Intel Iris Pro Graphics P580 vs NVIDIA GeForce GTX 750 Comparison

The NVIDIA GeForce GTX 750 and Intel Iris Pro Graphics P580 are both end-of-life graphics solutions, but they represent fundamentally different design philosophies: a discrete add-in card versus an integrated processor graphics unit. The data in the FACT PACK shows a near-tie in average benchmark scores, with the GTX 750 at 7222 and the Iris Pro P580 at 7170, a margin of only 0.7%. However, this surface-level parity masks significant architectural and performance differences that become apparent in specific workloads.

FAQ

Q: How does the average benchmark score of the GTX 750 compare to the Iris Pro P580?

A: The GTX 750 has an average benchmark score of 7222, while the Iris Pro P580 scores 7170. This puts the GTX 750 ahead by approximately 0.7%, based on the nearestRivals deltaPct values.

Q: Which GPU wins in the OpenCL benchmark, and by how much?

A: The GTX 750 wins the Geekbench OpenCL test with a score of 9315, compared to the Iris Pro P580's 9082. The delta is 2.6% in favor of NVIDIA.

Q: What is the biggest performance gap between these two GPUs in any test?

A: The largest gap is in the Geekbench Vulkan benchmark, where the GTX 750 scores 8078 versus the Iris Pro P580's 5258. This represents a 53.6% advantage for the GTX 750.

Q: What are the respective TDPs of these two GPUs?

A: The GTX 750 has a TDP of 55 W, while the Iris Pro P580 has a TDP of 15 W. The Intel part draws significantly less power.

Q: What is the memory configuration difference?

A: The GTX 750 uses 1024 MB of dedicated GDDR5 memory with a 128-bit bus and 80.19 GB/s bandwidth. The Iris Pro P580 uses System Shared memory, with bandwidth described as "System Dependent."

Q: How do their shader configurations differ?

A: The GTX 750 has 512 shading units, 32 TMUs, and 16 ROPs. The Iris Pro P580 has 576 shading units, 72 TMUs, and 9 ROPs.

Architecture Differences

The GTX 750 is built on NVIDIA's Maxwell architecture using the GM107 chip, fabricated on a 28 nm process at TSMC. It packs 1,870 million transistors into a 148 mm² die, yielding a transistor density of 12.6M per mm². The Iris Pro P580, by contrast, uses Intel's Generation 9.0 architecture with the Skylake GT4e chip, manufactured on a 14 nm+ process at Intel. No transistor count or die size data is available for the Intel part.

These architectural choices lead to different resource allocations. The GTX 750 has 512 shading units, 32 TMUs, and 16 ROPs. The Iris Pro P580 has more shading units (576) and significantly more TMUs (72), but fewer ROPs (9). This suggests the Intel part is designed for texture-heavy compute, while the NVIDIA part favors pixel throughput.

Clock speeds also differ substantially. The GTX 750 runs at a base clock of 1020 MHz with a boost of 1085 MHz, while the Iris Pro P580 has a base of 350 MHz and a boost of 1000 MHz. The NVIDIA card's memory runs at 1253 MHz (5 Gbps effective) on GDDR5, whereas the Intel GPU uses system shared memory with no dedicated clock.

The GTX 750 supports PCIe 3.0 x16 and is a single-slot card with a 145 mm length (5.7 inches), requiring no power connectors and a suggested 250 W PSU. The Iris Pro P580 is an IGP with a Ring Bus interface and motherboard-dependent display outputs. It also supports a newer DirectX version: 12 (12_1) versus the GTX 750's 12 (11_0).

The Verdict

The data points to a nuanced conclusion. In raw compute terms, the GTX 750 holds a slight edge in OpenCL (9315 vs 9082, a 2.6% delta) and a commanding lead in Vulkan (8078 vs 5258, a 53.6% delta). The GTX 750 wins both head-to-head benchmarks and has a higher avgBenchmarkScore.

However, the Iris Pro P580 is not without merit. Its 15 W TDP is dramatically lower than the GTX 750's 55 W, making it far more suitable for power-constrained environments. It also delivers higher texture rate (72.00 GTexel/s vs 34.72 GTexel/s) and comparable FP32 performance (1,152.0 GFLOPS vs 1,111.0 GFLOPS). The Intel part even exceeds the NVIDIA card in shading units and TMUs.

For users who prioritize graphics API performance, especially Vulkan, the GTX 750 is the clear choice. For those who need integrated simplicity and minimal power draw, the Iris Pro P580 offers a compelling alternative, despite its lower raw benchmark scores.

