Intel Iris Pro Graphics P580 vs NVIDIA GeForce MX230 Comparison

Head-to-Head Benchmarks

The recorded data splits cleanly between these two mobile graphics solutions, with each taking one of the two benchmark disciplines. In Geekbench OpenCL, the Intel Iris Pro Graphics P580 posts a score of 9082, while the NVIDIA GeForce MX230 manages 5739. That difference translates to a 58.3% advantage for the Intel part, a massive gap in raw compute throughput that suggests the older integrated solution has substantial headroom in workloads that scale with shading unit count and memory bandwidth.

The numbers align with the underlying hardware. The P580 carries 576 shading units, 72 texture mapping units, and 9 raster output pipelines. Its FP32 throughput is recorded at 1,152.0 GFLOPS, and its texture rate reaches 72.00 GTexel/s. The MX230, by contrast, has 256 shading units, 16 TMUs, and 16 ROPs, with FP32 at 783.9 GFLOPS and texture rate at 24.50 GTexel/s. The Intel chip's shading unit count is more than double that of the NVIDIA part, and its texture rate is nearly triple. OpenCL benchmarks tend to reward exactly these resources, so the 58.3% delta is consistent with the specification disparity.

The Vulkan result flips the narrative. Here, the MX230 scores 6414, while the P580 trails at 5258, a delta of 18% in favor of NVIDIA. This is notable because the P580's FP32 advantage should, in theory, carry over to graphics APIs. Yet the recorded data shows the opposite. The MX230's newer Pascal architecture, with its 16 ROPs compared to the P580's 9, likely explains part of the gap. Rasterization-heavy workloads depend on pixel throughput, and the MX230's pixel rate of 24.50 GPixel/s dwarfs the P580's 9.000 GPixel/s. The NVIDIA part also benefits from dedicated GDDR5 memory with 48.06 GB/s bandwidth, whereas the Intel solution relies on system-shared memory with bandwidth described as "System Dependent."

The average benchmark scores reflect these split results. The P580 averages 7170 across both tests, placing it in the 39th percentile of all GPUs. The MX230 averages 6077, sitting in the 35th percentile. Interestingly, the P580's nearest rivals in the database include the NVIDIA GeForce GTX 560 SE at 7171 (delta 0%), the GeForce GTX 970 at 7157 (delta 0.2%), and the AMD Radeon Vega 8 Mobile at 7203 (delta -0.5%). The MX230's nearest rivals include the NVIDIA RTX A400 at 6078 (delta 0%), the Quadro P2000 at 6049 (delta 0.5%), and the Intel Iris Pro Graphics 6200 at 6117 (delta -0.7%). These proximity figures suggest the P580 competes with discrete desktop-class parts from several generations ago, while the MX230 sits closer to entry-level mobile and workstation silicon.

The wins are balanced at one each, but the magnitude of the Intel victory in OpenCL (58.3%) is substantially larger than the NVIDIA margin in Vulkan (18%). This asymmetry matters. It means the P580's advantage in compute-heavy tasks is more pronounced than the MX230's advantage in graphics-API workloads. For users who prioritize raw compute, the Intel part is the clear choice. For those who prioritize Vulkan-based gaming or rendering, the NVIDIA part holds the edge, but by a smaller margin.

The Verdict

The data points to a nuanced conclusion. Neither GPU dominates outright, and the choice depends entirely on the workload profile. The Intel Iris Pro Graphics P580 wins decisively in OpenCL compute, with a 58.3% lead over the MX230. Its average benchmark score of 7170 also exceeds the MX230's 6077 by roughly 18%, placing it higher in the overall percentile ranking (39th vs 35th). If the primary use case involves OpenCL-accelerated applications, such as certain video encoding tools, scientific computing libraries, or productivity suites, the P580 is the superior option based on recorded measurements.

The NVIDIA GeForce MX230, however, wins in Vulkan by 18%. This suggests better driver optimization or architectural efficiency for modern graphics APIs. The MX230's dedicated GDDR5 memory with 48.06 GB/s bandwidth provides a consistent memory subsystem, whereas the P580's system-shared memory introduces variability depending on the host platform's memory configuration. For Vulkan-based games or applications that leverage this API, the MX230 is the safer choice.

The overall average benchmark score favors the P580, and its percentile placement is higher. But the MX230's Vulkan win cannot be ignored, especially since Vulkan is increasingly common in both gaming and professional visualization. The data does not support a universal recommendation. Instead, it suggests that buyers should match the GPU to their specific software stack. OpenCL-heavy workflows point to Intel; Vulkan-heavy workflows point to NVIDIA.

Where Each One Wins

The Intel Iris Pro Graphics P580 wins in OpenCL compute benchmarks. Its 9082 score in Geekbench OpenCL represents a 58.3% advantage over the MX230's 5739. This win likely stems from the P580's larger shading unit count (576 vs 256) and higher FP32 throughput (1,152.0 GFLOPS vs 783.9 GFLOPS). Applications that exercise general-purpose GPU compute, such as image processing, data analysis, or physics simulations, should favor the Intel part based on this metric.

The NVIDIA GeForce MX230 wins in Vulkan graphics benchmarks. Its 6414 score in Geekbench Vulkan beats the P580's 5258 by 18%. This win likely derives from the MX230's higher pixel rate (24.50 GPixel/s vs 9.000 GPixel/s), its dedicated GDDR5 memory with 48.06 GB/s bandwidth, and its newer Pascal architecture. Vulkan-based games, 3D renderers, or any software that relies on this cross-platform graphics API should perform better on the NVIDIA part.

