AMD Ryzen AI Embedded P164 vs Intel Core Ultra X7 358H Comparison

Head-to-Head Benchmarks

The benchmark data for these two mobile processors shows a decisive overall advantage for the Intel Core Ultra X7 358H, which claims 10 of 11 head-to-head wins. The AMD Ryzen AI Embedded P164 manages a single victory, but the margins in Intel's favor are often substantial. The most extreme gap appears in the passmark_find_prime_numbers test, where Intel scores 337 against AMD's 71, a 78.9% difference. This indicates a major advantage in integer-heavy, single-threaded calculation workloads for the Intel part.

The floating point math test reinforces Intel's dominance in compute-heavy tasks. The Core Ultra X7 358H posts a score of 103842, while the Ryzen AI Embedded P164 trails at 55799, a 46.3% deficit. This is one of the largest percentage deltas in the entire comparison, suggesting that the Intel processor delivers nearly double the floating-point throughput in this specific benchmark. Similarly, the physics test shows Intel at 3021 versus AMD's 1210, a 59.9% margin, which points to a strong performance advantage in simulation and physics-based workloads.

Data encryption is another clear Intel win. The Core Ultra X7 358H scores 26046, compared to the AMD's 16055, a 38.4% gap. This suggests the Intel processor has more capable cryptographic instruction handling or a more efficient implementation of encryption algorithms. The multithread benchmark also favors Intel, with a score of 33802 against AMD's 25889, a 23.4% difference. Given that both processors present the same thread count of 16, this result is particularly telling: Intel extracts more multithreaded performance from an equal number of threads.

The Intel part also wins in extended instructions with a score of 27274 versus AMD's 24193, an 11.3% margin, and in random string sorting at 40357 versus 34801, a 13.8% gap. These results indicate a consistent edge in memory manipulation and SIMD-style workloads. In data compression, Intel leads narrowly with 332508 versus 327891, a modest 1.4% difference. The single-thread benchmark also goes to Intel, but the margin is small: 4124 versus 4029, a 2.3% gap. This suggests that while Intel has the higher single-core score, the two chips are closely matched in basic single-threaded execution.

The AMD Ryzen AI Embedded P164's only win comes in integer math, where it scores 87940 against Intel's 83147, a 5.8% advantage. This means that in pure integer arithmetic operations, the AMD processor outperforms the Intel part despite losing every other benchmark in the head-to-head set. It is a narrow but meaningful win, and it shows that AMD's Zen 5 / Zen 5c architecture holds an edge in at least one fundamental computational domain.

The overall average benchmark scores place the AMD part higher in the database, with an average of 52901 versus Intel's 40967. The AMD processor also holds a higher percentile ranking, 91 versus 87. This apparent contradiction with the head-to-head results is explained by the fact that the two processors have different benchmark suites recorded. The AMD chip's PassMark tests are scored against a broader CPU population, and its nearest rivals include the AMD Ryzen 5 9500F (average score 52873, a 0.1% delta), the Intel Xeon 634 (52974, a 0.1% delta), the AMD EPYC 7313P (53206, a 0.6% delta), and the AMD Ryzen 9 7900X (53288, a 0.7% delta). The Intel part's nearest rivals are the AMD Ryzen AI 5 PRO 440 (41208, a 0.6% delta), the Intel Core Ultra 7 356H (41215, a 0.6% delta), the AMD Ryzen AI 5 PRO 435G (40718, a 0.6% delta), and the Intel Core Ultra 7 366H (41263, a 0.7% delta). These two rival groups occupy very different performance tiers, which explains why the AMD chip ranks higher overall despite losing most direct comparisons.

Architecture Differences

The two processors are built on fundamentally different manufacturing and design approaches. The AMD Ryzen AI Embedded P164 uses a 4 nm process node from TSMC, while the Intel Core Ultra X7 358H uses a 3 nm node from Intel's own foundry. The AMD chip's codename is Gorgon Point, and it belongs to the Ryzen AI Embedded generation based on Zen 5 and Zen 5c cores. The Intel part, codenamed Panther Lake, is part of the Ultra X7 line within the Panther Lake-H generation.

