Intel UHD Graphics 730 vs Intel UHD Graphics P630 Comparison

Head-to-Head Benchmarks

The benchmark data is unambiguous: Intel UHD Graphics 730 wins both recorded head-to-head tests against Intel UHD Graphics P630. In Geekbench OpenCL, the 730 scores 5988 against 5111 for the P630, a decisive 17.2% advantage. This is the largest margin between the two in any measured workload. The Vulkan test is closer, with the 730 at 5870 and the P630 at 5628, a 4.3% lead. While the gap narrows under Vulkan, the 730 still comes out ahead in every recorded scenario.

The average benchmark score reinforces this pattern. The 730 posts an average of 5929 across its benchmark suite, while the P630 averages 5370. That difference of roughly 10% in aggregate performance places the two chips in different percentile tiers: the 730 sits at the 33rd percentile of all GPUs, while the P630 sits at the 31st. Both are low-end parts by modern standards, but the 730 is meaningfully higher within that range.

Nearest rival data adds context. The 730's closest competitors are the AMD Radeon HD 8730M (average 5955, delta -0.4%), the NVIDIA Quadro K620M (5957, -0.5%), the AMD Radeon HD 8750M (5970, -0.7%), and the NVIDIA Quadro K4000 (5982, -0.9%). The 730 trails each of these by less than 1%, meaning it performs essentially on par with those older discrete mobile and workstation GPUs. The P630, by contrast, sits just above the AMD Radeon R7 M445 (5358, delta 0.2%) and the NVIDIA GeForce 930A (5317, 1%), while trailing the AMD Radeon R7 M365X (5416, -0.8%) and the NVIDIA GeForce 840M (5322, 0.9%). The P630 is competitive with those parts, but they are all lower-performing than the 730's rival set.

Breaking down the OpenCL result further: the 17.2% delta is not a minor edge. It suggests the 730's architectural changes translate into real compute throughput gains. The Vulkan result, at 4.3%, is smaller but still consistent. In neither API does the P630 manage to close the gap or overtake the 730.

The Verdict

The data says the Intel UHD Graphics 730 is the faster integrated GPU. It wins both head-to-head benchmarks, holds a higher average benchmark score (5929 versus 5370), and ranks higher in the percentile distribution (33rd versus 31st). The 730 also has a higher peak FP32 throughput at 499.2 GFLOPS versus 460.8 GFLOPS, and a higher pixel rate at 10.40 GPixel/s versus 3.600 GPixel/s. These are not subtle differences in the recorded metrics.

However, neither part is a gaming or compute powerhouse. Both sit near the bottom of the GPU percentile rankings. The 730's win is relative, not absolute. It is a better choice for anyone who must rely on integrated graphics and wants the stronger of these two options. The P630's only advantages in the raw data are its higher base clock (350 MHz versus 300 MHz) and its higher texture rate (28.80 GTexel/s versus 15.60 GTexel/s), but those do not translate into benchmark victories. Texture throughput alone cannot compensate for the P630's lower pixel rate and lower overall compute scores.

For system builders comparing these two as drop-in options, the 730 is the one to pick if benchmark performance is the priority. The P630 is the older architecture (Generation 9.5 versus Generation 12.1), and the recorded data reflects that generational gap. There is no scenario in the measurements where the P630 wins.

Where Each One Wins

The Intel UHD Graphics 730 wins in overall compute workloads. Its Geekbench OpenCL score of 5988 is 17.2% higher than the P630's 5111, making it the stronger choice for OpenCL-accelerated tasks. It also wins in Vulkan, scoring 5870 versus 5628, a 4.3% margin. Those two wins cover every benchmark in the database for this pairing.

The P630 does hold one theoretical advantage: texture rate. At 28.80 GTexel/s, it is nearly double the 730's 15.60 GTexel/s. In texture-bound workloads, the P630 could theoretically process more texels per second. But this advantage does not show up in the recorded benchmark results. The P630 also has a higher base clock (350 MHz versus 300 MHz), which could help in lightly loaded scenarios where the GPU does not boost. Its boost clock is lower, however: 1200 MHz versus 1300 MHz for the 730.

The 730 counters with a much higher pixel rate: 10.40 GPixel/s versus 3.600 GPixel/s. This suggests the 730 is better equipped for fill-rate-bound tasks, such as basic 2D compositing or low-resolution rendering. The 730 also has more ROPs: 8 versus 3. The P630 has more TMUs: 24 versus 12. Both have the same shading unit count at 192.

In practical terms, the 730 is the pick for general compute and newer API workloads. The P630's texture rate advantage is real on paper but does not produce a single benchmark win. The 730's higher FP32 and FP16 throughput (499.2 GFLOPS and 998.4 GFLOPS versus 460.8 GFLOPS and 921.6 GFLOPS) reinforces its position as the more capable compute part.

FAQ

Q: Which GPU is faster in OpenCL workloads?