Specification Differences

The two GPUs differ across nearly every major specification category. The GTX 750 uses a 28 nm process from TSMC, while the Iris Pro P580 uses 14 nm+ from Intel. The NVIDIA chip has 1,870 million transistors on a 148 mm² die; the Intel chip's transistor count and die size are not listed.

Clock speeds: GTX 750 base is 1020 MHz with boost at 1085 MHz; Iris Pro P580 base is 350 MHz with boost at 1000 MHz. Memory: GTX 750 has 1024 MB GDDR5 on a 128-bit bus with 80.19 GB/s bandwidth; Iris Pro P580 uses System Shared memory with System Dependent bandwidth.

Shading units: 512 (GTX 750) vs 576 (Iris Pro P580). TMUs: 32 vs 72. ROPs: 16 vs 9. Pixel rate: 17.36 GPixel/s vs 9.000 GPixel/s. Texture rate: 34.72 GTexel/s vs 72.00 GTexel/s. FP32: 1,111.0 GFLOPS vs 1,152.0 GFLOPS. The Intel part also has FP16 capability at 2.304 TFLOPS (2:1), which the GTX 750 lacks.

TDP: 55 W vs 15 W. Slot width: Single-slot vs IGP. Power connectors: None (GTX 750) vs not applicable. Suggested PSU: 250 W for NVIDIA, none listed for Intel. Bus interface: PCIe 3.0 x16 vs Ring Bus. Display outputs: 2x DVI and 1x mini-HDMI 1.4a vs Motherboard Dependent.

DirectX support: 12 (11_0) for GTX 750 vs 12 (12_1) for Iris Pro P580. Vulkan: 1.4 vs 1.3. OpenGL is 4.6 for both. The GTX 750 has a launch MSRP of 119 USD, while the Iris Pro P580 has no listed launch MSRP.

Head-to-Head Benchmarks

The head-to-head benchmark data shows a clear pattern. In Geekbench OpenCL, the GTX 750 scores 9315, edging out the Iris Pro P580's 9082. This 2.6% delta is modest, indicating that both GPUs are roughly comparable in general compute workloads. The GTX 750's dedicated GDDR5 memory with 80.19 GB/s bandwidth likely contributes to this advantage over the Intel part's system shared memory.

The Vulkan benchmark tells a very different story. The GTX 750 achieves 8078, while the Iris Pro P580 only manages 5258. This 53.6% delta is massive and suggests that the NVIDIA architecture is far more efficient at handling Vulkan's low-level API requirements. The GTX 750's support for Vulkan 1.4 versus the Intel part's Vulkan 1.3 may also play a role, though the magnitude of the gap points to deeper architectural advantages.

Across both tests, the GTX 750 wins 2 out of 2 benchmarks. The Iris Pro P580 does not win any. This is reflected in the avgBenchmarkScore: 7222 for NVIDIA versus 7170 for Intel. Interestingly, the nearestRivals data shows the GTX 750 is 0.3% ahead of AMD Radeon Vega 8 Mobile (7203) and 0.7% ahead of both the GTX 560 SE (7171) and the Iris Pro P580 (7170). The Iris Pro P580 is 0% different from the GTX 560 SE, 0.2% ahead of the GTX 970 (7157), and 0.5% behind the Vega 8 Mobile.

Where Each One Wins

The GTX 750 wins decisively in graphics API performance, particularly Vulkan, where it leads by 53.6%. It also wins in pixel rate (17.36 GPixel/s vs 9.000 GPixel/s), suggesting better performance in fill-rate-bound scenarios like traditional rasterization. The NVIDIA card's dedicated memory with 80.19 GB/s bandwidth provides a clear advantage over system shared memory, which shows in its higher benchmark scores. For users running Vulkan-based games or applications, the GTX 750 is the superior choice.

The Iris Pro P580 wins in power efficiency, with a 15 W TDP versus 55 W for the GTX 750. It also has a higher texture rate (72.00 GTexel/s vs 34.72 GTexel/s), more TMUs (72 vs 32), and slightly higher FP32 compute (1,152.0 GFLOPS vs 1,111.0 GFLOPS). This makes it better suited for texture-heavy compute workloads and environments where power draw is a primary concern. Its support for DirectX 12 (12_1) is also more advanced than the GTX 750's 12 (11_0).

The integration aspect matters too. The Iris Pro P580 is an IGP with a Ring Bus interface, requiring no separate card, power connectors, or PSU upgrade. The GTX 750 is a discrete card needing a PCIe 3.0 x16 slot and a 250 W suggested PSU. For compact or low-power systems, the Intel part is the only viable option. For dedicated graphics performance, the GTX 750's benchmark wins make it the data-backed recommendation.