AMD Radeon R5 M335 vs AMD Radeon Vega 3 Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon R5 M335 leads with an average benchmark score of 4752, while the AMD Radeon Vega 3 trails at 4268. That places the R5 M335 in the 28th percentile of all GPUs, versus the 25th percentile for the Vega 3.

Q: How large is the performance gap in the head-to-head tests?

A: The R5 M335 wins both recorded head-to-head tests. In Geekbench OpenCL, it scores 4745 against the Vega 3's 3963, a 19.7% advantage. In Geekbench Vulkan, it scores 4758 versus 3961, a 20.1% lead.

Q: Does the Vega 3 have any benchmark where it outperforms the R5 M335?

A: No. The database records two head-to-head tests, and the Vega 3 loses both. However, the Vega 3 does have a Geekbench Metal score of 4880, which is not a direct head-to-head comparison and cannot be matched against the R5 M335, which has no Metal result.

Q: What are the nearest rivals for each GPU, and how do they compare?

A: The R5 M335's closest rival is the AMD Radeon R8 M445DX, which scores 4727, just 0.5% behind. The Vega 3's nearest rival is the NVIDIA GeForce GTX 460M at 4282, which is 0.3% ahead of the Vega 3.

Q: Which GPU has the higher transistor count and smaller process node?

A: The Vega 3 uses a 12 nm process from GlobalFoundries and packs 4,940 million transistors. The R5 M335 uses a 28 nm process from TSMC with only 690 million transistors. The Vega 3 also has a higher transistor density at 23.5M per mm² versus 12.3M per mm².

Q: What is the memory configuration difference?

A: The R5 M335 has dedicated 2 GB of DDR3 memory on a 64-bit bus, delivering 14.40 GB/s bandwidth. The Vega 3 uses system-shared memory, with its bandwidth listed as system dependent and no dedicated VRAM.

The Verdict

The data points to a clear winner: the AMD Radeon R5 M335. Its average benchmark score of 4752 beats the Vega 3's 4268 by roughly 11%. In the two direct comparisons, the R5 M335 wins by nearly 20% in both OpenCL and Vulkan workloads. For anyone choosing between these two based solely on recorded performance, the R5 M335 is the stronger option.

That said, the Vega 3 is not without merits. It is a 12 nm part with a 15 W TDP, designed as an integrated GPU. Its architecture is GCN 5.0, a much newer design than the R5 M335's GCN 1.0. The Vega 3 also supports DirectX 12 (12_1) and Vulkan 1.3, while the R5 M335 only reaches DirectX 12 (11_1) and Vulkan 1.2.170. For systems where power efficiency and modern API support matter more than raw compute, the Vega 3 has a place.

The verdict is straightforward: pick the R5 M335 for raw benchmark performance, especially in OpenCL and Vulkan tasks. Pick the Vega 3 for a modern integrated solution with lower power draw and newer API support, accepting a significant performance deficit.

Head-to-Head Benchmarks

The two recorded head-to-head tests show a consistent pattern. In Geekbench OpenCL, the R5 M335 scores 4745, while the Vega 3 manages 3963. That is a 19.7% delta in favor of the R5 M335. The Vulkan test tells the same story: 4758 for the R5 M335 versus 3961 for the Vega 3, a 20.1% gap.

These margins are substantial. A 20% lead in Vulkan means the R5 M335 handles compute-heavy graphics workloads with noticeably more headroom. In OpenCL, the 19.7% advantage reinforces that the dedicated memory and higher shading unit count of the R5 M335 translate directly into compute throughput.

The Vega 3's scores are consistent across both tests, within 2 points of each other (3963 and 3961). The R5 M335 is equally consistent, with a 13-point spread (4745 and 4758). Neither GPU shows a workload-specific strength in these two tests; the R5 M335 simply outclasses the Vega 3 across the board.

Notably, the Vega 3 does post a Geekbench Metal score of 4880, which is higher than both of its OpenCL and Vulkan results. However, this is not a head-to-head metric, and the R5 M335 has no Metal score recorded, so no direct comparison is possible.

Specification Differences

The R5 M335 and Vega 3 differ in nearly every core specification. The R5 M335 has 320 shading units, 20 texture mapping units, and 8 raster output units. The Vega 3 has 192 shading units, 12 TMUs, and 4 ROPs. That gives the R5 M335 a 66.7% advantage in shading units, a 66.7% advantage in TMUs, and a 100% advantage in ROPs.

