AMD Ryzen 9 9955HX3D vs AMD Ryzen AI Embedded P164 Comparison

Head-to-Head Benchmarks

The benchmark data presents a decisive picture: the AMD Ryzen 9 9955HX3D wins all 11 recorded head-to-head tests against the AMD Ryzen AI Embedded P164. The scale of the victory varies dramatically by workload, but the direction is consistent across every metric.

The largest gap appears in prime number computation. The 9955HX3D scores 525 in PassMark's find prime numbers test, while the P164 manages only 71. That is a 639.4% advantage, the single biggest delta in the comparison. This type of workload is highly sensitive to memory latency and cache capacity, which is a strong indicator of how the 3D V-Cache configuration changes behavior.

Physics simulation also shows a massive split. The 9955HX3D records 4687 points versus 1210 for the P164, a 287.4% lead. Physics engines rely heavily on multi-threaded scaling and consistent memory bandwidth, both areas where the 16-core part has structural advantages.

Extended instruction performance favors the 9955HX3D by 159.6%, with scores of 62813 against 24193. Floating point math follows closely at 158.3% (144122 versus 55799), and integer math at 153% (222503 versus 87940). These are raw compute tests where core count and cache hierarchy matter more than clock speed alone.

Data compression and encryption show similar patterns. The 9955HX3D delivers 785811 in compression versus 327891, a 139.7% edge. Encryption shows 39446 versus 16055, a 145.7% advantage. Both workloads benefit from larger L3 cache and more parallel execution units.

The multi-thread test result is 62674 for the 9955HX3D against 25889 for the P164, a 142.1% gap. Random string sorting shows 83497 versus 34801, a 139.9% difference. These are consistent with the overall core and thread count disparity.

The narrowest margin is in single-thread performance. The 9955HX3D scores 4511, and the P164 scores 4029, a 12% lead. This is the only test where the P164 comes close, suggesting that per-core efficiency is relatively similar, but the 9955HX3D still holds a clear edge due to its higher boost clock of 5.40 GHz versus 5.00 GHz.

Where Each One Wins

Based on the recorded benchmarks, the 9955HX3D wins in every category. There is no test in the database where the P164 comes out ahead. However, the degree of dominance varies, which reveals different use-case profiles.

For single-threaded applications, such as older games or lightly threaded productivity tools, the 9955HX3D's 12% advantage is meaningful but not overwhelming. The P164 is closer here, indicating that its Zen 5 / Zen 5c cores are competitive on a per-core basis, but the higher boost clock of the 9955HX3D tips the balance.

For heavily parallel workloads, the gap widens dramatically. Multi-thread, physics, integer math, and floating point all show advantages between 142% and 287%. The 9955HX3D's 16 cores and 32 threads versus 8 cores and 16 threads create a structural difference that no clock speed adjustment can overcome.

Prime number finding is the standout case. The 639.4% advantage suggests that the 128 MB L3 cache on the 9955HX3D is doing substantial work. This workload typically involves large working sets that fit in cache, and the P164's 8 MB L3 is a fraction of that capacity.

The P164's only relative strength is power efficiency. With a 28 W TDP versus 55 W, it uses less than half the power envelope. For embedded or mobile scenarios where thermal limits are strict, that efficiency is a real advantage, even if the benchmark scores are lower.

Architecture Differences

The two processors share a common foundation: both are built on TSMC's 4 nm process, use Zen 5 cores, and support dual-channel DDR5 memory with 89.6 GB/s bandwidth. Both also offer ECC memory support and integrate Radeon graphics, though the specific iGPU differs.

The core configuration is the first major divergence. The 9955HX3D has 16 cores and 32 threads, while the P164 has 8 cores and 16 threads. The 9955HX3D also uses a homogeneous core design, whereas the P164's generation label mentions "Zen 5 / Zen 5c," implying a mix of full and compact cores. This hybrid approach explains why the P164's per-core performance is respectable despite lower raw scores.

Cache hierarchy is the second major difference. The 9955HX3D carries 128 MB of L3 cache, a figure that reflects its 3D V-Cache technology. The P164 has only 8 MB of L3. Both share 80 KB of L1 and 1 MB of L2 per core, but the L3 disparity is enormous. This directly explains the prime number result and likely contributes to the compression and encryption advantages.

Clock speeds differ as well. The 9955HX3D runs at 2.50 GHz base and 5.40 GHz boost. The P164 runs at 2.00 GHz base and 5.00 GHz boost. The 9955HX3D is also multiplier unlocked, while the P164 is locked, giving the former overclocking headroom.

