AMD Ryzen AI Embedded P164 vs Intel Core Ultra X9 378H Comparison

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra X9 378H has 16 cores and 16 threads, while the AMD Ryzen AI Embedded P164 has 8 cores and 16 threads. Both processors are listed in the mobile market segment.

Q: How do the two processors compare in average benchmark score?

A: The AMD Ryzen AI Embedded P164 has an average benchmark score of 52901, while the Intel Core Ultra X9 378H has an average score of 47468. The AMD processor ranks in the 91st percentile of all CPUs, while the Intel chip sits in the 89th percentile.

Q: Which processor supports ECC memory?

A: The AMD Ryzen AI Embedded P164 supports ECC memory, while the Intel Core Ultra X9 378H does not. Both support dual-channel memory, though the AMD chip supports DDR5 and LPDDR5X while the Intel chip supports only LPDDR5X.

Q: What process nodes do these processors use?

A: The AMD Ryzen AI Embedded P164 uses a 4 nm process from TSMC, while the Intel Core Ultra X9 378H uses a 3 nm process from Intel. The Intel chip also has a smaller physical footprint in terms of unspecified die area, as the database lists no die size for the Intel part versus 233 mm² for the AMD part.

Q: Which processor wins in single-thread performance?

A: The Intel Core Ultra X9 378H wins the PassMark single-thread test with a score of 4453, versus 4029 for the AMD Ryzen AI Embedded P164. That represents a 9.5% advantage for the Intel chip.

Q: What are the nearest rivals for each processor?

A: For the AMD Ryzen AI Embedded P164, the nearest rivals are the AMD Ryzen 5 9500F (average score 52873, delta 0.1%), Intel Xeon 634 (52974, delta -0.1%), AMD EPYC 7313P (53206, delta -0.6%), and AMD Ryzen 9 7900X (53288, delta -0.7%). For the Intel Core Ultra X9 378H, the nearest rivals are the AMD Ryzen 9 PRO 5945 (47527, delta -0.1%), Intel Core i7-13700KF (47330, delta 0.3%), Intel Core Ultra 7 265T (47697, delta -0.5%), and Intel Core i9-12900F (47176, delta 0.6%).

Architecture Differences

The AMD Ryzen AI Embedded P164 uses the Gorgon Point codename and belongs to the Ryzen AI Embedded generation built on Zen 5 and Zen 5c cores. The Intel Core Ultra X9 378H uses the Panther Lake codename and belongs to the Ultra X9 generation built on Panther Lake-H. These are fundamentally different designs: the AMD chip is a hybrid Zen 5 and Zen 5c arrangement, while the Intel chip is a Panther Lake-H part.

The process nodes differ significantly. AMD uses a 4 nm process at TSMC, while Intel uses a 3 nm process at its own foundry. The AMD die measures 233 mm², while the database records no die size for the Intel part. The AMD chip has 8 MB of L3 cache, while the Intel chip has 18 MB of shared L3 cache. Per-core cache also differs: the AMD part has 80 KB of L1 per core and 1 MB of L2 per core, while the Intel part has 192 KB of L1 per core and 2.5 MB of L2 per core.

Memory support separates the two as well. The AMD chip supports both DDR5 and LPDDR5X with dual-channel memory and a memory bandwidth of 89.6 GB/s. The Intel chip supports only LPDDR5X, also dual-channel, but with a much higher memory bandwidth of 153.6 GB/s. ECC memory is supported on the AMD side but not on the Intel side.

PCIe connectivity also differs. The AMD Ryzen AI Embedded P164 provides Gen 4 with 16 lanes (CPU only), while the Intel Core Ultra X9 378H provides Gen 5 with 4 lanes (CPU only). Integrated graphics differ too: the AMD chip uses a Radeon 880M, while the Intel chip uses an Arc B390.

Both processors share identical base clocks of 2.00 GHz and boost clocks of 5.00 GHz. Both are locked, with multiplier unlock set to false. Both are Active in production status. The AMD part uses AMD Socket FP8, while the Intel part uses Intel BGA 2540. The AMD chip has a TDP of 28 watts, and the Intel chip has a TDP of 25 watts.

Release dates place the AMD part at 2026-03-08 and the Intel part at 2026-04-03, roughly one month apart. Neither has a recorded launch MSRP in the database.

Where Each One Wins

The AMD Ryzen AI Embedded P164 wins in the overall average benchmark score, posting 52901 against 47468 for the Intel Core Ultra X9 378H. That puts the AMD chip in the 91st percentile of all CPUs, two points above the Intel chip's 89th percentile. The AMD part also sits closer to a higher-performing rival group, with its nearest rivals including the AMD Ryzen 9 7900X at an average score of 53288, while the Intel chip's nearest rivals top out near the Intel Core i7-13700KF at 47330.

