GPU Comparison

# Where Each One Wins

The single benchmark in the data, Geekbench OpenCL, is won by the NVIDIA GeForce GTX 580 with a score of 15,283 against the Intel Iris Xe MAX Graphics' 14,315. That is a 6.8% margin, which is meaningful but not decisive. The GTX 580 takes the only recorded win, while the Intel part records zero wins in this comparison.

The GTX 580 sits at the 58th percentile among all GPUs, while the Iris Xe MAX sits at the 56th. The overlap in their percentile ranges is telling: these are not cards from different performance tiers but rather adjacent performers. The GTX 580's nearest rivals, the NVIDIA GeForce RTX 2060 (15,290, 0% delta), AMD Radeon 680M (15,270, 0.1% delta), RTX 3050 OEM (15,199, 0.6% delta), and Radeon RX 7600 (15,171, 0.7% delta), all cluster within a 0.7% band. That places the GTX 580 in a crowded mid-pack of modern and older hardware where a few points separate competitors.

The Iris Xe MAX's nearest rivals tell a similar story from slightly below: AMD Radeon Vega 11 (14,352, -0.3% delta), NVIDIA GeForce GTX 1070 Ti (14,277, 0.3% delta), NVIDIA GeForce GTX TITAN (14,373, -0.4% delta), and AMD Radeon RX Vega 11 (14,385, -0.5% delta). The Intel part is sandwiched between a 2017 enthusiast card and integrated graphics, which shows its positioning as a low-power discrete option that lands in a strange middle ground.

In terms of use cases, the data does not support a clear workload split. With only one benchmark, the GTX 580 wins compute performance outright, but the margin is small enough that real-world differences would be minor for most tasks. The Iris Xe MAX does not win any recorded benchmark, so there is no basis to claim any compute advantage. However, its architectural features, such as dedicated ray tracing units absent in the GTX 580, may matter in specific workloads not covered by the Geekbench OpenCL test.

# Architecture Differences

The two chips come from fundamentally different eras and design philosophies. The GTX 580 uses the GF110 chip built on TSMC's 40 nm process, with 3,000 million transistors on a 520 mm² die. That yields a transistor density of 5.8 million per square millimeter. The Iris Xe MAX uses Intel's DG1 chip on Intel's 10 nm process, with a 95 mm² die; transistor count and density are not listed in the data.

The GTX 580 is a Fermi 2.0 architecture part from the GeForce 500 generation, while the Iris Xe MAX is Generation 12.1 Xe Graphics. The GTX 580 has 512 shading units, 64 texture mapping units, and 48 ROPs. The Iris Xe MAX has 768 shading units, 48 TMUs, and 24 ROPs. Despite fewer TMUs and ROPs, the Intel part posts higher fill rates: 39.60 GPixel/s pixel rate and 79.20 GTexel/s texture rate versus the GTX 580's 24.70 GPixel/s and 49.41 GTexel/s. The Intel part also has higher raw FP32 throughput at 2.534 TFLOPS versus 1.581 TFLOPS, a 60% advantage on paper.

Memory configurations diverge sharply. The GTX 580 uses 1536 MB of GDDR5 on a 384-bit bus, delivering 192.4 GB/s bandwidth. The Iris Xe MAX uses 4 GB of LPDDR4X on a 128-bit bus, delivering 68.26 GB/s, less than half the bandwidth. Clock speeds also differ: the GTX 580's memory runs at 1002 MHz (4 Gbps effective), while the Intel part's base clock is 300 MHz with a 1650 MHz boost, and memory runs at 2133 MHz (4.3 Gbps effective).

The power envelope is the starkest difference. The GTX 580 has a 244 W TDP, requires a dual-slot cooler, and needs 1x 6-pin plus 1x 8-pin power connectors with a 550 W suggested PSU. The Iris Xe MAX is an IGP with a 25 W TDP and a 200 W suggested PSU, with no power connectors or display outputs listed. The GTX 580 is a 267 mm long, 111 mm tall add-in card; the Intel part has no listed dimensions.

API support favors the Intel part. The Iris Xe MAX supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The GTX 580 supports DirectX 12 (11_0), OpenGL 4.6, and no Vulkan. The GTX 580 uses PCIe 2.0 x16; the Iris Xe MAX uses PCIe 4.0 x8.

# Head-to-Head Benchmarks

The only head-to-head benchmark is Geekbench OpenCL, where the GTX 580 scores 15,283 versus the Iris Xe MAX's 14,315. The delta is 6.8% in favor of NVIDIA. That is the entire measured difference between these two products.

