memory Bandwidth: 32.00 GB/s (R7 M445) vs 14.40 GB/s (R5 M335)
• Pixel Rate: 7.360 GPixel/s (R7 M445) vs 8.240 GPixel/s (R5 M335)
• Texture Rate: 18.40 GTexel/s (R7 M445) vs 20.60 GTexel/s (R5 M335)
• FP32: 588.8 GFLOPS (R7 M445) vs 659.2 GFLOPS (R5 M335)
• FP16: 588.8 GFLOPS (1:1) (R7 M445) vs null (R5 M335)
• DirectX: 12 (12_0) (R7 M445) vs 12 (11_1) (R5 M335)
• Release Date: 2016-05-14 (R7 M445) vs 2015-10-20 (R5 M335)

Head-to-Head Benchmarks

The database contains one direct head-to-head comparison: the Geekbench OpenCL test. In this test, the AMD Radeon R7 M445 scores 5358, while the AMD Radeon R5 M335 scores 4745. The R7 M445 wins by 12.9%. This is a substantial margin in a compute benchmark, indicating that the R7 M445 is significantly more capable in OpenCL workloads.

This result is interesting when compared to the theoretical specifications. The R5 M335 actually has a higher FP32 throughput (659.2 GFLOPS) and higher texture fillrate (20.60 GTexel/s) than the R7 M445 (588.8 GFLOPS and 18.40 GTexel/s). Yet, in the real-world OpenCL test, the R7 M445 comes out ahead by nearly 13%. This suggests that the R7 M445's architectural advantages, particularly its GDDR5 memory with 32.00 GB/s bandwidth versus the R5 M335's DDR3 at 14.40 GB/s, play a more critical role in this workload than raw clock-speed-derived throughput. The newer GCN 3.0 architecture likely also contributes to better instruction efficiency.

Looking at the nearest rivals for each GPU provides additional context. The R7 M445's closest rival is the Intel UHD Graphics P630, which scores 5370, a delta of -0.2% (the R7 M445 is slightly behind). It also sits near the NVIDIA GeForce 840M (5322, delta 0.7% favoring the R7 M445) and the NVIDIA GeForce 930A (5317, delta 0.8% favoring the R7 M445). The R7 M445 even edges out the NVIDIA GeForce GTX 980M (5308, delta 0.9% favoring the R7 M445).

The R5 M335's nearest rivals include the AMD Radeon R8 M445DX (4727, delta 0.5% favoring the R5 M335) and the AMD Radeon R5 M255 (4788, delta -0.7% favoring the R5 M255). It also sits near the NVIDIA Quadro P400 (4684, delta 1.5% favoring the R5 M335). A notable rival in its list is the NVIDIA GeForce RTX 3080 12 GB (4791, delta -0.8% favoring the RTX 3080), which shows how the percentile ranking can place very different classes of GPUs near each other in a specific compute test.

The Verdict

The data points to a single clear conclusion: the AMD Radeon R7 M445 is the superior GPU for compute performance. Its 12.9% lead in the Geekbench OpenCL test is decisive, despite the R5 M335 having higher theoretical fillrate and FP32 figures. The R7 M445's advantage likely stems from its newer GCN 3.0 architecture and significantly faster GDDR5 memory, which provides more than double the bandwidth of the R5 M335's DDR3.

Users should choose the AMD Radeon R7 M445 if their priority is OpenCL compute performance, as it is the only GPU of the two with a verified win in a head-to-head benchmark. It also offers double the memory capacity (4 GB vs 2 GB) and a more capable DirectX 12 (12_0) implementation.

The AMD Radeon R5 M335 holds a single advantage in the recorded data: it has a verified Geekbench Vulkan score of 4758, while the R7 M445 has no recorded Vulkan score. For a user specifically targeting Vulkan-based applications, the R5 M335 at least has a measured performance level, whereas the R7 M445's Vulkan performance is unknown. However, this is a weak basis for a recommendation, as it represents an absence of data rather than a performance win. Given the R7 M445's overall architectural superiority in memory bandwidth and its decisive OpenCL victory, the R5 M335 is only justifiable in the narrow scenario where Vulkan support is a hard requirement and no other data exists to compare.