Intel Iris Pro Graphics P580 vs NVIDIA GeForce GTX 770M Comparison

Head-to-Head Benchmarks

The only direct benchmark comparison in the database is Geekbench OpenCL, and it favors the Intel Iris Pro Graphics P580. The Intel part scores 9,082, while the NVIDIA GeForce GTX 770M scores 8,552. That is a 6.2% advantage for Intel. It is not a landslide, but it is a clear, measurable win in a compute-oriented workload.

The Intel GPU also has additional benchmark data points beyond the head-to-head. Its Geekbench Vulkan score is 5,258, which shows a solid API implementation for a processor-integrated solution. The GTX 770M has a Geekbench Metal score of 3,448, a test the Intel part cannot run due to platform constraints. The average benchmark score across all recorded tests tells a similar story: the Iris Pro P580 averages 7,170, while the GTX 770M averages 6,000. That is a 1,170-point gap, or roughly 19.5% higher average performance for Intel.

The percentile rankings reinforce this. The Intel part sits at the 39th percentile of all GPUs in the database, whereas the GTX 770M is at the 35th percentile. Both are mid-pack parts, but Intel sits visibly higher. The nearest rivals in the database for the Intel GPU include the NVIDIA GeForce GTX 560 SE at 7,171 (0% delta), the GTX 970 at 7,157 (0.2% ahead of Intel), and the AMD Radeon Vega 8 Mobile at 7,203 (0.5% ahead). For the GTX 770M, the nearest rivals are the AMD Radeon RX 6400 at 6,001 (0% delta), the NVIDIA RTX PRO 6000 Blackwell Server at 5,996 (0.1% ahead), and the AMD FirePro W4100 at 5,987 (0.2% ahead). The Intel part is effectively trading blows with much newer discrete GPUs, while the GTX 770M is matched with older workstation and entry-level parts.

The wins tally is one to zero in favor of Intel. That single win is the OpenCL benchmark, which is the only test where both parts have recorded data. The absence of additional shared benchmarks means the database cannot confirm wins in gaming, encoding, or other synthetic workloads, but for the data that exists, Intel holds the edge.

Architecture Differences

The architectural divide here is substantial. The Intel Iris Pro Graphics P580 is built on Intel's Generation 9.0 architecture, specifically the Skylake GT4e chip, and it is manufactured on a 14 nm+ process at Intel's own foundry. The NVIDIA GeForce GTX 770M uses the Kepler architecture with the GK106 chip, built on a 28 nm process at TSMC. The transistor counts differ sharply: the GTX 770M packs 2,540 million transistors on a 221 mm² die, with a transistor density of 11.5M per mm². The Intel part has no transistor or die size data recorded, but its 14 nm+ process is a full node ahead of NVIDIA's 28 nm node, which typically implies higher density and lower power per transistor.

The core configurations are also very different. The Intel GPU has 576 shading units, 72 texture mapping units, and 9 raster output units. The GTX 770M has 960 shading units, 80 TMUs, and 24 ROPs. NVIDIA has a clear advantage in raw core counts: 66% more shading units, 11% more TMUs, and 166% more ROPs. Yet the Intel part still wins the OpenCL benchmark, which suggests that its architecture is more efficient per core, or that the memory subsystem and driver overhead play a larger role than raw counts.

Clock speeds tell a contrasting story. The Intel Iris Pro P580 has a base clock of 350 MHz and a boost clock of 1,000 MHz. The GTX 770M has a base clock of 706 MHz and a boost clock of 797 MHz. Intel boosts about 25% higher, while NVIDIA runs a much higher base clock. The memory situation is where the two diverge most dramatically. The Intel GPU uses system shared memory, with the type, bus width, and capacity all listed as system dependent. The GTX 770M has dedicated 3 GB of GDDR5 memory on a 192-bit bus, delivering 96.19 GB/s of bandwidth. The Intel part's bandwidth is listed as system dependent, meaning it relies entirely on the host platform's memory configuration.

Pixel and texture rates reflect the ROP and TMU counts. The GTX 770M outputs 15.94 GPixel/s versus Intel's 9.00 GPixel/s. Texture rate goes the other way: Intel delivers 72.00 GTexel/s against NVIDIA's 63.76 GTexel/s. FP32 compute favors NVIDIA at 1.530 TFLOPS versus Intel's 1,152.0 GFLOPS, but Intel has a recorded FP16 capability of 2.304 TFLOPS (2:1), which NVIDIA does not list at all. Power consumption is heavily lopsided: the Intel IGP is rated at 15 W, while the GTX 770M is rated at 75 W. That is a fivefold difference in power draw.

