Intel Iris Pro Graphics P580 vs NVIDIA T600 Comparison

Intel Iris Pro Graphics P580 and NVIDIA T600 occupy very different positions in the GPU landscape, despite both landing at the 39th percentile in the overall benchmark distribution. The data shows a clear performance hierarchy, with the T600 dominating in every head-to-head comparison available. The Iris Pro P580, an integrated solution from Intel’s Skylake era, trails the dedicated Turing-based workstation card by substantial margins, though its integrated nature and low power envelope define its role.

Head-to-Head Benchmarks

The two available benchmark comparisons leave no ambiguity about the performance gap. In the Geekbench OpenCL test, the NVIDIA T600 scores 27,875 points against the Intel Iris Pro Graphics P580’s 9,082 points. This represents a 67.4% deficit for the Intel part, meaning the T600 delivers roughly three times the raw compute throughput in this API. The margin is even starker in the Geekbench Vulkan test, where the T600 posts 25,580 points versus the P580’s 5,258 points — a 79.4% gap. In Vulkan, the NVIDIA card outperforms the Intel integrated graphics by nearly fivefold.

These deltas are not marginal differences; they reflect fundamentally different hardware classes. The T600’s OpenCL score is more than triple the P580’s, while its Vulkan score is almost five times higher. No benchmark in the dataset favors the Intel part — the win count stands at 0 for the P580 and 2 for the T600. The average benchmark score across all recorded tests reinforces this: the T600 averages 7,035 points, while the P580 averages 7,170 points. Interestingly, the averages are nearly identical, but this is misleading because the T600’s average is dragged down by its Passmark DX9, DX10, DX11, and DX12 scores (114, 32, 49, and 25 respectively), which are low compared to its Geekbench results. The P580’s average comes solely from its two Geekbench tests.

Looking at the nearest rivals provides additional context. The P580’s closest competitor in average score is the NVIDIA GeForce GTX 560 SE (7,171 points, 0% delta), meaning the Intel part sits right at that performance tier. The T600, by contrast, is bracketed by the GeForce GTX 680M (7,023 points, 0.2% delta) and the GeForce GTX 970 (7,157 points, -1.7% delta). The T600’s average is slightly below the GTX 970 but above the GTX 680M, placing it in mid-range discrete territory.

Architecture Differences

The two GPUs come from different manufacturers, nodes, and design philosophies. The Intel Iris Pro Graphics P580 is built on the Skylake GT4e chip using Intel’s Generation 9.0 architecture, fabricated on a 14 nm+ process at Intel’s own foundry. It belongs to the HD Graphics-W (Skylake) generation and uses a Ring Bus interface, reflecting its integrated nature. The NVIDIA T600, conversely, uses the TU117 chip based on the Turing architecture, fabricated on a 12 nm process at TSMC. It belongs to the Quadro Turing (Tx000) generation and connects via PCIe 3.0 x16.

The Intel chip integrates 576 shading units, 72 texture mapping units, and just 9 raster output pipelines. Its pixel rate is 9.000 GPixel/s and texture rate is 72.00 GTexel/s. Floating-point performance is rated at 1,152.0 GFLOPS for FP32 and 2.304 TFLOPS for FP16 (2:1 ratio). The NVIDIA part features 640 shading units, 40 TMUs, and 32 ROPs. Its pixel rate reaches 42.72 GPixel/s, texture rate is 53.40 GTexel/s, and FP32 throughput is 1.709 TFLOPS, with FP16 at 3.418 TFLOPS (2:1). The T600 has higher pixel throughput despite fewer TMUs, a result of its higher clock speeds and more ROPs.

Memory configurations differ completely. The P580 uses system-shared memory with no dedicated VRAM, relying on the host system’s RAM with a system-dependent bandwidth. The T600 features 4 GB of GDDR6 memory on a 128-bit bus, delivering 160.0 GB/s of bandwidth. This dedicated memory bandwidth is critical for workloads that repeatedly access textures or geometry, as it avoids contention with the CPU.

Power and physical specifications also diverge. The P580 has a 15 W TDP and is an IGP (integrated graphics processor), meaning it has no slot width, power connectors, or display outputs of its own — everything is motherboard dependent. The T600 is a single-slot card with a 40 W TDP, no power connectors required, and a suggested 200 W PSU. It offers four mini-DisplayPort 1.4a outputs for multi-monitor setups.

API support shows some parity: both support DirectX 12 (12_1) and OpenGL 4.6. However, the T600 supports Vulkan 1.4, while the P580 is limited to Vulkan 1.3. This newer Vulkan version partially explains the T600’s massive lead in the Geekbench Vulkan test, although the hardware differences are the primary factor.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The Intel Iris Pro Graphics P580 averages 7,170 points, while the NVIDIA T600 averages 7,035 points. The Intel part is nominally ahead by 135 points, but this is within the margin of error given its nearest rival (GTX 560 SE) is at 7,171 points with a 0% delta.

Q: How much faster is the NVIDIA T600 in Geekbench OpenCL?

A: The T600 scores 27,875 versus the P580’s 9,082, a 67.4% advantage. This means the T600 achieves roughly three times the OpenCL performance.

