AMD Radeon 610M vs AMD Radeon R5 M335 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 610M has an average benchmark score of 5444, while the AMD Radeon R5 M335 scores 4752. This places the 610M in the 32nd percentile of all GPUs, versus the 28th percentile for the R5 M335.

Q: How do the two GPUs compare in OpenCL performance?

A: In the Geekbench OpenCL test, the AMD Radeon R5 M335 wins with a score of 4745, which is 4.4% higher than the Radeon 610M's 4535. This is the only benchmark where the older card comes out ahead.

Q: What about Vulkan performance?

A: The AMD Radeon 610M dominates in the Geekbench Vulkan test with a score of 6353, a 33.5% advantage over the R5 M335's 4758. This single result drives most of the 610M's overall lead.

Q: What are the architecture generations of each GPU?

A: The Radeon 610M uses RDNA 2.0 architecture on a 6 nm process (chip: Mendocino), while the Radeon R5 M335 uses GCN 1.0 architecture on a 28 nm process (chip: Exo). The 610M is from the Navi II IGP generation; the R5 M335 is from the Gem System (R5 M300) generation.

Q: What memory configurations do these GPUs use?

A: The Radeon 610M uses system shared memory with system-dependent bandwidth, while the Radeon R5 M335 has 2 GB of dedicated DDR3 memory on a 64-bit bus with 14.40 GB/s bandwidth and 1800 Mbps effective memory speed.

Q: Which GPU supports newer graphics APIs?

A: The Radeon 610M supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Radeon R5 M335 supports DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. The 610M has the newer DirectX and Vulkan feature sets.

Architecture Differences

The two GPUs come from completely different eras of AMD's design philosophy. The Radeon 610M is built on the RDNA 2.0 architecture, using the Mendocino chip manufactured on TSMC's 6 nm process. The die size measures 100 mm². In contrast, the Radeon R5 M335 uses the much older GCN 1.0 architecture, with the Exo chip built on a 28 nm process. Its die is 56 mm² and packs 690 million transistors at a density of 12.3M per mm².

The core configurations show an interesting inversion. The R5 M335 has significantly more raw compute units: 320 shading units, 20 texture mapping units, and 8 raster output units. The Radeon 610M has only 128 shading units, 8 TMUs, and 4 ROPs. However, the 610M includes 2 ray tracing cores, a feature entirely absent from the R5 M335. Despite fewer shading units, the 610M's RDNA 2.0 design achieves a pixel rate of 7.600 GPixel/s and a texture rate of 15.20 GTexel/s, while the R5 M335 posts 8.240 GPixel/s and 20.60 GTexel/s.

Clock behavior differs fundamentally. The Radeon 610M has defined base and boost clocks of 1500 MHz and 1900 MHz respectively, and its memory clock is described as system shared. The R5 M335 has no listed base or boost clock, but its memory runs at 900 MHz or 1800 Mbps effective. The 610M's FP32 throughput is 486.4 GFLOPS, while the R5 M335 reaches 659.2 GFLOPS. The 610M also lists FP16 performance at 972.8 GFLOPS (2:1), a capability the R5 M335 does not document.

Interface and power characteristics further separate the two. The Radeon 610M uses PCIe 4.0 x8 and has a 15 W TDP with no power connectors, typical of an integrated graphics processor. The R5 M335 uses PCIe 3.0 x8 and also has no power connectors, but its TDP is not recorded. The 610M is designated as an IGP in terms of slot width, whereas the R5 M335 has no slot width listed. Both are end-of-life products, with the 610M released in September 2022 and the R5 M335 in October 2015. The 610M's predecessor is listed as Vega II IGP and its successor as Navi III IGP; the R5 M335's predecessor is Solar System and its successor is Polaris Mobile.

The Verdict

The benchmark data supports a clear but nuanced conclusion. The Radeon 610M is the stronger overall GPU, with an average score of 5444 versus 4752 for the R5 M335, a lead of roughly 14.6%. Its Vulkan advantage of 33.5% is decisive for modern workloads that leverage this API. The 610M also brings a newer architecture, a smaller process node, ray tracing support, and a more recent DirectX feature level.

However, the R5 M335 is not without its own territory. In OpenCL, it beats the 610M by 4.4%, and it has substantially higher raw shading unit counts, FP32 throughput, and dedicated memory. Users whose applications rely heavily on OpenCL compute may find the older card adequate, especially given its dedicated 2 GB VRAM versus the 610M's system-shared memory.

For most general-purpose use, particularly gaming or applications that support Vulkan, the Radeon 610M is the better choice based on the recorded data. The R5 M335 is a niche pick for legacy OpenCL workloads where its compute throughput and dedicated memory give it an edge. The 610M's nearest rivals include the NVIDIA Quadro M4000 (delta -0.4%), AMD Radeon R7 M440 (delta -0.7%), and NVIDIA GeForce GTX 765M (delta -1%), all within 1% of its average score. The R5 M335's closest competitors are the AMD Radeon R8 M445DX (delta 0.5%), AMD Radeon R5 M255 (delta -0.7%), and NVIDIA Quadro P400 (delta 1.5%).

Specification Differences

The two GPUs differ across nearly every recorded specification category. Process technology: the 610M uses 6 nm TSMC, the R5 M335 uses 28 nm TSMC. Die size: 100 mm² versus 56 mm². The R5 M335 has a documented transistor count of 690 million and density of 12.3M per mm², while the 610M has no transistor count or density recorded.

