AMD Radeon HD 8790M vs Intel Iris Pro Graphics P6300 Comparison

Head-to-Head Benchmarks

The only direct benchmark comparison available is Geekbench OpenCL, and it shows a clear win for the Intel Iris Pro Graphics P6300. The Intel part scores 5712 points, while the AMD Radeon HD 8790M manages 5017 points. That is a 13.9% advantage for the Intel solution, which is a substantial margin in a single benchmark. When you consider that both GPUs are in the 33rd percentile of all GPUs, this result separates them meaningfully within their own performance tier.

Looking at the wider rival landscape, the Intel Iris Pro Graphics P6300 sits at 5712 points, landing just 0.1% behind the NVIDIA GeForce GTX 670MX (5721) and 0.3% behind the GTX 550 Ti (5731). The AMD Radeon HD 8790M, meanwhile, scores 5691 in its average benchmark score, which puts it 0.4% behind the Intel part. The Intel GPU also holds a 1.9% edge over the NVIDIA Quadro M500M (5604), while the AMD part is only 1.6% ahead of that same Quadro. What this means is that while the Intel GPU wins the head-to-head, both parts are clustered within a narrow band of performance, and the 13.9% OpenCL delta is the single largest gap between them in any measured metric.

The AMD GPU does have a Vulkan benchmark score of 6365, but there is no comparable Vulkan score for the Intel part in the data, so it cannot be used as a direct point of comparison. In terms of average benchmark score, the Intel part edges ahead at 5712 versus 5691 for the AMD, a 0.4% difference that is almost negligible in real-world terms. The head-to-head table confirms the Intel win with a 13.9% delta, but the average scores tell a slightly tighter story. That discrepancy suggests that the AMD part performs closer to the Intel GPU in some workloads not captured in the single OpenCL test.

Architecture Differences

The two GPUs come from completely different design philosophies and manufacturing processes. The Intel Iris Pro Graphics P6300 is built on the Broadwell GT3e chip, using Intel’s Generation 8.0 architecture with a 14 nm process node fabricated by Intel itself. The AMD Radeon HD 8790M uses the Mars chip, built on GCN 1.0 architecture with a 28 nm process node from TSMC. That process node difference is significant: 14 nm versus 28 nm means Intel has a density advantage that helps explain how it achieves competitive performance with a much lower power envelope.

The Intel chip operates at a base clock of 300 MHz with a boost up to 800 MHz, while the AMD chip runs at a base of 850 MHz and boosts to 900 MHz. Despite the lower clocks, the Intel part still wins the OpenCL benchmark, which points to architectural efficiency differences. Both GPUs have 384 shading units, but the Intel part has 48 texture mapping units (TMUs) versus only 24 on the AMD side. The ROP count flips in the other direction: Intel has 6 ROPs, while AMD has 8.

Memory configurations are starkly different. The Intel Iris Pro Graphics P6300 uses system shared memory with a system-dependent bandwidth, meaning it relies on the host system’s RAM and has no dedicated VRAM. The AMD Radeon HD 8790M comes with 2 GB of dedicated GDDR5 memory on a 128-bit bus, delivering 64.00 GB/s of bandwidth. That dedicated memory gives the AMD part a theoretical advantage in memory-bound scenarios, though the benchmark data does not show it overcoming the Intel part’s overall compute lead.

The AMD chip is significantly larger in terms of die size, coming in at 77 mm² with 950 million transistors, giving a transistor density of 12.3M per mm². The Intel part does not have published transistor or die size figures in the data. The AMD GPU also has a higher pixel rate at 7.200 GPixel/s versus 4.800 GPixel/s for Intel, and the texture rate favors Intel at 38.40 GTexel/s versus 21.60 GTexel/s. Floating-point performance (FP32) slightly favors AMD at 691.2 GFLOPS versus 614.4 GFLOPS for Intel.

API support differs as well. Both support DirectX 12 (11_1), but the AMD part offers OpenGL 4.6 versus OpenGL 4.4 for Intel, and Vulkan 1.2.170 versus Vulkan 1.0 for the Intel GPU. The AMD part has a higher level of API maturity, which could matter for compatibility with newer software.

Where Each One Wins

The Intel Iris Pro Graphics P6300 wins the only direct benchmark comparison, taking the Geekbench OpenCL test with 5712 points against the AMD’s 5017. That 13.9% margin indicates the Intel architecture is more efficient at compute tasks, likely due to its newer process node and higher TMU count. The Intel part also has a slightly better average benchmark score (5712 versus 5691), reinforcing its position as the stronger overall performer in the data available.

The AMD Radeon HD 8790M does not win any benchmark in the head-to-head comparison, but it does have specific hardware advantages that could translate to wins in certain workloads. Its 64.00 GB/s of dedicated GDDR5 memory bandwidth is a major asset for texture-heavy applications and higher resolutions, where system-shared memory on the Intel part could become a bottleneck. The AMD part also has a higher pixel rate (7.200 GPixel/s versus 4.800 GPixel/s), which benefits fill-rate-bound scenarios like simple geometry rendering or post-processing effects.

The AMD GPU’s higher FP32 throughput (691.2 GFLOPS versus 614.4 GFLOPS) suggests it should handle raw compute shaders better, yet the OpenCL benchmark does not reflect that. The 8 ROPs on the AMD part versus 6 on Intel could also help in certain rendering paths. However, the AMD part’s lower TMU count (24 versus 48) likely hurts it in texture-heavy workloads, and the benchmark data confirms the Intel part is the stronger compute performer overall.

