AMD Radeon HD 6950 vs Intel Iris Pro Graphics P580 Comparison

Where Each One Wins

The benchmark data presents a clear, if lopsided, picture. The Intel Iris Pro Graphics P580 wins the only head-to-head benchmark recorded in the database, the Geekbench OpenCL test, by a substantial margin. The AMD Radeon HD 6950 does not win any of the recorded comparisons.

This does not mean the AMD card has no role. Its strengths lie in its dedicated memory subsystem and raw fillrate capabilities, which are not fully captured by the single compute-oriented benchmark available. The Intel solution, being an integrated graphics processor, trades its superior compute score for flexibility and low power draw. The AMD card, by contrast, is a discrete, dual-slot board with its own VRAM, which historically would have been favored for memory-bandwidth-sensitive workloads such as high-resolution texture streaming.

The data suggests that for purely compute-centric tasks, as measured by OpenCL, the Intel part is the clear winner. For tasks that depend on dedicated video memory bandwidth and pixel throughput, the AMD card's specifications point to advantages that the current benchmark suite does not directly measure. The recorded numbers, however, only award the win to Intel.

Architecture Differences

The architectural gap between these two products is generationally wide. The Intel Iris Pro Graphics P580 is built on a 14 nm+ process by Intel, using the Skylake GT4e chip and the Generation 9.0 architecture. The AMD Radeon HD 6950 uses the much older 40 nm process from TSMC, with the Cayman chip and the TeraScale 3 architecture. This process difference alone accounts for a vast discrepancy in power efficiency and transistor density.

The transistor counts and die sizes reinforce this divide. The AMD chip packs 2,640 million transistors onto a 389 mm² die, yielding a density of 6.8M per mm². The Intel part, being an integrated GPU, does not have a listed transistor count or die size in the database, but its 15 W TDP versus the AMD's 200 W TDP tells a story of thermal design. The AMD card requires a dual-slot cooler, two 6-pin power connectors, and a suggested 550 W power supply. The Intel IGP, by contrast, is a ring-bus integrated solution with no separate power connectors.

Memory architecture is another fundamental divergence. The Intel GPU uses system-shared memory, with bandwidth described as "System Dependent." The AMD card has 2 GB of dedicated GDDR5 memory on a 256-bit bus, delivering 160.0 GB/s of bandwidth. This is a critical difference for gaming and texture-heavy workloads.

The compute units also differ. The Intel part has 576 shading units, 72 texture mapping units, and only 9 ROPs. The AMD card has 1,408 shading units, 88 TMUs, and 32 ROPs. The AMD card's pixel rate of 25.60 GPixel/s is nearly three times the Intel's 9.000 GPixel/s. However, the Intel part's FP32 throughput of 1,152.0 GFLOPS is roughly half of the AMD's 2.253 TFLOPS, yet the Intel part still wins the OpenCL benchmark, suggesting that the benchmark rewards more than raw shader count.

Feature support is also split. The Intel GPU supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The AMD card is limited to DirectX 11.2 (11_0) and OpenGL 4.4, with no Vulkan support listed. This makes the Intel part far more future-proof for modern APIs.

Head-to-Head Benchmarks

The database records exactly one head-to-head benchmark between these two GPUs: the Geekbench OpenCL test. In this test, the Intel Iris Pro Graphics P580 scores 9,082 points, while the AMD Radeon HD 6950 scores 6,210 points. The Intel part wins by a delta of 46.2%. This is a decisive margin, not a close call.

The Intel part's average benchmark score across all recorded tests is 7,170, while the AMD's average is 6,210. This means the Intel part's single OpenCL result is well above its average, while the AMD's only result is its average. The Intel part's Vulkan score is 5,258, which is not directly comparable to the AMD card since the AMD card has no Vulkan result in the database.

Contextualizing these scores against their nearest rivals clarifies the picture. The Intel part's average score of 7,170 puts it within 0.2% of the NVIDIA GeForce GTX 970 (7,157) and within 0.7% of the NVIDIA GeForce GTX 750 (7,222). It is also 0.5% behind the AMD Radeon Vega 8 Mobile (7,203). The AMD Radeon HD 6950's average of 6,210 places it 0.2% behind the AMD FirePro W600 (6,223) and 1.1% behind the NVIDIA Quadro K620 (6,282). Notably, the AMD card's nearest rivals include modern NVIDIA cards like the RTX 4070 Ti SUPER AD102 and RTX 5070 Ti SUPER, both scoring 6,270, which are 0.9% ahead of the HD 6950. This suggests the HD 6950's compute performance is roughly on par with some modern low-end to mid-range offerings, an impressive feat for a 2010 architecture.

The 46.2% delta in the head-to-head test is the single most important number in this comparison. It indicates that for OpenCL compute workloads, the Intel IGP is significantly faster than the older discrete card, despite the AMD card having more than twice the shading units and higher pixel rate.

