Intel UHD Graphics 770 Mobile vs NVIDIA H20 NVL16 Comparison

FAQ

Q: What are the two products compared here?

A: The comparison is between the Intel UHD Graphics 770 Mobile, an integrated GPU from Intel based on the Raptor Lake chip and Generation 12.2 architecture, and the NVIDIA H20 NVL16, a server-grade accelerator from NVIDIA based on the GH100 chip and Hopper architecture.

Q: What is the process node and foundry for each GPU?

A: The Intel UHD Graphics 770 Mobile is built on a 10 nm process at Intel, while the NVIDIA H20 NVL16 is built on a 5 nm process at TSMC.

Q: How much memory does each GPU have?

A: The Intel UHD Graphics 770 Mobile uses System Shared memory, meaning its memory size, type, and bus width are all dependent on the host system. The NVIDIA H20 NVL16 has 96 GB of HBM3 memory with a 6144-bit bus and 4.03 TB/s bandwidth.

Q: What are the power requirements for each GPU?

A: The Intel UHD Graphics 770 Mobile has a TDP of 15 W and is an IGP (integrated graphics processor). The NVIDIA H20 NVL16 has a TDP of 400 W, is an SXM Module, and has a suggested PSU of 800 W.

Q: What are the release dates?

A: The Intel UHD Graphics 770 Mobile was released on 2023-01-03, while the NVIDIA H20 NVL16 was released on 2025-09-01. Both are currently listed as Active in production status.

Q: What is the successor for each GPU?

A: The Intel UHD Graphics 770 Mobile has Arc Graphics-M as its successor. The NVIDIA H20 NVL16 has Server Blackwell as its successor, with Server Ada as its predecessor.

Architecture Differences

The two GPUs represent fundamentally different design philosophies and market positions. The Intel UHD Graphics 770 Mobile is an integrated graphics solution embedded in a Raptor Lake processor, targeting mobile and low-power systems. Its architecture is Generation 12.2, built on Intel's 10 nm process. The NVIDIA H20 NVL16 is a discrete server accelerator built on the Hopper architecture, manufactured by TSMC on a 5 nm process, and designed for high-throughput data center workloads.

The transistor counts and die sizes differ dramatically. The NVIDIA H20 NVL16 contains 80,000 million transistors on a 814 mm² die, resulting in a transistor density of 98.3M / mm². The Intel UHD Graphics 770 Mobile does not have published transistor or die size data in the database. This difference alone explains the gulf in raw compute resources.

Shader resources are heavily skewed toward the NVIDIA part. The Intel GPU has 256 shading units, 16 TMUs, and 8 ROPs. The NVIDIA H20 NVL16 has 9,984 shading units, 312 TMUs, and 24 ROPs. Furthermore, the NVIDIA part includes 312 tensor cores, a feature entirely absent from the Intel integrated GPU. The Intel part has no ray tracing cores, and the NVIDIA part also does not list RT cores in the database, but the tensor core presence is a defining architectural difference.

Memory architecture is another major divergence. The Intel UHD Graphics 770 Mobile relies on System Shared memory, with bandwidth described as System Dependent. The NVIDIA H20 NVL16 uses 96 GB of HBM3 with a 6144-bit bus and 4.03 TB/s bandwidth. Memory clock differs as well: the Intel GPU lists no dedicated memory clock (System Shared), while the NVIDIA part runs at 1313 MHz with 5.3 Gbps effective.

The clock speeds tell a story of power versus efficiency. The Intel part has a base clock of 300 MHz and a boost of 1600 MHz, while the NVIDIA part has a base of 1830 MHz and a boost of 1980 MHz. The NVIDIA part runs at a much higher sustained clock, which is remarkable given its vastly larger silicon area.

Bus interfaces and physical form factors are distinct. The Intel UHD Graphics 770 Mobile uses a Ring Bus interface and is an IGP with no slot width. The NVIDIA H20 NVL16 is an SXM Module using PCIe 5.0 x16. Display outputs differ as well: the Intel part is Portable Device Dependent, while the NVIDIA part has no outputs at all, indicating it is not intended for display purposes.

API support also separates them. The Intel GPU supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The NVIDIA H20 NVL16 lists N/A for DirectX, OpenGL, and Vulkan, reflecting its server-oriented role where graphics APIs are not the primary interface.

The Verdict

The data indicates two devices built for entirely different tasks. The Intel UHD Graphics 770 Mobile is a low-power integrated solution with 15 W TDP, designed for portable devices where display output is required and system memory is shared. The NVIDIA H20 NVL16 is a 400 W server accelerator with 96 GB of dedicated HBM3, built for compute-heavy data center workloads.

Benchmark scores in the database are zero for both, and percentile versus all GPUs is 50 for both, so no performance ranking can be derived from recorded measurements. However, the specification deltas are overwhelming. The NVIDIA H20 NVL16 has 39 times the shading units (9,984 vs 256), 19.5 times the TMUs (312 vs 16), and 3 times the ROPs (24 vs 8). Its FP32 throughput is 39.54 TFLOPS versus 819.2 GFLOPS, a difference of roughly 48 times. FP16 performance is 79.07 TFLOPS versus 1.638 TFLOPS, a difference of roughly 48 times as well.

For pixel and texture rates, the NVIDIA part delivers 47.52 GPixel/s versus 12.80 GPixel/s, and 617.8 GTexel/s versus 25.60 GTexel/s. The NVIDIA accelerator is 3.7 times faster in pixel rate and 24 times faster in texture rate.