The Intel GPU uses system shared memory, which means its performance is dependent on the host system’s memory configuration. In a laptop with fast dual-channel RAM, the Intel part could perform closer to its theoretical maximum. The AMD part, with its dedicated 2 GB GDDR5, is less sensitive to system memory speed and has a fixed bandwidth that does not vary with the host platform.

FAQ

Q: Which GPU has the higher Geekbench OpenCL score?

A: The Intel Iris Pro Graphics P6300 scores 5712, while the AMD Radeon HD 8790M scores 5017. That gives Intel a 13.9% advantage in the direct head-to-head comparison.

Q: Does the AMD Radeon HD 8790M have any benchmark where it wins?

A: In the head-to-head data, the AMD part does not win any benchmark. It does have a Vulkan score of 6365, but there is no comparable Vulkan score for the Intel part to make a direct comparison.

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

A: The Intel part is 0.1% behind the NVIDIA GeForce GTX 670MX and 0.3% behind the GTX 550 Ti. The AMD part is 0.5% behind the GTX 670MX and 0.7% behind the GTX 550 Ti. Both are ahead of the NVIDIA Quadro M500M, with Intel leading by 1.9% and AMD by 1.6%.

Q: What are the memory differences between the two GPUs?

A: The Intel Iris Pro Graphics P6300 uses system shared memory with system-dependent bandwidth. The AMD Radeon HD 8790M has 2 GB of dedicated GDDR5 memory on a 128-bit bus with 64.00 GB/s of bandwidth.

Q: Which GPU has more texture mapping units?

A: The Intel Iris Pro Graphics P6300 has 48 TMUs, while the AMD Radeon HD 8790M has only 24. This gives Intel a 38.40 GTexel/s texture rate versus 21.60 GTexel/s for AMD.

Q: What are the API differences between the two?

A: Both support DirectX 12 (11_1). The AMD part supports OpenGL 4.6 and Vulkan 1.2.170, while the Intel part supports OpenGL 4.4 and Vulkan 1.0.

Specification Differences

The process node differs significantly: Intel uses 14 nm, while AMD uses 28 nm. The Intel chip is fabricated by Intel, and the AMD chip is fabricated by TSMC. The AMD part has published transistor figures of 950 million on a 77 mm² die with a density of 12.3M per mm²; the Intel part has no such figures in the data.

Clock speeds differ, with the Intel part running at 300 MHz base and 800 MHz boost, versus 850 MHz base and 900 MHz boost for AMD. Memory is the largest divergence: Intel uses system shared memory with system-dependent bandwidth, while AMD has 2 GB of GDDR5 on a 128-bit bus with 64.00 GB/s bandwidth.

The shading unit count is identical at 384, but TMUs differ at 48 for Intel versus 24 for AMD, and ROPs differ at 6 versus 8. Pixel rate favors AMD at 7.200 GPixel/s versus 4.800 GPixel/s, while texture rate favors Intel at 38.40 GTexel/s versus 21.60 GTexel/s. FP32 compute favors AMD at 691.2 GFLOPS versus 614.4 GFLOPS.

The Intel part has a TDP of 15 W, while the AMD part has no published TDP in the data. The Intel GPU is an IGP with a Ring Bus interface, while the AMD GPU is an MXM Module with an MXM-A (3.0) bus interface. The AMD part has no power connectors listed, and both have display outputs that are dependent on the motherboard or portable device.

API support differs: Intel offers OpenGL 4.4 and Vulkan 1.0, while AMD offers OpenGL 4.6 and Vulkan 1.2.170. Both support DirectX 12 (11_1). The AMD part has a predecessor (London) and successor (Gem System), while the Intel part has neither listed. Release dates differ, with the AMD part released earlier, and the Intel part released later. Both are end-of-life products.

The Verdict

The benchmark data points to the Intel Iris Pro Graphics P6300 as the stronger GPU for compute workloads. It wins the only direct comparison with a 13.9% margin in Geekbench OpenCL, and it also holds a slight edge in average benchmark score (5712 versus 5691). For anyone relying on OpenCL compute tasks, the Intel part is the clear choice.

The AMD Radeon HD 8790M is not without merit, though. Its dedicated 2 GB GDDR5 memory with 64.00 GB/s bandwidth is a significant advantage for gaming scenarios where memory bandwidth is critical. The system-shared memory on the Intel part is a potential bottleneck that could erode its compute lead in real-world gaming. The AMD part also has a higher pixel rate and more ROPs, which could benefit certain rendering paths.

The AMD part supports newer API versions, with Vulkan 1.2.170 and OpenGL 4.6 versus Vulkan 1.0 and OpenGL 4.4 on the Intel side. For users running modern Vulkan titles, the AMD part might have better compatibility and feature support. However, the lack of a Vulkan benchmark for the Intel part means this cannot be quantified.

The Intel part uses 14 nm process technology against 28 nm on AMD, which likely explains its lower TDP of 15 W. That makes the Intel GPU more suitable for power-constrained systems. The AMD part, being an MXM module, is designed for upgradeable laptops and has its own dedicated memory, which could be preferable in systems where system RAM is limited.

For compute performance, the Intel Iris Pro Graphics P6300 wins decisively in the data. For gaming and memory-bandwidth-sensitive applications, the AMD Radeon HD 8790M has the hardware to compete, though the benchmark data does not confirm this. Both GPUs sit in the 33rd percentile of all GPUs, meaning neither is a high-end performer. The choice ultimately comes down to whether dedicated memory and newer API support matter more than raw OpenCL compute performance. The data says Intel for compute, AMD for memory bandwidth and API maturity.