Intel UHD Graphics P630 vs NVIDIA Quadro K620 Comparison

FAQ

Q: How do the average benchmark scores of the NVIDIA Quadro K620 and Intel UHD Graphics P630 compare?

A: The NVIDIA Quadro K620 records an average benchmark score of 6282, while the Intel UHD Graphics P630 records 5370. This places the Quadro K620 in the 36th percentile of all GPUs, compared to the 31st percentile for the Intel integrated solution.

Q: Which GPU wins in the OpenCL benchmark, and by how much?

A: The NVIDIA Quadro K620 wins the Geekbench OpenCL test with a score of 6693 against the Intel UHD Graphics P630's 5111. This represents a 31% advantage for the NVIDIA card.

Q: Is the Vulkan performance gap similar to the OpenCL gap?

A: No, the Vulkan gap is much narrower. The Quadro K620 scores 5870 versus 5628 for the Intel UHD Graphics P630, a lead of only 4.3%. This suggests the integrated GPU is comparatively stronger in Vulkan workloads than in OpenCL.

Q: What are the nearest rivals to each GPU according to the database?

A: The Quadro K620's closest rivals include the NVIDIA GeForce RTX 5070 Ti SUPER and RTX 4070 Ti SUPER AD102 (both with an average score of 6270, within 0.2%), plus the AMD Radeon R7 M350 and Radeon Pro WX 4100. The Intel UHD Graphics P630's nearest rivals are the AMD Radeon R7 M445 (5358, 0.2% ahead) and the NVIDIA GeForce 840M (5322, 0.9% behind).

Q: Which GPU has the higher boost clock speed?

A: The NVIDIA Quadro K620 has a boost clock of 1124 MHz, while the Intel UHD Graphics P630 boosts to 1200 MHz. Despite the Intel part's higher boost frequency, the Quadro K620 delivers higher overall performance in both recorded benchmarks.

Q: How do the shading unit counts differ between the two?

A: The NVIDIA Quadro K620 contains 384 shading units, exactly double the 192 shading units found in the Intel UHD Graphics P630. The Intel part compensates with a higher boost clock, but the raw shading unit advantage favors NVIDIA.

Architecture Differences

The NVIDIA Quadro K620 is built on the GM107 chip using the Maxwell architecture, manufactured on a 28 nm process at TSMC. The Intel UHD Graphics P630 uses the Comet Lake GT2 chip with the Generation 9.5 architecture, fabricated on Intel's 14 nm+++ process. These process nodes represent different design philosophies: NVIDIA's older but mature 28 nm node versus Intel's refined 14 nm+++ variant.

The transistor counts reveal a substantial difference in complexity. The Quadro K620 packs 1,870 million transistors into a 148 mm² die, yielding a transistor density of 12.6 million per square millimeter. Intel's integrated solution does not report transistor count or die size in the database, reflecting its nature as a subsystem within a larger processor rather than a standalone chip.

Clock behavior differs significantly. The Quadro K620 operates at a base clock of 1058 MHz with a boost to 1124 MHz, while the Intel UHD Graphics P630 starts at a much lower 350 MHz base but boosts to 1200 MHz. This indicates that the Intel part is designed to ramp up aggressively under load, whereas the NVIDIA card maintains a more consistent clock range.

Memory architecture presents a fundamental contrast. The Quadro K620 has dedicated 2 GB of DDR3 memory on a 128-bit bus, delivering 28.80 GB/s of bandwidth. The Intel UHD Graphics P630 uses System Shared memory, with its bus width also marked as System Shared and bandwidth described as System Dependent. This means the Intel GPU's memory performance varies with the host system's configuration, while the NVIDIA card has fixed, dedicated resources.

The shading unit count strongly favors NVIDIA: 384 versus 192. Texture mapping units are equal at 24 each, but the raster operation units differ dramatically, with the Quadro K620 featuring 16 ROPs versus just 3 on the Intel part. This ROP disparity directly impacts pixel throughput, which the numbers confirm: 17.98 GPixel/s for NVIDIA versus 3.600 GPixel/s for Intel.

Feature support shows generational differences. The Intel UHD Graphics P630 supports DirectX 12 (12_1) and Vulkan 1.3, while the Quadro K620 supports DirectX 12 (11_0) and Vulkan 1.4. Both support OpenGL 4.6. The Intel part also lists FP16 capability at 921.6 GFLOPS with a 2:1 ratio, a feature absent from the NVIDIA specifications.

Power and physical characteristics differ as expected for a discrete versus integrated GPU. The Quadro K620 has a 45 W TDP, is single-slot, requires no power connectors, and suggests a 200 W power supply. The Intel UHD Graphics P630 has a 15 W TDP and is an IGP with no separate slot or power requirements.

Head-to-Head Benchmarks

The Geekbench OpenCL test provides the clearest separation between these two GPUs. The NVIDIA Quadro K620 scores 6693, while the Intel UHD Graphics P630 scores 5111. This 31% delta is the largest performance gap recorded between them. The OpenCL workload appears to favor the Quadro K620's dedicated memory and higher ROP count, allowing it to pull far ahead of the integrated solution.

The Geekbench Vulkan test tells a different story. Here the Quadro K620 scores 5870 against 5628 for the Intel part, a margin of just 4.3%. This compressed gap suggests that Vulkan's lower-level API overhead benefits the Intel architecture more than OpenCL does. The Intel UHD Graphics P630's higher boost clock of 1200 MHz may help close the gap in this workload, as can the integrated GPU's direct access to system memory through the Ring Bus interface.