The split also extends to architectural characteristics. The P580's texture rate of 72.00 GTexel/s versus the MX230's 24.50 GTexel/s suggests the Intel part handles texture-heavy compute workloads better. Conversely, the MX230's 16 ROPs versus the P580's 9 ROPs indicates superior pixel-processing capability, which matters for final framebuffer output in graphics rendering.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The Intel Iris Pro Graphics P580, with an average score of 7170 across Geekbench OpenCL and Vulkan tests, compared to the NVIDIA GeForce MX230's average of 6077.

Q: How large is the Intel's lead in OpenCL?

A: The P580 scores 9082 in Geekbench OpenCL versus the MX230's 5739, a delta of 58.3% in favor of Intel.

Q: By how much does the NVIDIA win in Vulkan?

A: The MX230 scores 6414 in Geekbench Vulkan versus the P580's 5258, an 18% advantage for NVIDIA.

Q: What explains the Intel's OpenCL dominance?

A: The P580 has 576 shading units, 72 TMUs, and an FP32 throughput of 1,152.0 GFLOPS, all substantially higher than the MX230's 256 shading units, 16 TMUs, and 783.9 GFLOPS.

Q: What explains the NVIDIA's Vulkan advantage?

A: The MX230 has a pixel rate of 24.50 GPixel/s and 16 ROPs, compared to the P580's 9.000 GPixel/s and 9 ROPs, along with dedicated GDDR5 memory providing 48.06 GB/s bandwidth.

Q: How do these GPUs rank among all GPUs in the database?

A: The P580 sits in the 39th percentile, while the MX230 sits in the 35th percentile, based on their average benchmark scores.

Architecture Differences

The two GPUs come from different architectural eras and manufacturers. The Intel Iris Pro Graphics P580 uses the Skylake GT4e chip, built on Intel's Generation 9.0 architecture, specifically the HD Graphics-W (Skylake) generation. It is fabricated on a 14 nm+ process at Intel's own foundry. The NVIDIA GeForce MX230 uses the GP108 chip, based on the Pascal architecture, from the GeForce MX (2xx) generation. It is fabricated on a 14 nm process at Samsung's foundry.

The P580's memory subsystem is entirely system-shared, with both memory size and type listed as "System Shared" and bandwidth described as "System Dependent." This means performance varies with the host system's RAM configuration. The MX230, in contrast, has 2 GB of dedicated GDDR5 memory on a 64-bit bus, providing 48.06 GB/s of fixed bandwidth. This is a fundamental architectural difference: the Intel part relies on shared system memory, while the NVIDIA part has its own fast memory pool.

The P580's bus interface is a Ring Bus, typical of integrated graphics within Intel processors. The MX230 uses PCIe 3.0 x4, indicating it is a discrete chip, even though its slot width is listed as IGP. Display outputs differ as well: the P580's outputs are "Motherboard Dependent," while the MX230's are "Portable Device Dependent," reflecting their intended placements in desktop versus laptop systems.

The MX230 also has a defined transistor count of 1,800 million on a 74 mm² die, with a transistor density of 24.3M per mm². The P580 lacks such figures in the database. The MX230's memory clock is listed at 1502 MHz with 6 Gbps effective, while the P580's memory clock is simply "System Shared." These differences highlight the fundamental design split: a large integrated GPU with shared memory versus a smaller discrete GPU with dedicated VRAM.

Specification Differences

The two GPUs differ across nearly every measurable specification. The P580 has 576 shading units, 72 TMUs, and 9 ROPs, while the MX230 has 256 shading units, 16 TMUs, and 16 ROPs. The Intel part offers more compute-oriented resources, while the NVIDIA part offers more pixel-oriented resources.

Clock speeds differ substantially. The P580 runs at a base clock of 350 MHz and boosts to 1000 MHz. The MX230 runs at a base clock of 1519 MHz and boosts to 1531 MHz. The NVIDIA chip operates at much higher frequencies, though its lower core count means its overall throughput still trails in FP32 (783.9 GFLOPS vs 1,152.0 GFLOPS).

Pixel and texture rates follow the ROP and TMU counts. The P580 delivers 9.000 GPixel/s and 72.00 GTexel/s. The MX230 delivers 24.50 GPixel/s and 24.50 GTexel/s. The NVIDIA part excels at pixel fill, while the Intel part excels at texture fill.

FP16 performance also diverges sharply. The P580 achieves 2.304 TFLOPS with a 2:1 ratio, indicating half-rate FP16 execution. The MX230 achieves only 12.25 GFLOPS with a 1:64 ratio, meaning its FP16 throughput is severely limited. For workloads that leverage FP16 arithmetic, the Intel part is vastly superior.

The MX230 has a lower TDP at 10 W versus the P580's 15 W. Both are listed as IGP slot width, but the MX230 has no power connectors while the P580's connector status is not recorded. The MX230 supports Vulkan 1.4, while the P580 supports Vulkan 1.3. Both support DirectX 12 (12_1) and OpenGL 4.6. The MX230's release date is 2019-02-20, while the P580's is 2015-08-31, making the Intel part older by several years, yet still competitive in compute benchmarks.