Pixel rate and texture rate follow the same pattern. The R5 M335 achieves 8.240 GPixel/s and 20.60 GTexel/s, while the Vega 3 manages 4.400 GPixel/s and 13.20 GTexel/s. The R5 M335 is nearly 87% faster in pixel throughput and 56% faster in texture throughput.

FP32 performance also favors the R5 M335. It delivers 659.2 GFLOPS, while the Vega 3 delivers 422.4 GFLOPS, a 56% difference. The Vega 3 does have FP16 capability at 844.8 GFLOPS (2:1), which the R5 M335 lacks entirely, but this only matters for specific workloads.

Memory is a fundamental difference. The R5 M335 uses 2 GB of dedicated DDR3 on a 64-bit bus, with 14.40 GB/s bandwidth. The Vega 3 uses system-shared memory with system-dependent bandwidth. The R5 M335 also has a fixed memory clock of 900 MHz (1800 Mbps effective), while the Vega 3 has no dedicated memory clock.

The R5 M335 connects via PCIe 3.0 x8, while the Vega 3 is an IGP with no bus interface beyond the integrated connection. Display outputs are portable-device dependent for the R5 M335 and motherboard dependent for the Vega 3.

Architecture Differences

The two GPUs come from different architectural eras. The R5 M335 is built on GCN 1.0, using the "Exo" chip and belonging to the Gem System generation. The Vega 3 is built on GCN 5.0, using the "Picasso" chip and belonging to the Vega IGP generation. The R5 M335 was released in October 2015, while the Vega 3 arrived in November 2019.

The process node tells a story of manufacturing progress. The R5 M335 uses a 28 nm process from TSMC, while the Vega 3 uses a 12 nm process from GlobalFoundries. Transistor counts reflect the integration scale: 690 million for the R5 M335 versus 4,940 million for the Vega 3. Die size also differs dramatically, 56 mm² for the R5 M335 versus 210 mm² for the Vega 3, with transistor density at 12.3M per mm² versus 23.5M per mm².

API support is another generational gap. The Vega 3 supports DirectX 12 (12_1), while the R5 M335 only reaches DirectX 12 (11_1). Vulkan support is newer on the Vega 3 (1.3 versus 1.2.170). Both support OpenGL 4.6.

The Vega 3 is a 15 W part, listed as an IGP with no power connectors. The R5 M335 has no TDP recorded in the database, but it also has no power connectors, suggesting a low-power mobile design. The Vega 3's predecessor is listed as GCN 3.0 IGP, and its successor is Vega II IGP. The R5 M335's predecessor is Solar System and its successor is Polaris Mobile.

Where Each One Wins

The R5 M335 wins in raw compute performance. Its 659.2 GFLOPS FP32 output is 56% higher than the Vega 3's 422.4 GFLOPS, and its pixel rate of 8.240 GPixel/s is nearly double the Vega 3's 4.400 GPixel/s. For tasks that rely on traditional rasterization and general compute, the R5 M335 is the clear choice.

The R5 M335 also wins in memory bandwidth. With 14.40 GB/s of dedicated bandwidth, it avoids the system-dependent variability of the Vega 3's shared memory setup. This gives the R5 M335 a predictable performance floor in memory-intensive workloads.

The Vega 3 wins on architecture modernity. Its GCN 5.0 design, 12 nm process, and support for DirectX 12 (12_1) and Vulkan 1.3 make it a better fit for newer software stacks. The R5 M335 is limited to DirectX 12 (11_1) and Vulkan 1.2.170, which could cause compatibility issues with the latest titles.

The Vega 3 also wins on power efficiency. At 15 W TDP, it is designed for tight thermal envelopes. The R5 M335 has no recorded TDP, but its 28 nm process and dedicated memory suggest a higher power draw. For thin-and-light systems, the Vega 3 is the safer pick.

The Vega 3 has one unique advantage: FP16 compute at 844.8 GFLOPS. The R5 M335 has no FP16 capability. Applications that leverage half-precision math, such as certain machine learning inference tasks, will run better on the Vega 3.

In terms of benchmark wins, the R5 M335 takes both head-to-head tests. But the database also records the Vega 3's Metal score of 4880, which exceeds the R5 M335's best result of 4758. That suggests the Vega 3 performs well in Apple's Metal API, though no direct comparison exists.

For gaming and general compute, choose the R5 M335. For integrated systems needing modern API support, low power draw, and FP16 capability, choose the Vega 3. The performance gap is real and consistent, but the Vega 3's newer architecture makes it a viable alternative in specific scenarios.