The Intel UHD Graphics 770 Mobile is the only option for systems requiring integrated graphics with display outputs and standard graphics API support. The NVIDIA H20 NVL16 is the only option for server workloads requiring tensor cores, massive HBM3 capacity, and extreme FP32/FP16 throughput. Choosing between them is not a matter of performance preference; it is a matter of target platform and workload type.

Specification Differences

| Specification | Intel UHD Graphics 770 Mobile | NVIDIA H20 NVL16 |

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

| Manufacturer | Intel | NVIDIA |

| Chip | Raptor Lake | GH100 |

| Architecture | Generation 12.2 | Hopper |

| Generation | HD Graphics-M (Raptor Lake) | Server Hopper (Hxx) |

| Process Node | 10 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | Not listed | 80,000 million |

| Die Size | Not listed | 814 mm² |

| Transistor Density | Not listed | 98.3M / mm² |

| Base Clock | 300 MHz | 1830 MHz |

| Boost Clock | 1600 MHz | 1980 MHz |

| Memory Clock | System Shared | 1313 MHz 5.3 Gbps effective |

| Memory Size | System Shared | 96 GB |

| Memory Type | System Shared | HBM3 |

| Memory Bus Width | System Shared | 6144 bit |

| Memory Bandwidth | System Dependent | 4.03 TB/s |

| Shading Units | 256 | 9,984 |

| TMUs | 16 | 312 |

| ROPs | 8 | 24 |

| Tensor Cores | Not listed | 312 |

| Pixel Rate | 12.80 GPixel/s | 47.52 GPixel/s |

| Texture Rate | 25.60 GTexel/s | 617.8 GTexel/s |

| FP32 | 819.2 GFLOPS | 39.54 TFLOPS |

| FP16 | 1.638 TFLOPS (2:1) | 79.07 TFLOPS (2:1) |

| TDP | 15 W | 400 W |

| Slot Width | IGP | SXM Module |

| Suggested PSU | Not listed | 800 W |

| Bus Interface | Ring Bus | PCIe 5.0 x16 |

| Display Outputs | Portable Device Dependent | No outputs |

| DirectX | 12 (12_1) | N/A |

| OpenGL | 4.6 | N/A |

| Vulkan | 1.4 | N/A |

| Release Date | 2023-01-03 | 2025-09-01 |

| Successor | Arc Graphics-M | Server Blackwell |

| Predecessor | Not listed | Server Ada |

Head-to-Head Benchmarks

The database lists no recorded head-to-head benchmark results between these two GPUs, and both have zero average benchmark scores. The wins counter shows zero for each side. Therefore, any performance comparison must be derived from the specification data in the database.

The largest advantage for the NVIDIA H20 NVL16 is in FP32 compute. The NVIDIA part delivers 39.54 TFLOPS, which is approximately 48 times the 819.2 GFLOPS of the Intel UHD Graphics 770 Mobile. This is the single biggest relative gap among compute metrics. FP16 shows a similar ratio: 79.07 TFLOPS versus 1.638 TFLOPS, also roughly 48 times.

Texture rate is another area of massive divergence. The NVIDIA part sustains 617.8 GTexel/s versus 25.60 GTexel/s for the Intel part, a 24-fold difference. This aligns with the 19.5 times more TMUs on the NVIDIA chip. Pixel rate favors NVIDIA as well, at 47.52 GPixel/s versus 12.80 GPixel/s, a 3.7 times advantage, despite the NVIDIA part having only 3 times the ROP count.

Memory bandwidth is not comparable in a direct numeric sense because the Intel part uses System Dependent bandwidth. However, the NVIDIA part's 4.03 TB/s from HBM3 over a 6144-bit bus is a figure unmatched by any integrated solution that shares system memory. The Intel part's memory clock and bus width are both System Shared, meaning they vary by host platform.

Clock speed advantages go to NVIDIA on absolute terms. The base clock of 1830 MHz is 6.1 times the Intel base of 300 MHz, and the boost of 1980 MHz is 1.24 times the Intel boost of 1600 MHz. However, the Intel part operates within a 15 W TDP, whereas the NVIDIA part consumes 400 W. The efficiency per watt is not calculable from the given data, but the power envelope difference is stark.

The Intel UHD Graphics 770 Mobile holds advantages in areas that the NVIDIA part does not address. It supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the NVIDIA H20 NVL16 lists N/A for all three. The Intel part also has display outputs, described as Portable Device Dependent, whereas the NVIDIA part has no outputs. These are functional wins for the Intel product in client and mobile contexts.

The bus interface also favors Intel for integration: Ring Bus is typical for onboard graphics, while the NVIDIA SXM Module relies on PCIe 5.0 x16 and an 800 W suggested PSU. The Intel part needs no external power connectors, while the NVIDIA part is a module-level component for server platforms.

Release timing shows a generational gap. The Intel UHD Graphics 770 Mobile launched on 2023-01-03, and its successor, Arc Graphics-M, is already listed. The NVIDIA H20 NVL16 launched on 2025-09-01, with Server Ada as predecessor and Server Blackwell as successor. The Intel part is an older mobile integrated design, while the NVIDIA part is a newer server accelerator.

In summary, the recorded data shows no direct benchmark overlap, but the specification deltas indicate that the NVIDIA H20 NVL16 dominates in raw compute, memory, and throughput metrics, while the Intel UHD Graphics 770 Mobile dominates in API support, display capability, and power efficiency.