Core and thread counts differ significantly. The AMD processor has 8 cores and 16 threads, while the Intel processor has 16 cores and 16 threads. This means the Intel part has twice as many physical cores but no hyperthreading, whereas the AMD part relies on simultaneous multithreading to reach its 16-thread count. The Intel chip's higher core count likely contributes to its strong multithread and physics benchmark results.

Cache hierarchies also differ. The AMD processor has 80 KB of L1 cache per core, 1 MB of L2 cache per core, and 8 MB of L3 cache. The Intel processor has 192 KB of L1 per core, 3 MB of L2 per core, and a much larger 18 MB of shared L3 cache. The larger per-core L1 and L2 allocations on the Intel part, combined with the larger shared L3, provide a substantial caching advantage that likely explains its lead in data compression and random string sorting, both of which are memory-latency sensitive.

Memory support differs as well. The AMD chip supports both DDR5 and LPDDR5X memory across a dual-channel bus, with a memory bandwidth of 89.6 GB/s. The Intel chip supports only LPDDR5X, also dual-channel, but with a significantly higher memory bandwidth of 153.6 GB/s. That 63.9 GB/s gap in theoretical bandwidth is substantial and likely contributes to Intel's wins in bandwidth-intensive benchmarks. The AMD part supports ECC memory, while the Intel part does not, which makes the AMD chip more suitable for error-sensitive embedded workloads.

PCIe connectivity also differs. The AMD processor offers Gen 4 with 16 lanes (CPU only), while the Intel processor offers Gen 5 with 4 lanes (CPU only). This is a trade-off: AMD provides more lanes at a lower generation, while Intel provides fewer lanes but at a higher generation. For systems requiring multiple expansion devices, AMD's 16 lanes may be preferable; for maximum per-device bandwidth, Intel's Gen 5 support has the edge.

The integrated graphics solutions are different as well. The AMD chip uses the Radeon 880M, while the Intel chip uses the Arc B390. The database does not record benchmark scores for these iGPUs, so a direct performance comparison is not available from the recorded data. Both target the mobile segment and are currently Active in production status.

The AMD processor has a base clock of 2.00 GHz and a boost clock of 5.00 GHz, while the Intel processor has a base clock of 1.90 GHz and a boost clock of 4.80 GHz. The AMD part has higher clocks in both states, yet it still loses most benchmark comparisons. This suggests that the Intel architecture achieves higher instructions-per-clock in most workloads, or that its larger cache and higher memory bandwidth compensate for the lower clock speeds.

Thermal design power differs slightly: the AMD part is rated at 28 W, the Intel part at 25 W. Both are mobile-class processors with relatively modest power envelopes. Neither processor has an unlocked multiplier, so overclocking is not supported on either part. The AMD chip has a die size of 233 mm², while the Intel chip's die size is not recorded in the database.

The AMD processor was released on 2026-03-08, while the Intel processor was released on 2026-01-04. The Intel part came to market earlier by roughly two months. Neither processor has a recorded launch MSRP in the database.

FAQ

Q: Which processor has more cores?

A: The Intel Core Ultra X7 358H has 16 cores, while the AMD Ryzen AI Embedded P164 has 8 cores. Both have 16 threads.

Q: Which processor has the higher boost clock?

A: The AMD Ryzen AI Embedded P164 boosts to 5.00 GHz, while the Intel Core Ultra X7 358H boosts to 4.80 GHz.

Q: How large is the L3 cache on each processor?

A: The AMD processor has 8 MB of L3 cache, while the Intel processor has 18 MB of shared L3 cache.

Q: Which processor supports ECC memory?

A: The AMD Ryzen AI Embedded P164 supports ECC memory. The Intel Core Ultra X7 358H does not.

