AMD Ryzen AI Embedded P164 vs Intel Core 5 221TE Comparison

The AMD Ryzen AI Embedded P164 and Intel Core 5 221TE are both active processors with 16 threads, but the recorded benchmark data separates them sharply. The AMD part wins all 11 PassMark head-to-head tests, averages 52901, and sits in the 91st percentile of all CPUs. The Intel part averages 17860 and sits in the 71st percentile, with 0 wins in the head-to-head set.

Head-to-Head Benchmarks

The head-to-head set contains 11 PassMark workloads, and the AMD Ryzen AI Embedded P164 records 11 wins. The Intel Core 5 221TE records 0 wins, so there is no workload in the data where the Intel part comes out ahead.

The largest margin is in extended instructions: AMD scores 24193 against Intel's 9655, a delta of 150.6%. The second-largest margin is single-thread performance, where AMD scores 4029 and Intel scores 1734, a delta of 132.4%. Data compression is also one-sided: AMD scores 327891, Intel scores 156682, a 109.3% lead. Integer math shows AMD at 87940 versus 42303, a 107.9% lead. Random string sorting is 34801 to 16929, a 105.6% lead. Multithread performance is 25889 to 13301, a 94.6% lead. Encryption is 16055 to 8963, a 79.1% lead. Floating point math is 55799 to 31661, a 76.2% lead.

The narrower margins still favor AMD. Physics is 1210 to 977, a 23.8% lead, and find prime numbers is 71 to 59, a 20.3% lead. The Intel part's best relative result is in prime number search, where the gap shrinks to 20.3%, but it remains a clear AMD win.

The average-score context reinforces the gap. AMD's nearest rival in the database is the AMD Ryzen 5 9500F, which averages 52873; AMD leads that part by 0.1%. The Intel Xeon 634 averages 52974, and AMD trails by 0.1%. The AMD EPYC 7313P averages 53206, a 0.6% deficit for AMD, and the AMD Ryzen 9 7900X averages 53288, a 0.7% deficit. For the Intel part, the AMD Ryzen 5 3600XT averages 17891, and Intel trails by 0.2%. The Intel Core 5 120U averages 17898, also a 0.2% deficit. The Intel Core 7 350 averages 17779, and Intel leads that part by 0.5%. The AMD Ryzen 5 1600 averages 17994, a 0.7% deficit for Intel.

FAQ

Q: Which processor wins the most head-to-head benchmarks?

A: The AMD Ryzen AI Embedded P164 wins all 11 PassMark tests in the head-to-head set. The Intel Core 5 221TE records 0 wins.

Q: What is the overall average benchmark score for each part?

A: The AMD part averages 52901 and ranks in the 91st percentile of all CPUs. The Intel part averages 17860 and ranks in the 71st percentile.

Q: Which test shows the largest gap between the two?

A: Extended instructions shows the largest gap: AMD scores 24193, Intel scores 9655, a delta of 150.6%. Single-thread performance is next at 132.4%, with AMD at 4029 and Intel at 1734.

Q: Which test shows the smallest gap?

A: Find prime numbers shows the smallest gap: AMD scores 71 and Intel scores 59, a 20.3% lead. Physics is the second closest at 23.8%, with AMD at 1210 and Intel at 977.

Q: How does each part compare to its nearest rivals?

A: The AMD part leads the AMD Ryzen 5 9500F by 0.1% at 52873 and trails the Intel Xeon 634 by 0.1% at 52974, the AMD EPYC 7313P by 0.6% at 53206, and the AMD Ryzen 9 7900X by 0.7% at 53288. The Intel part trails the AMD Ryzen 5 3600XT by 0.2% at 17891, the Intel Core 5 120U by 0.2% at 17898, and the AMD Ryzen 5 1600 by 0.7% at 17994, while leading the Intel Core 7 350 by 0.5% at 17779.

Q: Do the two parts have the same core and thread counts?

A: No. AMD has 8 cores and 16 threads; Intel has 10 cores and 16 threads. The Intel part has two more cores but the same thread count.

Where Each One Wins

Because the Intel Core 5 221TE has no wins in the head-to-head data, the use-case split is not balanced. The AMD Ryzen AI Embedded P164 is the faster part in every recorded workload. The largest advantages are in extended instruction throughput, single-thread tasks, data compression, integer math, and random string sorting; all of those deltas exceed 100%. For tasks that stress encryption and floating point math, AMD still leads by 79.1% and 76.2%, respectively. The only workloads where Intel comes within 25% are prime number search and physics, but it still loses those by 20.3% and 23.8%. The recorded data shows no workload in which the Intel part takes the lead.

Specification Differences

| Field | AMD Ryzen AI Embedded P164 | Intel Core 5 221TE |

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

| Manufacturer | AMD | Intel |

| Cores | 8 | 10 |

| Base clock | 2.00 | 1.80 |

| TDP | 28 | 45 |

| Socket | AMD Socket FP8 | Intel Socket 1700 |

| Codename | Gorgon Point | Bartlett Lake |

| Generation | Ryzen AI Embedded (Zen 5 / Zen 5c) | Core 5 (Bartlett Lake) |

| Process node | 4 nm | 10 nm |

| Foundry | TSMC | Intel |

| Die size | 233 mm² | 215 mm² |

| L2 cache | 1 MB (per core) | 1.25 MB (per core) |

| L3 cache | 8 MB | 24 MB (shared) |

| Memory support | DDR5, LPDDR5X | DDR4, DDR5 |

| Memory bandwidth | 89.6 GB/s | 76.8 GB/s |

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

| Integrated graphics | Radeon 880M | UHD Graphics 730 |

| Market segment | Mobile | Desktop |

| Release date | 2026-03-08 | 2025-01-12 |

| Launch MSRP | not recorded | $232 |

| Part number | unknown | SRVQS |

Architecture Differences

The two parts come from different design families. AMD's Gorgon Point is listed under the Ryzen AI Embedded generation with Zen 5 / Zen 5c cores, fabricated by TSMC on a 4 nm process. Intel's Bartlett Lake belongs to the Core 5 generation and is fabricated by Intel on a 10 nm process. The process node difference is substantial: 4 nm versus 10 nm.

The core topology differs as well. AMD uses 8 cores and 16 threads, while Intel uses 10 cores and 16 threads. Despite having two fewer cores, the AMD part produces higher multithread performance in the recorded data, 25889 versus 13301, a 94.6% lead. The AMD part also has a higher base clock, 2.00 versus 1.80, while both parts list a 5.00 boost clock. The benchmark results indicate that the AMD part's per-core advantage is enough to overcome the Intel part's additional cores.

The cache hierarchy is organized differently. AMD assigns 1 MB L2 per core and an 8 MB L3. Intel assigns 1.25 MB L2 per core and a 24 MB shared L3. The larger Intel L3 does not translate into a win in any recorded workload. Memory architecture also differs: AMD supports DDR5 and LPDDR5X with 89.6 GB/s bandwidth, while Intel supports DDR4 and DDR5 with 76.8 GB/s bandwidth. Both support ECC memory.

Platform features separate the two as well. AMD uses PCIe Gen 4 with 16 CPU lanes, while Intel uses PCIe Gen 5 with 16 CPU lanes. The integrated graphics are Radeon 880M for AMD and UHD Graphics 730 for Intel. AMD's part is a mobile design on AMD Socket FP8, while Intel's part is a desktop design on Intel Socket 1700. Both parts are active production parts with locked multipliers.