AMD Radeon HD 8850M vs Intel Iris Pro Graphics P580 Comparison

The Verdict

The data positions these two as near-parity parts, but with a clear split in workload type. The Intel Iris Pro Graphics P580 wins the only shared benchmark, Geekbench OpenCL, by a decisive margin: 9,082 versus 7,447 for the AMD Radeon HD 8850M, a delta of -18% from the AMD's perspective. However, the Intel part's average benchmark score across all its tests (7,170) is lower than the AMD's single-score average (7,447), because the Iris Pro's Vulkan result (5,258) drags its average down. The percentile rankings tell a similar story: both sit in the lower-middle tier of all GPUs, with the AMD at the 40th percentile and the Intel at the 39th. The verdict from the data: pick the Intel Iris Pro P580 for OpenCL compute workloads and DirectX 12 feature-level support; pick the AMD Radeon HD 8850M for its consistent single-benchmark performance and higher average score. Neither part is a clear all-around winner; the Intel wins the head-to-head compute test, but the AMD holds a slight edge in overall standing relative to the full GPU landscape.

Architecture Differences

The two GPUs come from fundamentally different design philosophies and process technologies. The AMD Radeon HD 8850M is built on TSMC's 28 nm process, packing 1,500 million transistors into a 123 mm² die, yielding a transistor density of 12.2 million per square millimeter. Its architecture is GCN 1.0, with the chip codenamed Venus, part of the Solar System generation (HD 8800M family). In contrast, the Intel Iris Pro Graphics P580 uses Intel's 14 nm+ process and the Skylake GT4e chip, from the Generation 9.0 architecture, belonging to the HD Graphics-W (Skylake) generation. The Intel part's transistor count and die size are not listed, but its process node is more advanced, which partially explains its higher clock ceiling.

Core configuration differs sharply. The AMD Radeon HD 8850M has 640 shading units, 40 texture mapping units, and 16 ROPs. The Intel Iris Pro P580 has fewer shading units at 576, but more TMUs at 72, and only 9 ROPs. This imbalance explains the theoretical throughput rates: the AMD delivers 25.00 GTexel/s texture fill and 10.00 GPixel/s pixel fill, while the Intel reaches 72.00 GTexel/s texture fill (nearly 3x higher) but only 9.000 GPixel/s pixel fill. In raw FP32 compute, the Intel is ahead: 1,152.0 GFLOPS versus 800.0 GFLOPS. The Intel also lists FP16 capability at 2.304 TFLOPS (2:1 ratio), which the AMD does not report. Memory architecture is entirely different: the AMD uses 2 GB of dedicated GDDR5 on a 128-bit bus with 64.00 GB/s bandwidth, while the Intel uses System Shared memory with System Dependent bandwidth. Clock behavior also diverges: the AMD's base is 575 MHz with a 625 MHz boost, while the Intel runs at a 350 MHz base but boosts to 1000 MHz.

Where Each One Wins

Looking strictly at benchmark wins, the Intel Iris Pro Graphics P580 is the sole victor in the head-to-head OpenCL test. Its 9,082 score is 18% higher than the AMD's 7,447. That is a substantial margin for a single compute workload. The Intel also has a second benchmark result, a Vulkan score of 5,258, which the AMD does not have listed. This suggests the Intel part is better suited for modern API compute tasks, and its DirectX 12 (12_1) support is a full feature level above the AMD's DirectX 12 (11_1). For Vulkan, the Intel lists API version 1.3, while the AMD lists 1.2.170.

The AMD Radeon HD 8850M wins on consistency and average standing. Its single benchmark score of 7,447 gives it an average of 7,447, which is higher than the Intel's average of 7,170 across two tests. The AMD's nearest rival is the NVIDIA GeForce GTX 1650 with an average score of 7,472, a delta of only -0.3%, meaning the AMD is essentially at parity with that much newer discrete card. In contrast, the Intel's nearest rival is the NVIDIA GeForce GTX 560 SE at 7,171, a delta of 0%, and it sits just below the AMD Radeon Vega 8 Mobile (7,203, -0.5%). The AMD's percentile rank (40) is one point higher than the Intel's (39). For texture-heavy workloads that are fill-rate bound, the Intel's 72.00 GTexel/s should be advantageous, while the AMD's higher pixel rate (10.00 GPixel/s) may help in rasterization-bound scenes. The AMD's dedicated 64.00 GB/s bandwidth is a fixed resource, while the Intel's System Shared memory bandwidth is System Dependent, meaning it could be a bottleneck or an advantage depending on the platform.

FAQ

Q: Which GPU has the higher OpenCL benchmark score?

A: The Intel Iris Pro Graphics P580 scores 9,082 in Geekbench OpenCL, while the AMD Radeon HD 8850M scores 7,447. The Intel wins by 18%.