Core counts diverge sharply. The R5 M335 has 320 shading units, 20 TMUs, and 8 ROPs, while the 610M has 128 shading units, 8 TMUs, and 4 ROPs. The 610M adds 2 ray tracing cores; the R5 M335 has none. Clock speeds: the 610M lists base 1500 MHz and boost 1900 MHz, with system shared memory; the R5 M335 has no base or boost clock recorded, and its memory runs at 900 MHz or 1800 Mbps effective.

Memory configuration is a major split. The 610M uses system shared memory with system-dependent bandwidth; the R5 M335 has 2 GB of DDR3 on a 64-bit bus with 14.40 GB/s bandwidth. Pixel and texture rates favor the R5 M335: 8.240 GPixel/s versus 7.600 GPixel/s, and 20.60 GTexel/s versus 15.20 GTexel/s. FP32 throughput also favors the R5 M335 at 659.2 GFLOPS versus 486.4 GFLOPS. The 610M has FP16 at 972.8 GFLOPS (2:1); the R5 M335 has no FP16 figure.

Power and interface: the 610M has a 15 W TDP and is classified as an IGP, while the R5 M335 has no TDP or slot width recorded. Both use PCIe x8, but the 610M uses PCIe 4.0 and the R5 M335 uses PCIe 3.0. API support: the 610M has DirectX 12 Ultimate (12_2) and Vulkan 1.4; the R5 M335 has DirectX 12 (11_1) and Vulkan 1.2.170. Both support OpenGL 4.6. Release dates: September 2022 for the 610M, October 2015 for the R5 M335.

Head-to-Head Benchmarks

Two benchmark tests were recorded for this comparison. The results split evenly, one win for each GPU, but the magnitude of the victories is far from equal.

In Geekbench OpenCL, the Radeon R5 M335 scores 4745 against the Radeon 610M's 4535. This is a 4.4% margin in favor of the older card. The R5 M335's higher shading unit count and FP32 throughput, combined with its dedicated 2 GB DDR3 memory, likely explain this result. The 610M's system-shared memory may introduce latency that OpenCL compute workloads are sensitive to. This win is modest in percentage terms, but it is the only recorded benchmark where the R5 M335 comes out ahead.

In Geekbench Vulkan, the Radeon 610M delivers a far more emphatic result. It scores 6353, while the R5 M335 manages only 4758. The 33.5% delta represents a substantial performance gap. This is the dominant result of the comparison, and it aligns with the 610M's newer RDNA 2.0 architecture, its support for Vulkan 1.4, and its ray tracing cores. Modern Vulkan titles and compute workloads that leverage the newer API features are likely to show a significant advantage for the 610M.

The average scores reflect this split. The 610M's average of 5444 is 692 points higher than the R5 M335's 4752. In percentile terms, the 610M sits at the 32nd percentile of all GPUs, the R5 M335 at the 28th. The nearest rivals for the 610M cluster tightly: NVIDIA Quadro M4000 at 5467 (delta -0.4%), AMD Radeon R7 M440 at 5483 (delta -0.7%), and NVIDIA GeForce GTX 765M at 5501 (delta -1%). The R5 M335's nearest rivals include AMD Radeon R8 M445DX at 4727 (delta 0.5%), AMD Radeon R5 M255 at 4788 (delta -0.7%), and NVIDIA Quadro P400 at 4684 (delta 1.5%).

Where Each One Wins

The Radeon 610M wins in scenarios that favor modern API usage and architectural efficiency. Its Vulkan score of 6353 is the standout result, and users running Vulkan-based games or compute applications should expect a 33.5% advantage over the R5 M335. The 610M also supports DirectX 12 Ultimate (12_2), which brings features like ray tracing and mesh shaders to supported titles. Its 2 ray tracing cores, while modest, are entirely absent from the R5 M335. The 6 nm process and RDNA 2.0 architecture suggest better power efficiency, reflected in its 15 W TDP. The 610M's PCIe 4.0 x8 interface provides double the bandwidth of the R5 M335's PCIe 3.0 x8, which can benefit system-shared memory access patterns.

The Radeon R5 M335 wins in raw compute throughput and legacy OpenCL scenarios. Its OpenCL score of 4745 beats the 610M by 4.4%, and its hardware configuration supports this: 320 shading units, 20 TMUs, 8 ROPs, and 659.2 GFLOPS of FP32 performance. The dedicated 2 GB DDR3 memory with 14.40 GB/s bandwidth offers predictable, isolated memory access that system-shared designs cannot match. Applications that rely on OpenCL and do not support Vulkan will not see the 610M's architectural advantages. The R5 M335 also has a higher pixel rate (8.240 GPixel/s) and texture rate (20.60 GTexel/s), which can help in fill-rate-bound legacy workloads.

The decision framework is straightforward. Choose the Radeon 610M for Vulkan-based gaming, modern DirectX 12 Ultimate workloads, and any application that can use its ray tracing or newer API features. Choose the Radeon R5 M335 for OpenCL compute tasks that favor its higher shading unit count and dedicated memory, or for legacy software stacks that predate Vulkan's widespread adoption. The data does not support the R5 M335 as a general-purpose winner, but its one benchmark victory is legitimate and should not be ignored.