Q: Does the Intel Iris Pro have any benchmark win over the T600?

A: No. In the head-to-head dataset, the T600 wins both available tests (Geekbench OpenCL and Vulkan). The Intel part has zero wins.

Q: What memory does the Intel Iris Pro use, and how does it compare to the T600?

A: The P580 uses system-shared memory with a system-dependent bandwidth, meaning it borrows from the host’s RAM. The T600 has 4 GB of dedicated GDDR6 memory on a 128-bit bus with 160.0 GB/s bandwidth.

Q: What are the power requirements for each?

A: The Intel P580 has a 15 W TDP and is an integrated part. The NVIDIA T600 has a 40 W TDP, requires no power connectors, and suggests a 200 W power supply.

Q: Which GPU supports a newer Vulkan version?

A: The NVIDIA T600 supports Vulkan 1.4, while the Intel P580 supports Vulkan 1.3.

Specification Differences

The specifications where the two GPUs differ are substantial. The process node differs: Intel uses 14 nm+ versus TSMC’s 12 nm for NVIDIA. The transistor count and die size are only listed for the T600 (4,700 million transistors, 200 mm², 23.5M / mm² density), while the P580 has no data. Clock speeds show the T600 running higher: base 735 MHz versus 350 MHz, boost 1,335 MHz versus 1,000 MHz. The T600 also has a dedicated memory clock of 1,250 MHz (10 Gbps effective), whereas the P580’s memory is system shared.

Shading units favor the T600 (640 vs 576), as do ROPs (32 vs 9). The P580 has more TMUs (72 vs 40). Pixel rate is dramatically higher on the T600 (42.72 GPixel/s vs 9.000 GPixel/s), while texture rate is higher on the P580 (72.00 GTexel/s vs 53.40 GTexel/s). FP32 performance favors the T600 (1.709 TFLOPS vs 1,152.0 GFLOPS), and FP16 also favors the T600 (3.418 TFLOPS vs 2.304 TFLOPS). TDP is 15 W for Intel and 40 W for NVIDIA.

Form factor differs: the P580 is an IGP with no slot width, while the T600 is single-slot. The T600 has no power connectors and lists a suggested PSU of 200 W; the P580 has none listed. Bus interface: Ring Bus for Intel, PCIe 3.0 x16 for NVIDIA. Display outputs: motherboard dependent for the P580, 4x mini-DisplayPort 1.4a for the T600. Memory size, type, bus width, and bandwidth all differ (shared vs 4 GB GDDR6, 128-bit, 160.0 GB/s). Release dates are 2015-08-31 for Intel and 2021-04-11 for NVIDIA. The T600 has a predecessor (Quadro Volta) and successor (Workstation Ampere); the P580 has none listed.

The Verdict

The benchmark data is unambiguous: the NVIDIA T600 is the superior performer for any task that stresses the GPU. Its 79.4% lead in Vulkan and 67.4% lead in OpenCL demonstrate a level of compute capability that the Intel Iris Pro P580 cannot approach. The T600’s dedicated 4 GB GDDR6 memory with 160.0 GB/s bandwidth removes the bottleneck of system-shared memory, making it suitable for texture-heavy or memory-bandwidth-sensitive workloads.

The Intel part’s only statistical advantage is its average benchmark score (7,170 vs 7,035), but this is an artifact of test distribution — the P580 has only two scores, both from Geekbench, while the T600 includes lower Passmark DX scores. Real-world performance, as measured by the head-to-head tests, favors the T600 overwhelmingly. The P580’s 15 W TDP and integrated nature make it a low-power solution for basic graphics or legacy systems, but it is end-of-life and lacks the hardware resources (fewer ROPs, lower clocks, shared memory) to compete.

For a user who needs reliable workstation graphics, multi-display output (4x mini-DisplayPort), or modern API support (Vulkan 1.4), the T600 is the clear choice. The P580 is only relevant for systems where a discrete GPU cannot be installed or where power draw must stay minimal. The data does not support any scenario where the P580 outperforms the T600 in compute or graphics benchmarks.

Where Each One Wins

NVIDIA T600 wins on: Geekbench OpenCL (27,875 vs 9,082), Geekbench Vulkan (25,580 vs 5,258), FP32 throughput (1.709 TFLOPS), FP16 throughput (3.418 TFLOPS), pixel rate (42.72 GPixel/s), memory bandwidth (160.0 GB/s), dedicated memory capacity (4 GB), ROP count (32 vs 9), Vulkan API version (1.4 vs 1.3), display outputs (4x mini-DisplayPort), and clock speeds (boost 1,335 MHz vs 1,000 MHz).

Intel Iris Pro Graphics P580 wins on: TMU count (72 vs 40), texture rate (72.00 GTexel/s vs 53.40 GTexel/s), TDP (15 W vs 40 W), and average benchmark score (7,170 vs 7,035). The texture rate advantage is notable for texture-heavy workloads, but the T600’s higher pixel rate and memory bandwidth compensate in most scenarios. The lower TDP makes the P580 suitable for fanless or ultra-portable designs, but the T600’s 40 W is still modest for a discrete card. The average score win is marginal and not reflected in any head-to-head test.