Intel HD Graphics P530 vs NVIDIA GeForce GTX 650 Comparison

# Intel HD Graphics P530 vs. NVIDIA GeForce GTX 650: A Data-Driven Comparison

Benchmark results for the Intel HD Graphics P530 and the NVIDIA GeForce GTX 650 show two very different contenders. The P530 is an integrated graphics solution built into Intel's Skylake-generation processors, while the GTX 650 is a discrete add-in card based on NVIDIA's Kepler architecture. Their recorded scores place them in different performance tiers, with the Intel part ranking in the 26th percentile among all GPUs in our database, and the GTX 650 landing in the 22nd percentile. On average, the P530 scores 4560 points, which is 19.3% higher than the GTX 650's average of 3823 points.

Head-to-Head Benchmarks

The database includes two specific head-to-head tests between these two parts. The results are split, with the Intel solution winning both recorded matchups. In the Geekbench OpenCL test, the P530 scored 4549 against 4545 for the GTX 650, a margin of 0.1%. In the Geekbench Vulkan test, the P530 again came out on top, scoring 4571 against 4524, a 1% advantage. These are narrow wins, but they show the integrated part can keep pace with, and slightly edge, this particular discrete competitor in compute-heavy workloads.

Looking at the broader rival set, the P530 sits close to several other parts. The AMD Radeon RX 560 scores 4569, which is 0.2% higher than the Intel part. The AMD Radeon R5 Graphics scores 4577, a 0.4% advantage. On the other side, the AMD FirePro W4190M scores 4505, which is 1.2% lower than the Intel part, and the NVIDIA Quadro M3000M scores 4621, a 1.3% deficit. These figures indicate a tightly clustered group of parts where the P530 can trade places with its immediate rivals depending on the workload.

Against the GTX 650, the margins are more substantial. The NVIDIA part holds a 1.9% advantage over the Intel part in the database's average score. The Intel part, in turn, leads the NVIDIA Quadro 2000 by 1.5% in the average. The closest competitor from the opposing side, the AMD Radeon RX 550, trails the Intel part by 0.8%.

Architecture Differences

The silicon fundamentals of these two parts are entirely different. Intel's chip is codenamed Skylake GT2 and is built on a "14 nm+" process at Intel's foundry, with a die size of 123 mm². It belongs to the Generation 9.0 graphics architecture. The NVIDIA part, in contrast, is built on the Kepler architecture, uses the GK106 chip, and is manufactured at TSMC on a 28 nm process. The die size is substantially larger at 221 mm², and it packs 2,540 million transistors, for a transistor density of 11.5 million per mm².

The Intel part has 192 shading units, 16 texture mapping units, and 3 ROPs. The NVIDIA part has double the shading units at 384, double the TMUs at 32, and more than five times the ROPs at 16. That gives the discrete part a clear advantage in raw geometry throughput, though the integrated part counters with higher clock frequencies. The Intel part's base clock is 350 MHz, which can boost up to 1000 MHz. The GTX 650's base and boost clocks are not listed in the database. Memory speed tells a similar story: the Intel part uses system-shared memory, while the GTX 650 uses 1024 MB of GDDR5 on a 128-bit bus, producing 80.00 GB/s of bandwidth. The Intel part's memory bandwidth is system dependent and shared with the CPU.

The Intel part also has a much lower power draw: its 15 W TDP is a fraction of the 65 W TDP of the GTX 650. The integrated part is a single slot solution, while the discrete part requires one 6-pin power connector and a suggested 250 W power supply. The bus interface is the Ring Bus on the Intel side, while the NVIDIA part uses PCIe 3.0 x16. Display outputs are also different: the Intel part is motherboard dependent, while the NVIDIA part has 1x DVI and 2x DisplayPort 1.2 connections.

FAQ

Q: Which is faster in raw compute, the Intel HD P530 or the NVIDIA GTX 650?

A: The recorded FP32 performance is higher on the Intel part. The Intel HD P530 records 384.0 GFLOPS, while the NVIDIA GeForce GTX 650 is rated at 812.5 GFLOPS. However, the Intel part's synthetic average is actually 66 points higher.

