Intel UHD Graphics P630 vs NVIDIA GeForce GTX 770M Comparison

The NVIDIA GeForce GTX 770M and the Intel UHD Graphics P630 represent two very different approaches to graphics processing, one a dedicated mobile GPU and the other an integrated solution. The recorded data shows a decisive performance gap between them, but each part holds a specific place in the hardware landscape. The GTX 770M is a dedicated Kepler-based processor designed for gaming laptops, while the P630 is an integrated graphics engine built into Intel's Comet Lake processors, intended for basic display output and light workloads.

Where Each One Wins

The performance split between these two GPUs is stark, with the dedicated NVIDIA part winning the only directly comparable benchmark. In the Geekbench OpenCL test, the GTX 770M scores 8552, while the Intel UHD Graphics P630 scores 5111. This gives the NVIDIA GPU a 67.3% advantage in raw compute performance, making it the clear winner for any task that leverages general-purpose GPU compute. The GTX 770M also holds a significant hardware advantage across nearly every architectural metric, which translates directly into its benchmark superiority.

The Intel UHD Graphics P630, however, wins in the areas of efficiency and integration. Its thermal design power is 15 W, compared to the GTX 770M's 75 W, meaning it draws far less power while still providing functional graphics output. The P630 is an IGP with a Ring Bus interface, meaning it shares system memory and requires no dedicated VRAM, no power connectors, and no MXM module slot. This makes it the only viable option for systems where a dedicated GPU cannot be physically installed or where power consumption must be minimized. The P630 also supports a newer DirectX feature level, 12_1, compared to the GTX 770M's 12 (11_0), and it supports Vulkan 1.3 versus the GTX 770M's Vulkan 1.2.175, giving it an API compatibility edge for modern software titles.

The benchmark data also shows that the P630 has a Vulkan score of 5628, a test that the GTX 770M does not have recorded. While this is not a direct head-to-head comparison, it indicates that the Intel part has at least one functional benchmark result where it is not trailing by a massive margin. The GTX 770M's Metal score of 3448 is likewise a unique data point, though it belongs to the Apple-specific API and does not apply to the Intel part.

The Verdict

The data unequivocally points to the NVIDIA GeForce GTX 770M as the superior graphics processor for demanding workloads. Its 67.3% lead in OpenCL, combined with its 960 shading units versus the P630's 192, its 80 texture mapping units versus 24, and its 24 ROPs versus 3, means it will outperform the Intel part in any GPU-accelerated application. The GTX 770M also has a dedicated 3 GB of GDDR5 memory on a 192-bit bus with 96.19 GB/s of bandwidth, whereas the P630 relies on system shared memory with bandwidth that is system dependent. For gaming, 3D rendering, or compute tasks, the GTX 770M is the only choice from this comparison.

The Intel UHD Graphics P630 is the correct selection for a different use case entirely. Its 15 W TDP, integrated form factor, and system shared memory make it suitable for office productivity, media playback, and basic desktop acceleration in thin-and-light or embedded systems. It has no power connectors and is motherboard dependent for display outputs, reinforcing its role as a low-power companion to a CPU. The P630's API support, including DirectX 12_1 and Vulkan 1.3, is more current than the GTX 770M's, which may matter for compatibility with newer graphics frameworks, but this does not overcome the massive raw performance deficit.

The average benchmark scores in the database also reflect this hierarchy. The GTX 770M has an average benchmark score of 6000, placing it in the 35th percentile of all GPUs, while the P630 averages 5370, sitting in the 31st percentile. The GTX 770M's nearest rivals include the AMD Radeon RX 6400 at a 0% delta, the NVIDIA RTX PRO 6000 Blackwell Server at 0.1%, and the AMD FirePro W4100 at 0.2%, showing it sits in a cluster of similarly performing parts. The P630's nearest rivals, such as the AMD Radeon R7 M365X at -0.8% and the NVIDIA GeForce 840M at 0.9%, are all older or lower-tier mobile parts, confirming its position at the lower end of the performance spectrum.

Head-to-Head Benchmarks

The single recorded head-to-head benchmark is Geekbench OpenCL, and the result is a landslide. The NVIDIA GeForce GTX 770M scores 8552, compared to the Intel UHD Graphics P630's 5111, a delta of 67.3% in favor of the NVIDIA part. This score reflects the fundamental differences in hardware: the GTX 770M's 960 shading units and 1.530 TFLOPS of FP32 performance dwarf the P630's 192 shading units and 460.8 GFLOPS. The NVIDIA GPU also has a texture rate of 63.76 GTexel/s and a pixel rate of 15.94 GPixel/s, versus the Intel part's 28.80 GTexel/s and 3.600 GPixel/s.

The GTX 770M's memory subsystem is another major factor. Its 3 GB of GDDR5 memory with a 192-bit bus delivers 96.19 GB/s of bandwidth, while the P630's memory bandwidth is listed as system dependent, meaning it is limited by the host system's RAM and bus configuration. This disparity in memory bandwidth directly impacts compute performance, as the GPU must feed its shading units with data from memory. The GTX 770M's memory clock is 1002 MHz, with 4 Gbps effective speed, while the P630's memory clock is tied to the system.