A: The Intel UHD Graphics 730 scores 5988 in Geekbench OpenCL, 17.2% higher than the Intel UHD Graphics P630's 5111.

Q: How do the two compare in Vulkan performance?

A: The 730 scores 5870 in Geekbench Vulkan, 4.3% ahead of the P630's 5628.

Q: Do both GPUs have the same number of shading units?

A: Yes, both have 192 shading units. They also share the same TDP of 15 W and use system-shared memory.

Q: What is the main architectural difference between the two?

A: The 730 uses Generation 12.1 architecture on the Rocket Lake chip, while the P630 uses Generation 9.5 architecture on the Comet Lake GT2 chip. Both are built on Intel's 14 nm+++ process.

Q: Does the P630 have any benchmark win over the 730?

A: No. The recorded head-to-head data shows the 730 winning both tests, with 2 wins for the 730 and 0 for the P630.

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

A: The 730 trails its closest rivals by less than 1%, with the AMD Radeon HD 8730M at 5955 (-0.4%) and the NVIDIA Quadro K620M at 5957 (-0.5%). The P630 is within 1% of the AMD Radeon R7 M445 (5358, delta 0.2%) and the NVIDIA GeForce 840M (5322, delta 0.9%).

Architecture Differences

The two GPUs come from different Intel generations. The Intel UHD Graphics 730 is built on Generation 12.1 architecture and uses the Rocket Lake chip. The Intel UHD Graphics P630 uses Generation 9.5 architecture with the Comet Lake GT2 chip. Both are fabricated on Intel's 14 nm+++ process node and both are produced by Intel's foundry.

The clock behavior differs. The 730 has a base clock of 300 MHz and a boost clock of 1300 MHz. The P630 has a higher base clock at 350 MHz but a lower boost clock at 1200 MHz. This means the P630 starts at a higher frequency but cannot reach the same peak as the 730.

The core configuration differs in TMU and ROP counts. The 730 has 12 texture mapping units and 8 raster output units. The P630 has 24 TMUs and only 3 ROPs. Both have 192 shading units. The TMU count being higher on the P630 gives it a texture rate of 28.80 GTexel/s, while the 730's lower TMU count yields 15.60 GTexel/s. The ROP count being higher on the 730 gives it a pixel rate of 10.40 GPixel/s, far exceeding the P630's 3.600 GPixel/s.

Compute throughput also differs. The 730 delivers 499.2 GFLOPS in FP32 and 998.4 GFLOPS in FP16 (2:1). The P630 delivers 460.8 GFLOPS in FP32 and 921.6 GFLOPS in FP16 (2:1). The 730 is ahead in both precision levels.

The API support is nearly identical, with both supporting DirectX 12 (12_1) and OpenGL 4.6. The Vulkan version differs: the 730 supports Vulkan 1.4, while the P630 supports Vulkan 1.3. Both use a Ring Bus interface and have motherboard-dependent display outputs.

Memory handling is the same for both: system shared memory, system shared type, system shared bus width, and system-dependent bandwidth. Neither has dedicated VRAM. Both are IGP slot width parts with no power connectors and a 15 W TDP.

Specification Differences

The following specifications differ between the Intel UHD Graphics 730 and the Intel UHD Graphics P630:

• Chip: Rocket Lake for the 730, Comet Lake GT2 for the P630.
• Architecture: Generation 12.1 for the 730, Generation 9.5 for the P630.
• Generation label: HD Graphics (Rocket Lake) for the 730, HD Graphics-W (Comet Lake) for the P630.
• Base clock: 300 MHz for the 730, 350 MHz for the P630.
• Boost clock: 1300 MHz for the 730, 1200 MHz for the P630.
• TMUs: 12 for the 730, 24 for the P630.
• Texture rate: 15.60 GTexel/s for the 730, 28.80 GTexel/s for the P630.
• ROPs: 8 for the 730, 3 for the P630.
• Pixel rate: 10.40 GPixel/s for the 730, 3.600 GPixel/s for the P630.
• FP32 throughput: 499.2 GFLOPS for the 730, 460.8 GFLOPS for the P630.
• FP16 throughput: 998.4 GFLOPS (2:1) for the 730, 921.6 GFLOPS (2:1) for the P630.
• Vulkan version: 1.4 for the 730, 1.3 for the P630.
• Release date: 2021-03-29 for the 730, 2020-05-12 for the P630.
• Average benchmark score: 5929 for the 730, 5370 for the P630.
• Percentile versus all GPUs: 33rd for the 730, 31st for the P630.

Fields that are identical include shading units (192), TDP (15 W), slot width (IGP), bus interface (Ring Bus), display outputs (motherboard dependent), memory configuration (all system shared), DirectX support (12 (12_1)), OpenGL support (4.6), process node (14 nm+++), foundry (Intel), and production status (end-of-life for both).