The Verdict

The data points to a clear conclusion: the Intel Iris Pro Graphics P580 is the superior choice for compute-oriented tasks as measured by the database's benchmark suite. It wins the only head-to-head comparison by 46.2%, and its average score is about 15.5% higher than the AMD card's average. Its modern API support, including DirectX 12 and Vulkan, makes it a more viable option for contemporary software.

However, the AMD Radeon HD 6950 should not be dismissed outright. Its dedicated 2 GB GDDR5 memory with 160.0 GB/s bandwidth is a major advantage for tasks that require large texture datasets or high-resolution framebuffers. Its 32 ROPs and 25.60 GPixel/s pixel rate suggest it would outperform the Intel part in pure rasterization throughput. The AMD card also has a higher FP32 throughput at 2.253 TFLOPS, though this did not translate to a higher OpenCL score.

The Intel part's 15 W TDP versus the AMD's 200 W TDP is a decisive factor for system builders. The Intel IGP requires no additional cooling, no power connectors, and no bulky dual-slot cooler. The AMD card demands a 550 W power supply and a 286 mm long, 126 mm tall, 42 mm wide physical footprint.

For users building a compact, low-power system with modern API support, the Intel Iris Pro Graphics P580 is the clear winner. For users with legacy software that relies on dedicated memory bandwidth and who have the power budget and physical space, the AMD Radeon HD 6950 retains some theoretical advantages. The benchmark data, however, only supports the Intel part's superiority in compute performance. The AMD card's launch MSRP was 299 USD, a figure that matters little now given both parts are end-of-life.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The Intel Iris Pro Graphics P580 scores 9,082 in the Geekbench OpenCL test, while the AMD Radeon HD 6950 scores 6,210. The Intel part leads by 46.2%.

Q: Does the AMD Radeon HD 6950 have more shading units?

A: Yes, the AMD card has 1,408 shading units, compared to 576 on the Intel Iris Pro Graphics P580. Despite this, the Intel part wins the compute benchmark.

Q: What memory configurations do these GPUs use?

A: The Intel part uses system-shared memory with bandwidth described as "System Dependent." The AMD card has 2 GB of dedicated GDDR5 memory on a 256-bit bus with 160.0 GB/s bandwidth.

Q: Which GPU supports more modern graphics APIs?

A: The Intel Iris Pro Graphics P580 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The AMD Radeon HD 6950 only supports DirectX 11.2 (11_0) and OpenGL 4.4, with no Vulkan support.

Q: How do these GPUs compare to their nearest rivals?

A: The Intel part's average score of 7,170 is within 0.7% of the NVIDIA GeForce GTX 750 and 0.5% behind the AMD Radeon Vega 8 Mobile. The AMD card's average of 6,210 is 1.1% behind the NVIDIA Quadro K620 and 0.9% behind the NVIDIA RTX 4070 Ti SUPER AD102.

Q: What are the power requirements for each GPU?

A: The Intel Iris Pro Graphics P580 has a TDP of 15 W and no power connectors. The AMD Radeon HD 6950 has a TDP of 200 W, requires 2x 6-pin power connectors, and a suggested 550 W power supply.

Specification Differences

| Specification | Intel Iris Pro Graphics P580 | AMD Radeon HD 6950 |

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

| Architecture | Generation 9.0 | TeraScale 3 |

| Process Node | 14 nm+ | 40 nm |

| Foundry | Intel | TSMC |

| Transistors | Not listed | 2,640 million |

| Die Size | Not listed | 389 mm² |

| Transistor Density | Not listed | 6.8M / mm² |

| Base Clock | 350 MHz | Not listed |

| Boost Clock | 1000 MHz | Not listed |

| Memory Clock | System Shared | 1250 MHz, 5 Gbps effective |

| Memory Size | System Shared | 2 GB |

| Memory Type | System Shared | GDDR5 |

| Memory Bus | System Shared | 256 bit |

| Memory Bandwidth | System Dependent | 160.0 GB/s |

| Shading Units | 576 | 1408 |

| TMUs | 72 | 88 |

| ROPs | 9 | 32 |

| Pixel Rate | 9.000 GPixel/s | 25.60 GPixel/s |

| Texture Rate | 72.00 GTexel/s | 70.40 GTexel/s |

| FP32 Performance | 1,152.0 GFLOPS | 2.253 TFLOPS |

| TDP | 15 W | 200 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 2x 6-pin |

| Suggested PSU | Not listed | 550 W |

| Bus Interface | Ring Bus | PCIe 2.0 x16 |

| DirectX | 12 (12_1) | 11.2 (11_0) |

| OpenGL | 4.6 | 4.4 |

| Vulkan | 1.3 | Not listed |

| Length | Not listed | 286 mm (11.3 inches) |

| Height | Not listed | 126 mm (5 inches) |

| Width | Not listed | 42 mm (1.7 inches) |

| Release Date | 2015-08-31 | 2010-12-13 |

| Launch MSRP | Not listed | 299 USD |