The GTX 770M also benefits from a higher boost clock of 797 MHz versus the P630's 1200 MHz. While the Intel part has a higher frequency, its much smaller execution resource pool means it cannot translate that clock speed into competitive throughput. The GTX 770M's base clock is 706 MHz, and its boost clock is 797 MHz, while the P630's base clock is 350 MHz and its boost clock is 1200 MHz. The GTX 770M's superior pixel rate of 15.94 GPixel/s versus the P630's 3.600 GPixel/s means it can fill a frame buffer over four times faster, which is critical for gaming at higher resolutions.

The GTX 770M's FP32 throughput of 1.530 TFLOPS is over three times the P630's 460.8 GFLOPS. The Intel part does have an FP16 rating of 921.6 GFLOPS with a 2:1 ratio, but the GTX 770M has no recorded FP16 value, and OpenCL results indicate the NVIDIA part is still ahead in real-world compute. The texture rate of 63.76 GTexel/s for the GTX 770M versus 28.80 GTexel/s for the P630 further confirms the NVIDIA part's dominance in fill-rate-limited scenarios.

FAQ

Q: Which GPU has a higher average benchmark score?

A: The NVIDIA GeForce GTX 770M has an average benchmark score of 6000, while the Intel UHD Graphics P630 has an average of 5370.

Q: What is the performance difference in the Geekbench OpenCL test?

A: The GTX 770M scores 8552, and the P630 scores 5111. This gives the GTX 770M a 67.3% lead in that test.

Q: Which GPU supports a newer version of Vulkan?

A: The Intel UHD Graphics P630 supports Vulkan 1.3, while the NVIDIA GeForce GTX 770M supports Vulkan 1.2.175.

Q: How do the shading unit counts compare?

A: The GTX 770M has 960 shading units, while the P630 has 192 shading units.

Q: Which GPU has dedicated video memory?

A: The GTX 770M has 3 GB of GDDR5 memory on a 192-bit bus. The P630 uses system shared memory with system dependent bandwidth.

Q: What is the thermal design power of each GPU?

A: The GTX 770M has a TDP of 75 W, while the P630 has a TDP of 15 W.

Architecture Differences

The architectural divide between these two GPUs is fundamental. The NVIDIA GeForce GTX 770M is built on the GK106 chip using the Kepler architecture, fabricated on a 28 nm process at TSMC. The chip contains 2,540 million transistors on a 221 mm² die, yielding a transistor density of 11.5M per mm². The Intel UHD Graphics P630 uses the Comet Lake GT2 chip with the Generation 9.5 architecture, built on Intel's 14 nm+++ process. The Intel part has no recorded transistor count or die size, reflecting its integrated nature on a larger processor package.

The GTX 770M is a discrete MXM module with an MXM-B (3.0) bus interface, meaning it is a removable, upgradeable card designed for gaming laptops. It has no power connectors listed, drawing all its power through the MXM slot. The P630 is an IGP with a Ring Bus interface, meaning it is built into the CPU die and shares the processor's power delivery and cooling. Its display outputs are motherboard dependent, while the GTX 770M's outputs are portable device dependent, reflecting its laptop-only usage.

Compute resources differ drastically. The GTX 770M features 960 shading units, 80 TMUs, and 24 ROPs, while the P630 has 192 shading units, 24 TMUs, and only 3 ROPs. The GTX 770M's pixel rate is 15.94 GPixel/s, and its texture rate is 63.76 GTexel/s. The P630's pixel rate is 3.600 GPixel/s, and its texture rate is 28.80 GTexel/s. The FP32 performance of the GTX 770M is 1.530 TFLOPS, whereas the P630 delivers 460.8 GFLOPS, with the Intel part offering FP16 at 921.6 GFLOPS with a 2:1 ratio.

Memory architecture is another major differentiator. The GTX 770M uses 3 GB of GDDR5 with a 192-bit bus, providing 96.19 GB/s of bandwidth and a 1002 MHz memory clock with 4 Gbps effective speed. The P630 uses system shared memory, with its bandwidth listed as system dependent, meaning performance varies based on the host platform's RAM configuration. The GTX 770M's memory is dedicated, while the P630 must contend with the CPU for memory access.

API support shows a generational split. The GTX 770M supports DirectX 12 (11_0), OpenGL 4.6, and Vulkan 1.2.175. The P630 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. The Intel part's higher DirectX feature level and newer Vulkan version are notable, but they do not compensate for the NVIDIA part's overwhelming hardware advantage. The GTX 770M is from the GeForce 700M generation, released on May 29, 2013, while the P630 is from the HD Graphics-W (Comet Lake) generation, released on May 12, 2020. The GTX 770M is now end-of-life, as is the P630, with the GTX 770M having a predecessor in the GeForce 600M series and a successor in the GeForce 800M series. The P630 has no recorded predecessor or successor.