Q: What is the memory bandwidth of each processor?

A: The AMD processor has a memory bandwidth of 89.6 GB/s, while the Intel processor has a memory bandwidth of 153.6 GB/s.

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen AI Embedded P164 has an average benchmark score of 52901, compared to 40967 for the Intel Core Ultra X7 358H. The AMD chip also ranks in the 91st percentile of all CPUs, while the Intel chip ranks in the 87th.

Specification Differences

The two processors differ in the following recorded specifications:

| Field | AMD Ryzen AI Embedded P164 | Intel Core Ultra X7 358H |

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

| Cores | 8 | 16 |

| Threads | 16 | 16 |

| Base clock | 2.00 GHz | 1.90 GHz |

| Boost clock | 5.00 GHz | 4.80 GHz |

| TDP | 28 W | 25 W |

| Socket | AMD Socket FP8 | Intel BGA 2540 |

| Codename | Gorgon Point | Panther Lake |

| Generation | Ryzen AI Embedded (Zen 5 / Zen 5c) | Ultra X7 (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) | 3 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 | Yes | No |

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

| Integrated graphics | Radeon 880M | Arc B390 |

| Release date | 2026-03-08 | 2026-01-04 |

| Part number | Unknown | SA4RAQ9ET |

Where Each One Wins

The Intel Core Ultra X7 358H is the stronger choice for most compute-heavy workloads based on the recorded benchmarks. Its wins in floating point math, physics, encryption, and multithread performance make it well-suited for applications that stress the CPU with scientific calculations, physics simulations, and data encryption. The 46.3% lead in floating point math and the 59.9% lead in physics are the standout margins, indicating a substantial advantage in these areas. The Intel part also wins in data compression, extended instructions, random string sorting, and single-thread performance, making it the more versatile processor for general-purpose computing.

The Intel chip's larger L3 cache, higher memory bandwidth, and 16 physical cores likely explain its broad benchmark dominance. For tasks that are memory-latency sensitive or that scale with core count, the Intel processor has clear advantages. Its 153.6 GB/s memory bandwidth is especially notable, as it more than doubles the AMD part's theoretical ceiling, and its 18 MB shared L3 cache is more than twice the size of AMD's 8 MB. These architectural resources appear to translate directly into benchmark wins.

The AMD Ryzen AI Embedded P164 is the better option in a narrower set of circumstances. Its only head-to-head win is in integer math, where it leads by 5.8%, so for integer-heavy workloads the AMD processor holds a measurable advantage. The AMD chip also offers ECC memory support, which the Intel part lacks, making it the more appropriate choice for embedded applications where data integrity is critical. Its PCIe Gen 4 implementation provides 16 lanes, compared to Intel's 4 lanes of PCIe Gen 5, so systems that need to connect multiple PCIe devices may prefer the AMD part for its lane count alone.

The higher average benchmark score and 91st percentile ranking of the AMD processor also suggest that its overall standing in the database is strong, even though it loses the direct comparison to the Intel part. The two processors occupy different performance neighborhoods relative to their nearest rivals, and the AMD chip's closest competitors are significantly faster than the Intel chip's closest competitors. This indicates that the AMD part is positioned in a higher tier of overall CPU performance, even if the Intel part wins the specific benchmarks they share.

For single-threaded workloads, the Intel chip leads by a narrow 2.3%, so neither processor has a decisive edge in this category. The Intel part has a slightly lower base clock and boost clock, yet still comes out ahead in most tests, which suggests its architectural efficiency compensates for the clock deficit. The AMD part's higher clocks help it win integer math, but not enough to overcome Intel's advantages elsewhere. In summary, the Intel Core Ultra X7 358H is the processor of choice for users who prioritize raw benchmark performance across a broad range of tasks, while the AMD Ryzen AI Embedded P164 retains specific advantages in integer math, ECC support, and PCIe lane count for embedded and reliability-focused applications.