However, the Intel Core Ultra X9 378H wins every single head-to-head benchmark in the database. The Intel chip dominates in workloads that scale with core count and parallel throughput. In PassMark multithread, the Intel chip scores 38298 versus 25889 for the AMD chip, a 32.4% advantage. In PassMark physics, the Intel chip scores 3404 versus 1210, a 64.5% lead. In floating point math, the Intel chip scores 114500 versus 55799, a 51.3% lead.

The Intel chip also wins in tasks involving encryption and compression. PassMark data encryption shows the Intel chip at 29840 versus 16055, a 46.2% advantage. PassMark data compression shows the Intel chip at 386591 versus 327891, a 15.2% lead. In random string sorting, the Intel chip scores 44648 versus 34801, a 22.1% advantage.

For single-threaded and light-threaded workloads, the Intel chip also holds the edge. PassMark single-thread shows the Intel chip at 4453 versus 4029, a 9.5% lead. PassMark extended instructions shows the Intel chip at 31315 versus 24193, a 22.7% advantage. In integer math, the Intel chip leads by a narrower margin: 92603 versus 87940, or 5%. The largest Intel win is in find prime numbers, where the Intel chip scores 357 versus 71, an 80.1% margin.

The AMD part's advantage is thus concentrated in the aggregate benchmark average, not in any individual recorded workload. The database records zero wins for the AMD chip across the head-to-head tests, while the Intel chip records wins in all 11.

Specification Differences

| Specification | AMD Ryzen AI Embedded P164 | Intel Core Ultra X9 378H |

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

| Cores | 8 | 16 |

| Threads | 16 | 16 |

| TDP | 28 watts | 25 watts |

| Socket | AMD Socket FP8 | Intel BGA 2540 |

| Codename | Gorgon Point | Panther Lake |

| Generation | Ryzen AI Embedded (Zen 5 / Zen 5c) | Ultra X9 (Panther Lake-H) |

| Process node | 4 nm | 3 nm |

| Foundry | TSMC | Intel |

| Die size | 233 mm² | Not recorded |

| L1 cache | 80 KB per core | 192 KB per core |

| L2 cache | 1 MB per core | 2.5 MB per core |

| L3 cache | 8 MB | 18 MB shared |

| Memory support | DDR5, LPDDR5X | LPDDR5X |

| Memory bandwidth | 89.6 GB/s | 153.6 GB/s |

| ECC memory | Supported | Not supported |

| PCIe | Gen 4, 16 lanes (CPU only) | Gen 5, 4 lanes (CPU only) |

| Integrated graphics | Radeon 880M | Arc B390 |

| Release date | 2026-03-08 | 2026-04-03 |

| Percentile vs all CPUs | 91 | 89 |

| Average benchmark score | 52901 | 47468 |

Head-to-Head Benchmarks

The head-to-head data is one-sided. Across all 11 recorded PassMark tests, the Intel Core Ultra X9 378H wins every contest, but the size of the margin varies dramatically by workload.

The largest Intel advantage appears in the find prime numbers test, where the Intel chip scores 357 against 71 for the AMD chip, an 80.1% gap. This is an extreme outlier relative to the rest of the results, suggesting the Intel architecture handles that specific workload far more efficiently than the AMD implementation.

The physics test shows the second-largest gap. The Intel chip scores 3404 versus 1210, a 64.5% lead. Floating point math follows with the Intel chip at 114500 versus 55799, a 51.3% margin. These three workloads, prime finding, physics, and floating point math, are all heavily parallel and benefit from the Intel chip's 16 cores versus the AMD chip's 8 cores.

Encryption shows a substantial Intel advantage as well. The Intel chip scores 29840 versus 16055, a 46.2% lead. This suggests the Intel part has a strong advantage in cryptographic throughput, likely related to its core count and architecture design.

Multithread performance shows the Intel chip at 38298 versus 25889, a 32.4% lead. Extended instructions show the Intel chip at 31315 versus 24193, a 22.7% advantage. Random string sorting shows the Intel chip at 44648 versus 34801, a 22.1% lead.

The narrowest Intel wins come in integer math and single-thread performance. In integer math, the Intel chip scores 92603 versus 87940, only a 5% edge. In single-thread, the Intel chip scores 4453 versus 4029, a 9.5% lead. These smaller margins indicate that the AMD chip's Zen 5 and Zen 5c cores are competitive per-thread, but the Intel chip's higher core count and larger cache configuration push it ahead in aggregate throughput.

Data compression is the remaining test, with the Intel chip at 386591 versus 327891 for the AMD chip, a 15.2% advantage.

The aggregate picture is notable: the AMD chip holds a higher average benchmark score (52901 versus 47468) and a higher percentile rank (91 versus 89), yet it loses every individual head-to-head test. This apparent contradiction stems from the composition of the average score, which includes a broader set of benchmarks beyond the PassMark suite, and from the different nearest-rival pools that anchor each chip's percentile position. The AMD part's nearest rivals include the AMD Ryzen 9 7900X at an average score of 53288, while the Intel part's nearest rivals include the Intel Core i9-12900F at 47176, which places the Intel chip in a lower-scoring reference group despite its per-test wins against the AMD part.