Q: Which GPU supports DirectX 12?

A: Both do. The Intel HD Graphics P530 supports DirectX 12 (12_1) and OpenGL 4.6. The NVIDIA GeForce GTX 650 supports DirectX 12 (11_0) and OpenGL 4.6.

Q: What is the memory configuration of the GTX 650?

A: The GeForce GTX 650 has 1024 MB of GDDR5 memory on a 128-bit bus, producing 80.00 GB/s of bandwidth.

Q: Does the Intel P530 use system RAM?

A: Yes, the Intel HD Graphics P530 uses System Shared memory, with bandwidth and bus width being system dependent.

Q: Which GPU has higher pixel and texture throughput?

A: The NVIDIA GeForce GTX 650 is rated at 8.464 GPixel/s for pixel fill and 33.86 GTexel/s for texture. The Intel HD P530 is rated at 3.000 GPixel/s and 16.00 GTexel/s. The NVIDIA part is more than double in both measured rates.

Q: What are the power requirements for the two cards?

A: The Intel HD P530 is rated at 15 W TDP and uses a motherboard dependent power connection. The GTX 650 is rated at 65 W and requires a 250 W power supply plus a 6-pin PCIe power connector.

Specification Differences

| Specification | Intel HD P530 | NVIDIA GTX 650 |

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

| Process node | 14 nm+ | 28 nm |

| Die size | 123 mm² | 221 mm² |

| Shading units | 192 | 384 |

| TMUs | 16 | 32 |

| ROPs | 3 | 16 |

| Pixel fillrate | 3.000 GPixel/s | 8.464 GPixel/s |

| Texture fillrate | 16.00 GTexel/s | 33.86 GTexel/s |

| Memory | System shared | 1024 MB GDDR5 |

| Memory bus | System shared | 128 bit |

| Bandwidth | System dependent | 80.00 GB/s |

| Power | 15 W | 65 W |

| Interface | Ring Bus | PCIe 3.0 x16 |

| Power connectors | None | 1x 6-pin |

| Outputs | Motherboard dependent | 1x DVI, 2x DisplayPort 1.2 |

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

| OpenGL | 4.6 | 4.6 |

| Vulkan | 1.3 | 1.2.175 |

The Verdict

The data points in a clear direction. For anyone performing general compute workloads, especially OpenCL or Vulkan based tasks, the Intel HD Graphics P530 is the better pick. Its average score is 66% higher, and it won both head-to-head tests against the GTX 650. It also offers a lower power envelope, a smaller feature process, and a more modern API feature set.

The NVIDIA GeForce GTX 650, on the other hand, is the choice when the workload is heavy on traditional graphics rasterization. Its raw shading, TMU, and ROP counts are double, double, and five times higher, respectively. That results in a pixel fill rate and texture fill rate that are more than double the Intel part. It also has dedicated GDDR5 memory with dedicated bandwidth, which helps in gaming scenarios where textures are large and complex.

Where Each One Wins

Pick the Intel HD Graphics P530 if you need:

• A modern, efficient, low-power integrated graphics solution with no extra power.
• OpenCL or Vulkan tasks that produce higher synthetic scores.
• A smaller footprint and a motherboard-dependent output (the GPU does not have any display output itself).
• DirectX 12_1 feature support and a more recent API feature set.

Pick NVIDIA GeForce GTX 650 if you need:

• Maximum raw rasterization performance: the data shows 2x more shading units, TMUs and ROPs.
• A dedicated fast GDDR5 memory with 80.00 GB/s of bandwidth.
• An excellent choice for classic 3D gaming with high resolution.
• A discrete board that plugs into a PCIe 3.0 x16 slot with a 6-pin power connector, applicable to an existing system.

For those who want a single integrated GPU that uses little power and solid compute in synthetic tests, the Intel HD Graphics P530 is the measured winner. For workloads that are mainly 3D rendering, and where a dedicated discrete card is already in the budget, the GeForce GTX 650 is the stronger option. The choice in the end is between a power-efficient, compute-focused integrated graphics or a more powerful rasterization discrete solution.