Looking at the nearest rival data provides context for each GPU's position. The Quadro K620's average score of 6282 places it within 0.2% of the NVIDIA GeForce RTX 5070 Ti SUPER and RTX 4070 Ti SUPER AD102, both scoring 6270. It trails the AMD Radeon R7 M350 by 0.7% and the Radeon Pro WX 4100 by 0.8%. This tight clustering indicates the Quadro K620 sits in a competitive performance band despite its age.

The Intel UHD Graphics P630's average score of 5370 puts it 0.2% behind the AMD Radeon R7 M445 (5358), 0.8% behind the Radeon R7 M365X (5416), and ahead of the NVIDIA GeForce 840M by 0.9% and GeForce 930A by 1%. The Intel part's rival grouping spans a narrow range, similar to the Quadro's situation, though at a lower absolute performance level.

The benchmark data shows a consistent winner across both tests: the NVIDIA Quadro K620 takes 2 wins to 0. However, the margin varies widely by API, which hints that the Intel integrated GPU is not uniformly slower. Its Vulkan result comes within striking distance, suggesting that API selection matters significantly when comparing these two parts.

Specification Differences

The two GPUs differ across nearly every major specification category. The process node shows 28 nm for NVIDIA versus 14 nm+++ for Intel. Transistor count is 1,870 million for the Quadro K620, while the Intel part reports none. Die size is 148 mm² for NVIDIA, with no figure for Intel.

Clock speeds differ substantially: the Quadro K620 has a 1058 MHz base and 1124 MHz boost, while the Intel UHD Graphics P630 has a 350 MHz base and 1200 MHz boost. Memory configurations are entirely different, with 2 GB DDR3 on a 128-bit bus for NVIDIA versus System Shared memory for Intel.

Compute resources show a 384 versus 192 split in shading units, equal TMUs at 24, and a 16 versus 3 ROP disparity. Pixel rate is 17.98 GPixel/s versus 3.600 GPixel/s, while texture rate is 26.98 GTexel/s for NVIDIA versus 28.80 GTexel/s for Intel. FP32 performance is 863.2 GFLOPS versus 460.8 GFLOPS, with only Intel listing FP16 at 921.6 GFLOPS.

The TDP figures are 45 W versus 15 W. The bus interface is PCIe 2.0 x16 for NVIDIA and Ring Bus for Intel. Display outputs are 1x DVI and 1x DisplayPort 1.2 for the Quadro, while the Intel part is motherboard dependent. API support differs in DirectX (12_1 for Intel versus 12 (11_0) for NVIDIA) and Vulkan (1.3 versus 1.4).

Release dates are separated by nearly six years: July 2014 for the Quadro K620 and May 2020 for the Intel UHD Graphics P630. Both are marked end-of-life in the database.

The Verdict

The recorded data points to the NVIDIA Quadro K620 as the stronger GPU overall. It wins both head-to-head benchmarks, holds a higher average score of 6282 versus 5370, and sits at a higher percentile ranking of 36 versus 31. The 31% OpenCL advantage is decisive, and even the narrower 4.3% Vulkan lead favors NVIDIA.

The Quadro K620's dedicated 2 GB of DDR3 memory with 28.80 GB/s bandwidth provides a clear structural advantage over the Intel part's system-shared memory. This likely explains much of the OpenCL performance gap. The 384 shading units versus 192, combined with 16 ROPs versus 3, gives NVIDIA a substantial compute and pixel throughput edge.

The Intel UHD Graphics P630 is not without merit. Its Vulkan score of 5628 comes within 4.3% of the Quadro, and its texture rate of 28.80 GTexel/s actually exceeds the NVIDIA card's 26.98 GTexel/s. The 15 W TDP makes it far more power-efficient on paper, and its integrated nature requires no additional power connectors or cooling considerations.

For workloads that rely heavily on OpenCL or pixel processing, the Quadro K620 is the clear choice based on the data. For scenarios where Vulkan performance is primary, the Intel part's closer showing suggests the gap narrows considerably, though it still trails. The Quadro's end-of-life status and 2014 release date mean it offers no future driver support, while the Intel part's 2020 release gives it a more recent baseline.

Where Each One Wins

The NVIDIA Quadro K620 wins in raw compute performance. Its FP32 throughput of 863.2 GFLOPS is nearly double the Intel part's 460.8 GFLOPS. Pixel rate also favors NVIDIA decisively at 17.98 GPixel/s versus 3.600 GPixel/s, making it the better choice for any workload involving heavy rasterization or framebuffer operations.

The Intel UHD Graphics P630 wins in texture throughput, posting 28.80 GTexel/s against 26.98 GTexel/s for the Quadro. This small 1.82 GTexel/s advantage suggests the Intel part handles texture-heavy operations slightly better per clock, despite having far fewer shading units and ROPs.

The Intel part also wins on power efficiency. Its 15 W TDP is one-third of the Quadro's 45 W, meaning it draws significantly less power for the performance it delivers. For systems where power budgets are tight or cooling is limited, the integrated solution is the more practical choice.

The Quadro K620 wins on dedicated memory. Its 2 GB of DDR3 with a fixed 128-bit bus and 28.80 GB/s bandwidth provides consistent performance regardless of system configuration. The Intel part's system-shared memory means its bandwidth is system dependent, introducing variability that could hurt performance in memory-constrained configurations.

The Intel UHD Graphics P630 wins on API modernity. Its DirectX 12 (12_1) support exceeds the Quadro's 12 (11_0), and its FP16 capability at 921.6 GFLOPS enables mixed-precision workloads that the Quadro cannot handle. These features may matter for specific modern applications despite the overall performance deficit.

The Quadro K620 wins on Vulkan compatibility, supporting version 1.4 versus 1.3 for Intel. This newer Vulkan version may provide access to additional extensions and optimizations in Vulkan-based applications.