Q: What is the average benchmark score for each GPU?

A: The AMD Radeon HD 8850M has an average benchmark score of 7,447 (from a single OpenCL test). The Intel Iris Pro Graphics P580 has an average score of 7,170, averaged from its OpenCL score of 9,082 and its Vulkan score of 5,258.

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

A: The AMD Radeon HD 8850M is 0.1% ahead of the Intel UHD Graphics 750 (7,441) and 0.3% ahead of the AMD Radeon R7 350 (7,425), but 0.3% behind the NVIDIA GeForce GTX 1650 (7,472). The Intel Iris Pro Graphics P580 is exactly tied with the NVIDIA GeForce GTX 560 SE (7,171), 0.2% ahead of the GTX 970 (7,157), and 0.5% behind the AMD Radeon Vega 8 Mobile (7,203).

Q: Which GPU has better DirectX feature-level support?

A: The Intel Iris Pro Graphics P580 supports DirectX 12 (12_1), which is a higher feature level than the AMD Radeon HD 8850M's DirectX 12 (11_1). Both support OpenGL 4.6, but the Intel supports Vulkan 1.3 while the AMD supports Vulkan 1.2.170.

Q: What are the memory configurations?

A: The AMD Radeon HD 8850M has 2 GB of dedicated GDDR5 memory on a 128-bit bus, providing 64.00 GB/s of fixed bandwidth. The Intel Iris Pro Graphics P580 uses System Shared memory, with System Shared type and bus width, and System Dependent bandwidth.

Q: Which GPU has higher theoretical texture and pixel fill rates?

A: The Intel Iris Pro Graphics P580 has a texture rate of 72.00 GTexel/s, which is nearly three times the AMD's 25.00 GTexel/s. However, the AMD has a higher pixel rate at 10.00 GPixel/s versus the Intel's 9.000 GPixel/s.

Head-to-Head Benchmarks

The only direct comparison available is the Geekbench OpenCL test, and it is a decisive win for the Intel Iris Pro Graphics P580. The Intel scores 9,082 against the AMD's 7,447, producing a deltaPct of -18% from the AMD's perspective. That is a significant gap in a compute-heavy workload. To put it in context, the AMD's nearest rival in the OpenCL arena is the NVIDIA GeForce GTX 1650 at 7,472, meaning the AMD is only 0.3% behind a modern discrete card. The Intel's OpenCL score of 9,082, if compared to its own nearest rivals, would place it well above the GTX 560 SE (7,171) and GTX 970 (7,157), indicating the Intel's compute throughput is far ahead of its average standing suggests. The Intel also has a Vulkan score of 5,258, which is lower than its OpenCL score, but there is no AMD Vulkan result to compare. The wins tally confirms the split: winsA is 0 and winsB is 1. The AMD's sole benchmark score of 7,447 is actually higher than the Intel's average score of 7,170, but the Intel's peak OpenCL score is the single biggest number in either dataset. For users focused on peak compute performance in a single API, the Intel is the clear choice. For users who want a consistent, if lower, performance profile, the AMD offers that.

Specification Differences

The two GPUs differ across nearly every specification field. The AMD Radeon HD 8850M uses a 28 nm TSMC process with 1,500 million transistors on a 123 mm² die, achieving a density of 12.2M transistors per mm². The Intel Iris Pro Graphics P580 uses a 14 nm+ Intel process, but its transistor count and die size are not listed. The AMD's base clock is 575 MHz with a 625 MHz boost, while the Intel's base is 350 MHz with a 1000 MHz boost — a far wider dynamic range. Memory is the starkest contrast: the AMD has 2 GB of GDDR5 on a 128-bit bus with 64.00 GB/s bandwidth, whereas the Intel uses System Shared memory with System Dependent bandwidth. The AMD has 640 shading units, 40 TMUs, and 16 ROPs; the Intel has 576 shading units, 72 TMUs, and 9 ROPs. Pixel rates are close (10.00 GPixel/s vs 9.000 GPixel/s), but texture rates diverge sharply (25.00 GTexel/s vs 72.00 GTexel/s). FP32 compute favors the Intel at 1,152.0 GFLOPS versus 800.0 GFLOPS, and the Intel lists FP16 at 2.304 TFLOPS (2:1) while the AMD does not report FP16. The Intel's TDP is listed at 15 W and it is an IGP with a Ring Bus interface, while the AMD's TDP, slot width, and power connectors are not specified; the AMD uses PCIe 3.0 x16. The Intel's display outputs are Motherboard Dependent, while the AMD lists none. API support differs: the AMD offers DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170, while the Intel offers DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. Release dates are separated by roughly two and a half years, with the AMD launching on 2013-03-31 and the Intel on 2015-08-31. Both are end-of-life products. The AMD's predecessor is London and successor is Gem System, while the Intel lists no predecessor or successor.