API support is another differentiator. The Intel part supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The GTX 770M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. Intel has the newer Vulkan version and the higher DirectX feature level. The bus interface differs as well: Intel uses a Ring Bus, while NVIDIA uses MXM-B (3.0). The GTX 770M is an MXM module, meaning it is a replaceable mobile graphics card, while the Intel part is an IGP integrated into the processor package.

The Verdict

The data points to the Intel Iris Pro Graphics P580 as the stronger part in compute workloads. It wins the only shared benchmark by 6.2%, has a higher average benchmark score by 1,170 points, and sits four percentile points higher in the database ranking. It does all of this while consuming 15 W against the GTX 770M's 75 W, a fivefold efficiency advantage. The Intel part also supports newer API versions, including Vulkan 1.3 and DirectX 12 (12_1), versus the GTX 770M's Vulkan 1.2.175 and DirectX 12 (11_0).

The GTX 770M is not without merit. It has far higher pixel throughput at 15.94 GPixel/s, a dedicated 3 GB GDDR5 frame buffer with 96.19 GB/s bandwidth, and more shading units, TMUs, and ROPs. Its FP32 compute is higher at 1.530 TFLOPS. These are traits that typically matter for rasterization-heavy workloads like gaming. But the recorded benchmark data does not include a gaming test, and the only compute test available favors Intel.

For a system builder or buyer, the choice depends on what the hardware will do. If the workload is OpenCL compute, the Intel part wins outright. If the workload requires dedicated memory bandwidth and high pixel fill rates, the GTX 770M has the architectural edge. But the database's overall average score, which aggregates all recorded benchmarks, strongly favors Intel. The 39th percentile versus 35th percentile ranking is a small but consistent gap.

The GTX 770M's nearest rival, the AMD Radeon RX 6400, scores 6,001, which is essentially identical. The Intel part's nearest rival, the GTX 560 SE, scores 7,171, and the GTX 970 scores 7,157. The Intel part is competing with desktop GPUs from a much higher tier, while the GTX 770M is matched with entry-level and older parts. That context matters. The Intel IGP is punching above its class, while the GTX 770M is holding its historical position.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The Intel Iris Pro Graphics P580 scores 9,082 in Geekbench OpenCL, while the NVIDIA GeForce GTX 770M scores 8,552. Intel leads by 6.2%.

Q: How much memory does each GPU have?

A: The GTX 770M has 3 GB of dedicated GDDR5 memory on a 192-bit bus with 96.19 GB/s bandwidth. The Intel Iris Pro P580 uses system shared memory, with capacity, type, and bus width all dependent on the host system.

Q: What is the power consumption difference?

A: The Intel Iris Pro P580 is rated at 15 W, while the GTX 770M is rated at 75 W. Intel draws one-fifth the power.

Q: Which GPU has better API support?

A: Intel supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. NVIDIA supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. Intel has the newer DirectX feature level and Vulkan version.

Q: What is the average benchmark score for each?

A: The Intel part averages 7,170 across all recorded benchmarks, while the GTX 770M averages 6,000. That is a 1,170-point gap in Intel's favor.

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

A: The Intel Iris Pro P580 is at the 39th percentile, and the GTX 770M is at the 35th percentile.

Where Each One Wins

The Intel Iris Pro Graphics P580 wins in compute performance. The OpenCL benchmark, the only shared test, goes to Intel by 6.2%. The average benchmark score, which includes additional Vulkan and Metal tests, also favors Intel by a wide margin. The Intel part wins on efficiency, with a 15 W TDP versus 75 W, and it wins on API modernity, supporting Vulkan 1.3 and DirectX 12 (12_1). For users running OpenCL workloads, Vulkan applications, or any compute task that leverages the 2.304 TFLOPS FP16 throughput, the Intel part is the clear choice.

The GTX 770M wins on raw rasterization hardware. It has 960 shading units versus 576, 24 ROPs versus 9, and 80 TMUs versus 72. Its pixel rate is 15.94 GPixel/s against Intel's 9.00 GPixel/s. The dedicated 3 GB GDDR5 frame buffer with 96.19 GB/s bandwidth is a major advantage for texture-heavy workloads, since it does not compete with the CPU for memory access. The GTX 770M's FP32 compute of 1.530 TFLOPS also exceeds Intel's 1,152.0 GFLOPS. For scenarios where pixel fill rate, dedicated memory, or raw FP32 throughput matters more than overall compute efficiency, the GTX 770M has the architectural edge.

The use-case split is sharp. The Intel part suits low-power systems, compact form factors, and compute-oriented tasks where API support and efficiency are priorities. The GTX 770M suits systems that need a replaceable MXM module with dedicated graphics memory and high pixel throughput. The database's recorded data, however, favors Intel in every shared benchmark, so the practical recommendation leans toward the Iris Pro P580 for most mixed workloads.