AMD Radeon Vega 10 Mobile vs Intel UHD Graphics 750 Comparison

# Intel UHD Graphics 750 vs AMD Radeon Vega Mobile 10

The recorded benchmark data places these two integrated graphics processors in the same performance neighborhood, with the Intel UHD 750 holding a narrow edge in the primary OpenCL workload. The Intel part posts a Geekbench OpenCL score of 6,743 against 6,476 for the AMD Radeon Vega Mobile 10, a delta of 4.1%. In the broader database, the Intel part ranks at the 40th percentile among all GPUs, while the AMD part sits at the 37th percentile. Their average recorded scores, 7,441 and 6,476 respectively, are separated by less than 15%, which places them in direct contention in any workload that leans on compute throughput.

Head-to-Head Benchmarks

The sole head-to-head result in the database is the Geekbench OpenCL test. Intel UHD Graphics 750 scored 6,743, while the AMD Radeon Vega Mobile 10 scored 6,476. Intel wins this run by 267 points, or 4.1%. This margin is consistent with the overall average score gap, where Intel leads by approximately 965 points (7,441 vs 6,476 in average terms). The OpenCL workload is the only recorded measurement, so conclusions about gaming or synthetic graphics workloads must be drawn from the architecture and feature set rather than direct measured results.

The nearest rival data reinforces the narrow band. The AMD Radeon HD 8850M averages 7,447 (0.1% behind the Intel part), the AMD Radeon R7 350 averages 7,425 (0.2% behind), and the NVIDIA GeForce GTX 1650 averages 7,472 (0.4% behind). Only the Intel Arc A310 sits ahead among close rivals, averaging 7,550, which is 1.4% higher. The positioning is tight: a difference of a few percent separates four of the five closest parts, indicating that driver optimization, thermal behavior, and platform-specific factors can outweigh raw silicon math.

Where Each One Wins

Based on recorded wins, the Intel part wins the only head-to-head test. There is no recorded test in the database where the AMD part comes out ahead. However, the AMD Radeon Vega Mobile 10 has structural advantages that can translate to wins in different workloads. Its GCN 5.0 architecture with Vega IGP silicon (Raven Ridge-M) has 640 shading units, 40 texture memory units, and 8 ROPs, versus Intel's 256 shading units, 16 texture memory units, and 8 ROPs. A higher shading unit count (640 vs 256) and texture unit count (40 vs 16) gives the AMD design a 2.5x and 2.5x respective advantage on paper. The Vega's pixel rate is 10.41 GPixel/s and texture fill rate 52.04 GTexel/s, both above Intel's 10.40 GPixel/s and 20.80 GTexel/s. Those specifications matter in shader-heavy games and general desktop composition, where the AMD part can push more parallel work per clock.

The Intel part hits back with a much higher boost clock: 1,300 MHz vs 1,301 MHz for the AMD part (the data records 1,300 MHz for Intel boost, 1,301 MHz for AMD boost). Clock-for-clock, the Intel architecture completes more than twice (2:1) of the FP32 and FP16 rate: 665.6 GFLOPS vs 1,665 TFLOPS, and in FP16 (2:1) of 1,331.2 GFLOPS. The Intel's 14 nm+++ process versus AMD's 14 nm also points to a power efficiency advantage, but the database does not include measured TDP for the AMD part. The Intel part is rated at 15 W TDP. It is a dedicated IGP with a ring bus interface, while the AMD uses an IGP bus interface.

For direct use cases, the AMD part is more appropriate when the workload is heavy in pixel fill and texture: its 52.04 GTexel/s texture rate is 2.5x the Intel's 20.80 GTexel/s, and its 10.41 GPixel/s pixel rate is essentially identical to Intel's. The AMD part is the choice for older DX11-style feature levels with the GCN architecture, since it has a higher shading unit count. However, the Intel part is the one to pick for OpenCL compute and for workloads that run on the newer, more efficient architecture: its cache structure and the 1,300 MHz boost help to hit 6,743 in Geekbench, which is a 4.1% advantage. The database also shows a production status: Intel is end-of-life, while AMD is end-of-life, and both have no successor listed.

Architecture Differences

The Intel UHD Graphics 750 is built on the Rocket Lake chip, using Intel's Generation 12.1 architecture, and made on Intel's 14 nm+++ process. It integrates 256 shading units, 16 texture memory units, and 8 ROPs. Its clock base is 300 MHz, boosting to 1,300 MHz. The memory subsystem is shared with the system, with bus width and bandwidth dependent on the system. API support includes DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. It was released March 29, 2021.

The AMD Radeon Vega Mobile 10 uses the Raven-M chip, GCN 5.0 architecture, from GlobalFoundries at 14 nm. It packs 640 shading units, 40 texture units, and 8 ROPs. The base clock is 300 MHz, boost 1,301 MHz. Memory is shared system memory, again with system dependent bus width and bandwidth. API support covers DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.3. It was released January 7, 2019, and its predecessor is GCN 3.0 IGP. Production status is end-of-life.

Where they differ in architecture, the AMD part has 2.5x more shading units and 2.5x more texture units, and its fixed-function throughput is rated higher. The Intel part has a higher boost clock (1,300 vs 1,301 MHz, a negligible 1 MHz difference) and is built on a smaller process (14nm vs 14nm). The transistor count tells a bigger story: the AMD die integrates 4,940 million transistors on a 210 mm² die, for a density of 23.5M/mm², while the Intel die transistor count is not recorded in the database.

Specification Differences

Both parts share many specifications: system-shared memory (type, size, bus width, bandwidth), no power connectors, motherboard/portable dependent display outputs, and no dedicated launch MSRP. The differences are listed below.

| Field | Intel UHD Graphics 750 | AMD Radeon Vega Mobile 10 |

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

| Chip | Rocket Lake | Raven-M |

| Architecture | Generation 12.1 | GCN 5.0 |

| Process | 14nm+++ | 14nm |

| Transistors | Not recorded | 4,940 million |

| Die size | Not recorded | 210mm² |

| Transistor density | Not recorded | 23.5M/mm² |

| Shading units | 256 | 640 |

| TMUs | 16 | 40 |

| Boost clock | 1300 MHz | 1301 MHz |

| Pixel rate | 10.40 GPixel/s | 10.41 GPixel/s |

| Texture rate | 20.80 GTexel/s | 52.04 GTexel/s |

| FP32 | 665.6 GFLOPS | 1.665 TFLOPS |

| FP16 | 1,331.2 GFLOPS (2:1) | 3.331 TFLOPS (2:1) |

| Slot width | IGP | IGP |

| Bus interface | Ring Bus | IGP |

FAQ

Q: Which part is faster in synthetic Compute workloads?

A: The Intel UHD 750 wins the only recorded head-to-head, scoring 6,743 vs 6,476 in Geekbench OpenCL, a 4.1% advantage. Its average score of 7,441 vs 6,476 is also higher.

Q: When does the AMD Radeon Vega Mobile 10 have an edge?

A: In shading and texture-heavy tasks. Its 640 shading units versus 256, 40 TMUs versus 16, and 52.04 GTexel/s vs 20.80 GTexel/s make it stronger for fill-bound workloads, despite lower average compute.

Q: Are the APIs the same?

A: Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan. Intel adds Vulkan 1.4 and AMD 1.3.

Q: What is the power draw?

A: The database lists the Intel part at 15 W TDP. No TDP is recorded for the AMD part.

Q: Are both discontinued?

A: Yes. Intel is end-of-life, AMD is end-of-life. No successors are listed for either.