Package and socket differences are notable. The 9955HX3D uses AMD Socket FL1, while the P164 uses AMD Socket FP8. The 9955HX3D uses a chiplet design with a die size of 2x 70.6 mm² and 16,630 million transistors. The P164 is monolithic at 233 mm², with transistor count not recorded. The 9955HX3D supports PCIe Gen 5 with 28 lanes, while the P164 supports PCIe Gen 4 with 16 lanes.

Memory support differs slightly: the 9955HX3D lists DDR5 only, while the P164 lists DDR5 and LPDDR5X. Both use dual-channel memory buses. The integrated graphics also differ: the 9955HX3D uses Radeon 610M, while the P164 uses Radeon 880M, which is likely a more capable iGPU given the embedded focus.

The P164 has a later release date of March 2026, while the 9955HX3D launched in January 2025. Both are listed as active in production.

FAQ

Q: Which processor has the higher single-thread performance?

A: The AMD Ryzen 9 9955HX3D scores 4511 in PassMark single-thread tests, which is 12% higher than the AMD Ryzen AI Embedded P164's 4029. The delta comes from the 5.40 GHz boost clock versus 5.00 GHz.

Q: How large is the multi-thread performance gap?

A: The 9955HX3D records 62674 in PassMark multi-thread, while the P164 records 25889, a 142.1% advantage. This aligns with the core count difference of 16 versus 8.

Q: Why is the prime number test result so lopsided?

A: The 9955HX3D scores 525 versus 71 for the P164, a 639.4% advantage. The primary factor is likely the L3 cache: 128 MB on the 9955HX3D versus 8 MB on the P164, which matters greatly for this workload.

Q: Do both processors support ECC memory?

A: Yes, both the AMD Ryzen 9 9955HX3D and the AMD Ryzen AI Embedded P164 support ECC memory and use dual-channel DDR5 with 89.6 GB/s bandwidth.

Q: What are the TDP differences?

A: The 9955HX3D has a TDP of 55 W, while the P164 has a TDP of 28 W. The P164 consumes less power, which is relevant for embedded thermal constraints.

Q: Which processor has a higher percentile ranking?

A: The 9955HX3D ranks in the 96th percentile among all CPUs, while the P164 ranks in the 91st percentile. The average benchmark score is 97453 for the 9955HX3D and 52901 for the P164.

Specification Differences

| Field | AMD Ryzen 9 9955HX3D | AMD Ryzen AI Embedded P164 |

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

| Cores | 16 | 8 |

| Threads | 32 | 16 |

| Base clock | 2.50 GHz | 2.00 GHz |

| Boost clock | 5.40 GHz | 5.00 GHz |

| TDP | 55 W | 28 W |

| Socket | AMD Socket FL1 | AMD Socket FP8 |

| Codename | Fire Range | Gorgon Point |

| Generation | Ryzen 9 (Zen 5 (Fire Range)) | Ryzen AI Embedded (Zen 5 / Zen 5c) |

| Die size | 2x 70.6 mm² | 233 mm² |

| Transistors | 16,630 million | Not recorded |

| L3 cache | 128 MB | 8 MB |

| Memory support | DDR5 | DDR5, LPDDR5X |

| PCIe | Gen 5, 28 lanes | Gen 4, 16 lanes |

| Integrated graphics | Radeon 610M | Radeon 880M |

| Multiplier unlocked | Yes | No |

| Part number | 100-000001030 | Unknown |

| Release date | January 2025 | March 2026 |

The Verdict

The data supports a clear split based on workload and power budget. The AMD Ryzen 9 9955HX3D is the superior performer in every recorded benchmark, with advantages ranging from 12% in single-thread to 639.4% in prime number finding. Its 16 cores, 32 threads, and 128 MB L3 cache deliver a comprehensive multi-threaded advantage, and its 96th percentile ranking confirms its position among high-end CPUs.

The AMD Ryzen AI Embedded P164 is not without merit. Its 28 W TDP makes it suitable for power-constrained embedded designs, and its 91st percentile ranking is respectable. The Radeon 880M iGPU and LPDDR5X support suggest a focus on integrated graphics and memory flexibility. For single-threaded tasks, its 4029 score is only 12% behind, indicating competitive per-core efficiency.

The choice depends on the use case. For high-performance mobile computing, content creation, or any workload that scales with cores and cache, the 9955HX3D is the clear pick. Its 142.1% multi-thread lead and 158.3% floating point advantage leave no ambiguity. For embedded systems where power consumption is the primary constraint and raw compute is secondary, the P164's lower TDP and compact monolithic design offer a more suitable fit. The benchmark results confirm that these are different classes of processors, and the data does not suggest any scenario where the P164 outperforms the 9955HX3D in compute tasks.