GPU Comparison

Intel HD Graphics 630 and AMD Radeon R5 Graphics are both end-of-life integrated graphics solutions, and the benchmark data shows a near-total stalemate. The Intel part wins one API test decisively, the AMD part wins the other by a similar margin, and their average scores land within 5% of each other. The Intel HD Graphics 630 edges ahead on overall average score (4075 vs. 3883, a 4.9% gap), but the AMD Radeon R5 Graphics delivers a higher peak score in a single benchmark (5183 vs. 5099). Neither GPU escapes the bottom quarter of the performance distribution, with the Intel part sitting at the 24th percentile and the AMD part at the 23rd percentile of all GPUs.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel HD Graphics 630, with an average score of 4075 compared to the AMD Radeon R5 Graphics' 3883. That is a 4.9% difference in Intel's favor.

Q: How do the two compare in the Geekbench OpenCL test?

A: The AMD Radeon R5 Graphics wins decisively, scoring 5183 against Intel's 3587. That is a 30.8% advantage for AMD in this specific workload.

Q: Which GPU wins the Geekbench Vulkan test?

A: The Intel HD Graphics 630 takes the Vulkan test, scoring 3540 versus AMD's 2582. Intel leads by 37.1% in this API.

Q: What are the respective process nodes for these chips?

A: The Intel HD Graphics 630 uses a 14 nm++ process from Intel, while the AMD Radeon R5 Graphics uses a 28 nm process from GlobalFoundries. The AMD chip is built on a larger node.

Q: Do both GPUs support DirectX 12?

A: Yes, both support DirectX 12, but at different feature levels. The Intel HD Graphics 630 supports DirectX 12 (12_1), while the AMD Radeon R5 Graphics supports DirectX 12 (12_0).

Q: What is the transistor count difference between the two?

A: The AMD Radeon R5 Graphics has 2,410 million transistors on a 245 mm² die, while the Intel HD Graphics 630 does not have transistor or die size data listed in the specifications.

Where Each One Wins

The AMD Radeon R5 Graphics is the clear winner in OpenCL compute workloads. Its Geekbench OpenCL score of 5183 is not just higher than Intel's 3587; it is also higher than any single benchmark score the Intel part achieves across all three tests. This indicates that the GCN 2.0 architecture in the AMD chip is more efficient at raw parallel computation when accessed through OpenCL, making it the stronger choice for applications that leverage that specific API.

The Intel HD Graphics 630 wins the Vulkan API test with a score of 3540, which is 37.1% ahead of the AMD Radeon R5 Graphics' 2582. This is a significant margin and suggests that Intel's Generation 9.5 architecture is better optimized for Vulkan's lower-level hardware access. For any workload or game that uses Vulkan, the Intel part has a substantial performance advantage.

In terms of peak theoretical throughput, the two are nearly identical. The Intel HD Graphics 630 delivers 384.0 GFLOPS FP32, while the AMD Radeon R5 Graphics delivers 388.1 GFLOPS FP32. The AMD part also has a slightly higher pixel rate (3.032 GPixel/s vs. 3.000 GPixel/s), but the Intel part has a significantly higher texture rate (24.00 GTexel/s vs. 12.13 GTexel/s). The texture rate advantage for Intel is notable, as it is nearly double that of AMD.

Architecture Differences

The architectural divide is fundamental. The Intel HD Graphics 630 is built on the Generation 9.5 architecture (chip: Kaby Lake GT2) using a 14 nm++ process node. The AMD Radeon R5 Graphics uses the GCN 2.0 architecture (chip: Spectre SL) on a 28 nm process, fabricated by GlobalFoundries. This means Intel is using a significantly more advanced manufacturing process, allowing for higher clock speeds and efficiency at the same 15 W TDP.

The shader configurations differ in composition but not in raw output. The Intel HD Graphics 630 has 192 shading units, 24 TMUs, and 3 ROPs. The AMD Radeon R5 Graphics has 256 shading units, 16 TMUs, and 4 ROPs. Despite having fewer shaders, Intel achieves nearly the same FP32 throughput (384.0 GFLOPS vs. 388.1 GFLOPS) due to its higher clock speeds (boost: 1000 MHz vs. base clocks not listed for AMD). The Intel part also has double the texture rate (24.00 GTexel/s vs. 12.13 GTexel/s), which is a direct result of having more TMUs.

