Intel Iris Xe MAX Graphics vs NVIDIA GeForce GTX 590 Comparison

Head-to-Head Benchmarks

The recorded data contains a single direct comparison between the Intel Iris Xe MAX Graphics and the NVIDIA GeForce GTX 590, using the Geekbench OpenCL test. In that benchmark, the Intel part scored 14315 points against 12830 for the NVIDIA card. That works out to an 11.6% advantage for the Intel solution. The win count reflects this: one benchmark win for Intel, zero for NVIDIA.

The magnitude of that lead is worth putting in context. The Intel GPU sits at the 56th percentile of all GPUs in the database, while the GTX 590 sits at the 52nd percentile. The gap between those percentile positions aligns with the raw score difference. In practical terms, an 11.6% margin in a compute-oriented benchmark like OpenCL suggests that the Intel part has a real performance edge in workloads that scale well across its execution resources.

Looking at the nearest rivals for each product sharpens the picture. The Intel Iris Xe MAX Graphics lands within a tight cluster: it is 0.3% behind the AMD Radeon Vega 11, 0.3% ahead of the NVIDIA GeForce GTX 1070 Ti, 0.4% behind the NVIDIA GeForce GTX TITAN, and 0.5% behind the AMD Radeon RX Vega 11. That is a remarkably narrow band, meaning the Intel part trades blows with some heavy hitters from both AMD and NVIDIA, even though those are older or lower-tier products in their respective lineups.

The GTX 590, by contrast, sits in a slightly lower neighborhood. It matches the AMD Radeon Pro 455 exactly, is 0.1% ahead of the AMD FirePro W5100, 0.3% ahead of the AMD Radeon 740M, and 0.4% behind the NVIDIA GeForce GTX 670. None of those rivals are particularly close to the performance class of the GTX 1070 Ti or GTX TITAN, which reinforces that the Intel part is operating in a higher performance bracket despite its compact form factor.

The delta between the two products, 11.6%, is not enormous, but it is consistent. The Intel GPU wins the only recorded head-to-head contest. There is no benchmark in the database where the GTX 590 comes out ahead. That is a clean sweep, albeit over a single test. The data does not support a narrative where the GTX 590 has any compute advantage in OpenCL.

Architecture Differences

The two GPUs come from completely different technological eras, and the architectural splits are stark. Intel uses a DG1 chip built on Generation 12.1 architecture, which is part of the Xe Graphics generation. The process node is 10 nm, fabricated by Intel itself. The die size is 95 mm². NVIDIA uses a GF110 chip built on Fermi 2.0 architecture, part of the GeForce 500 generation. That chip is fabricated by TSMC on a 40 nm process, and the die is much larger at 520 mm². NVIDIA also lists a transistor count of 3,000 million for the GF110, with a transistor density of 5.8 million per mm². Intel does not provide a transistor count or density figure in the database.

The memory subsystems are equally divergent. Intel pairs its GPU with 4 GB of LPDDR4X on a 128-bit bus, delivering 68.26 GB/s of bandwidth. NVIDIA uses 1536 MB of GDDR5 on a much wider 384-bit bus, delivering 164.0 GB/s. That is a massive bandwidth advantage for NVIDIA, more than double. However, the Intel part uses a faster effective memory clock: 4.3 Gbps versus 3.4 Gbps. The wider bus wins in raw throughput, but the Intel memory clock is higher per pin.

Core resource counts tell a mixed story. Intel fields 768 shading units, 48 texture mapping units, and 24 ROPs. NVIDIA fields 512 shading units, 64 TMUs, and 48 ROPs. So Intel has more shaders, but NVIDIA has more TMUs and ROPs. The pixel rate favors Intel at 39.60 GPixel/s versus 19.46 GPixel/s, and the texture rate also favors Intel at 79.20 GTexel/s versus 38.91 GTexel/s. The FP32 compute rating goes to Intel as well: 2.534 TFLOPS versus 1,244.2 GFLOPS (which is 1.244 TFLOPS). NVIDIA does not list an FP16 figure, while Intel lists 5.069 TFLOPS with a 2:1 ratio.

Thermal and power characteristics are on entirely different scales. The Intel part draws 25 W TDP and is listed as an IGP (integrated graphics processor) with no power connectors. NVIDIA draws 365 W TDP, is a dual-slot card, and requires two 8-pin power connectors. The suggested PSU ratings reflect that: 200 W for Intel, 750 W for NVIDIA. The bus interface also differs: Intel uses PCIe 4.0 x8, while NVIDIA uses PCIe 2.0 x16.

API support is a point of separation. Intel lists DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. NVIDIA lists DirectX 12 (11_0), OpenGL 4.6, and no Vulkan entry at all. The DirectX feature level is higher on the Intel side, which matters for modern game compatibility.

Where Each One Wins

The data points to a clear split. The Intel Iris Xe MAX Graphics wins the only benchmark recorded, and it wins on compute throughput metrics: FP32, pixel rate, texture rate, and shading unit count. It also wins on efficiency, with a 25 W TDP versus 365 W, and on modern API support, with a higher DirectX feature level and Vulkan support.