The GTX 580's score places it within 0.1% of the RTX 2060 (15,290) and the Radeon 680M (15,270), effectively a tie. It sits 0.6% above the RTX 3050 OEM and 0.7% above the RX 7600. The Iris Xe MAX, at 14,315, is 0.3% behind the Radeon Vega 11, 0.3% ahead of the GTX 1070 Ti, 0.4% behind the GTX TITAN, and 0.5% behind the RX Vega 11.

A 6.8% delta in a single OpenCL test is small but consistent with the percentile positioning. The GTX 580's 58th percentile and the Iris Xe MAX's 56th percentile are adjacent, meaning the performance gap between them is smaller than the gap between either card and the next tier of GPUs. If the benchmark were run multiple times, variance alone might shift results; the delta is within a range where driver versions or thermal conditions could alter the outcome.

The GTX 580's higher memory bandwidth (192.4 GB/s vs 68.26 GB/s) likely contributes to its lead in this OpenCL workload, as compute kernels often stress memory throughput. However, the Iris Xe MAX's higher FP32 throughput (2.534 TFLOPS vs 1.581 TFLOPS) suggests it should win compute-bound tasks. The fact that it does not indicates the benchmark is not purely arithmetic, memory bandwidth and driver efficiency play decisive roles.

# FAQ

Q: Which GPU has the higher benchmark score?

A: The NVIDIA GeForce GTX 580 scores 15,283 in Geekbench OpenCL, which is 6.8% higher than the Intel Iris Xe MAX Graphics' 14,315.

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

A: The GTX 580 is within 0.7% of the RTX 2060, Radeon 680M, RTX 3050 OEM, and RX 7600. The Iris Xe MAX is within 0.5% of the Radeon Vega 11, GTX 1070 Ti, GTX TITAN, and RX Vega 11.

Q: Which GPU has more shader units?

A: The Intel Iris Xe MAX has 768 shading units, while the NVIDIA GeForce GTX 580 has 512 shading units.

Q: What is the memory bandwidth difference?

A: The GTX 580 has 192.4 GB/s bandwidth from 1536 MB of GDDR5 on a 384-bit bus. The Iris Xe MAX has 68.26 GB/s from 4 GB of LPDDR4X on a 128-bit bus.

Q: Which GPU has higher power consumption?

A: The GTX 580 has a 244 W TDP and requires a 550 W suggested PSU. The Iris Xe MAX has a 25 W TDP with a 200 W suggested PSU.

Q: Does either GPU support ray tracing?

A: Neither GPU lists ray tracing cores in the data. The Iris Xe MAX supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 580 supports DirectX 12 (11_0) with no Vulkan.

# The Verdict

The data supports a narrow victory for the NVIDIA GeForce GTX 580 in raw OpenCL compute performance. A 6.8% lead is real but not transformative; both cards cluster with modern competitors within a few percentage points. The GTX 580's 58th percentile and the Iris Xe MAX's 56th percentile confirm they belong to the same performance tier.

For users who need maximum compute throughput in OpenCL workloads, the GTX 580 is the choice, but its 244 W TDP, dual-slot size, and 550 W PSU requirement make it a demanding part. The Iris Xe MAX delivers 60% higher FP32 theoretical throughput (2.534 TFLOPS vs 1.581 TFLOPS) and 60% higher texture fill rate (79.20 GTexel/s vs 49.41 GTexel/s), yet loses the measured benchmark. That suggests the GTX 580's 192.4 GB/s memory bandwidth is the deciding factor in this test.

The Iris Xe MAX is the only one of the two that is an IGP with no display outputs, meaning it must be paired with a system that has its own graphics output. The GTX 580 has 2x DVI and 1x mini-HDMI 1.3a outputs. For a system builder in 2025, the GTX 580 requires a legacy PCIe 2.0 slot and power connectors; the Iris Xe MAX requires PCIe 4.0 and virtually no power.

There is no winner in absolute terms. The GTX 580 wins the only recorded benchmark, but the Iris Xe MAX wins on efficiency, API support, and theoretical compute. If the question is which GPU scores higher on Geekbench OpenCL, the answer is the GTX 580 by 6.8%. If the question is which GPU is more capable per watt, the data shows a 25 W part scoring within 6.8% of a 244 W part, a remarkable efficiency difference. The GTX 580 is end-of-life, as is the Iris Xe MAX, so neither is a future-proof purchase. Choose the GTX 580 for the benchmark lead and memory bandwidth; choose the Iris Xe MAX for low power consumption and modern API support, accepting the measured compute deficit.