Memory handling is identical in design: both use "System Shared" memory with "System Dependent" bandwidth. Neither has dedicated VRAM. The API support also diverges slightly: Intel supports DirectX 12 (12_1) and Vulkan 1.3, while AMD supports DirectX 12 (12_0) and Vulkan 1.2.170. The Intel part has a slightly higher DirectX feature level and a newer Vulkan version.

Specification Differences

| Specification | Intel HD Graphics 630 | AMD Radeon R5 Graphics |

|---|---|---|

| Architecture | Generation 9.5 | GCN 2.0 |

| Process Node | 14 nm++ | 28 nm |

| Foundry | Intel | GlobalFoundries |

| Shading Units | 192 | 256 |

| TMUs | 24 | 16 |

| ROPs | 3 | 4 |

| Pixel Rate | 3.000 GPixel/s | 3.032 GPixel/s |

| Texture Rate | 24.00 GTexel/s | 12.13 GTexel/s |

| FP32 | 384.0 GFLOPS | 388.1 GFLOPS |

| FP16 | 768.0 GFLOPS (2:1) | Not listed |

| DirectX Support | 12 (12_1) | 12 (12_0) |

| Vulkan Support | 1.3 | 1.2.170 |

| Transistors | Not listed | 2,410 million |

| Die Size | Not listed | 245 mm² |

| Transistor Density | Not listed | 9.8M / mm² |

| Release Date | 2016-08-29 | 2014-09-16 |

| Predecessor | Not listed | TeraScale 3 IGP |

| Successor | Not listed | GCN 3.0 IGP |

Head-to-Head Benchmarks

The two available head-to-head benchmarks split exactly one win each, making the overall comparison a tie on wins. The AMD Radeon R5 Graphics takes the Geekbench OpenCL test with a score of 5183, beating the Intel HD Graphics 630's 3587 by a delta of -30.8% (from Intel's perspective). This is a dominant victory for AMD in OpenCL, showcasing the GCN 2.0 architecture's strength in this specific compute API.

The Intel HD Graphics 630 counters in the Geekbench Vulkan test. Intel scores 3540 against AMD's 2582, a delta of 37.1% in Intel's favor. This is a similarly dominant victory, indicating that Intel's architecture is far better suited to Vulkan's explicit multi-threading and lower overhead model. The margin of victory in each test is substantial, but they cancel each other out in a pure win/loss tally.

Looking at the broader benchmark suite, the Intel HD Graphics 630 also has a Geekbench Metal score of 5099, which is not available for the AMD part. This additional data point gives Intel a higher average score (4075 vs. 3883) and pushes it to the 24th percentile, one point above AMD's 23rd percentile. The Intel part's nearest rival is the AMD Radeon RX 9060 XT 8 GB (avg score 4093, delta -0.4%), while AMD's nearest rival is the NVIDIA Quadro 2000 (avg score 3898, delta -0.4%). Both are clustered with other low-end mobile and workstation parts.

The Verdict

The data does not support a universal winner. The AMD Radeon R5 Graphics is the better choice for OpenCL-based compute tasks, delivering a 30.8% higher score than Intel. If your primary workload is OpenCL, the AMD part is the clear pick, as its 5183 score outpaces anything the Intel part can produce.

The Intel HD Graphics 630 is the better choice for Vulkan-based workloads, with a 37.1% higher score than AMD. For gaming or applications that leverage Vulkan, the Intel part is the only rational selection. It also has a higher average benchmark score (4075 vs. 3883) and a higher percentile ranking (24th vs. 23rd), plus the availability of a Metal benchmark score (5099) which AMD lacks entirely.

For everything else, the choice is less clear. The FP32 throughput is nearly identical (384.0 GFLOPS vs. 388.1 GFLOPS), and the pixel rates are within 1% of each other. The Intel part has a massive texture rate advantage (24.00 GTexel/s vs. 12.13 GTexel/s), which could benefit texture-heavy workloads, but the AMD part has a newer predecessor/successor lineage (TeraScale 3 IGP to GCN 3.0 IGP) suggesting a more modern design at its release. Ultimately, the decision comes down to API preference: pick AMD for OpenCL, pick Intel for Vulkan and general average performance.