The GTX 590 has advantages in memory bandwidth, TMU and ROP counts, and physical size. Its 164.0 GB/s bandwidth is a substantial margin over the Intel part, which could matter in bandwidth-sensitive workloads like high-resolution texture streaming or certain compute tasks. The dual-slot form factor and 279 mm length are typical for a high-end card of its era, but they make it a poor fit for small cases. The Intel part, being an IGP with no display outputs, is not a drop-in replacement for a discrete card, but it carries no physical footprint at all.

In terms of use cases, the Intel part is better suited for compute-oriented tasks that leverage its higher shader count and FP32 throughput. The GTX 590, despite losing the OpenCL test, still holds a bandwidth advantage that could help in specific scenarios. However, the database does not list any benchmark where the GTX 590 wins, so any such advantage remains theoretical.

The Verdict

The recorded data is unambiguous: the Intel Iris Xe MAX Graphics outperforms the NVIDIA GeForce GTX 590 in the only benchmark where they meet. The 11.6% lead in Geekbench OpenCL is backed by higher pixel rate, texture rate, FP32 compute, and a more modern architecture on a smaller process node. The GTX 590 is an older, power-hungry design with a wider memory bus, but that does not translate into a win in the database.

If the choice is between these two for a compute workload, the Intel part is the better performer. It also uses dramatically less power, requires no external power connectors, and supports newer APIs. The GTX 590 has a launch MSRP of 699 USD, which the database records, but that figure is historical and not a current purchasing guide.

For anyone building a system where the GPU is integrated and power is limited, the Intel Iris Xe MAX Graphics is the clear pick. For anyone who needs a dual-slot card with multiple DVI outputs, the GTX 590 is the only option among the two, since the Intel part has no display outputs. But on raw benchmark performance, the data favors Intel.

FAQ

Q: Which GPU wins the Geekbench OpenCL benchmark?

A: The Intel Iris Xe MAX Graphics scores 14315, while the NVIDIA GeForce GTX 590 scores 12830. Intel wins by 11.6%.

Q: How does the Intel Iris Xe MAX compare to the GTX 1070 Ti?

A: The Intel part scores 0.3% higher than the NVIDIA GeForce GTX 1070 Ti in the database, with scores of 14315 and 14277 respectively.

Q: What is the memory bandwidth difference between the two?

A: The GTX 590 has 164.0 GB/s on a 384-bit bus, while the Intel Iris Xe MAX has 68.26 GB/s on a 128-bit bus. The GTX 590 has more than double the bandwidth.

Q: Does the GTX 590 support Vulkan?

A: No. The database lists no Vulkan support for the GTX 590. The Intel Iris Xe MAX lists Vulkan 1.4.

Q: What are the power requirements for each GPU?

A: The Intel Iris Xe MAX has a 25 W TDP and a suggested PSU of 200 W. The GTX 590 has a 365 W TDP and a suggested PSU of 750 W, with two 8-pin power connectors.

Q: What is the DirectX feature level for each card?

A: The Intel Iris Xe MAX supports DirectX 12 (12_1). The GTX 590 supports DirectX 12 (11_0), which is a lower feature level.

Specification Differences

| Specification | Intel Iris Xe MAX Graphics | NVIDIA GeForce GTX 590 |

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

| Chip | DG1 | GF110 |

| Architecture | Generation 12.1 | Fermi 2.0 |

| Process Node | 10 nm | 40 nm |

| Foundry | Intel | TSMC |

| Die Size | 95 mm² | 520 mm² |

| Transistors | Not listed | 3,000 million |

| Transistor Density | Not listed | 5.8M / mm² |

| Memory Size | 4 GB | 1536 MB |

| Memory Type | LPDDR4X | GDDR5 |

| Memory Bus Width | 128 bit | 384 bit |

| Memory Bandwidth | 68.26 GB/s | 164.0 GB/s |

| Memory Clock | 4.3 Gbps effective | 3.4 Gbps effective |

| Shading Units | 768 | 512 |

| TMUs | 48 | 64 |

| ROPs | 24 | 48 |

| Pixel Rate | 39.60 GPixel/s | 19.46 GPixel/s |

| Texture Rate | 79.20 GTexel/s | 38.91 GTexel/s |

| FP32 | 2.534 TFLOPS | 1,244.2 GFLOPS |

| FP16 | 5.069 TFLOPS (2:1) | Not listed |

| TDP | 25 W | 365 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 2x 8-pin |

| Suggested PSU | 200 W | 750 W |

| Bus Interface | PCIe 4.0 x8 | PCIe 2.0 x16 |

| Display Outputs | No outputs | 3x DVI, 1x mini-DisplayPort |

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

| Vulkan | 1.4 | Not listed |

| Dimensions | Not listed | 279 mm x 111 mm x 40 mm |

| Release Date | 2020-10-30 | 2011-03-23 |

| Launch MSRP | Not listed | 699 USD |