AMD Ryzen AI Embedded P164 vs Intel Core 5 220H Comparison

Head-to-Head Benchmarks

The recorded benchmark data shows a decisive overall performance advantage for the AMD Ryzen AI Embedded P164, which wins 9 of the 11 shared tests. The Intel Core 5 220H manages two wins, though both are narrow and isolated to specific workload types.

The largest single margin in the entire comparison appears in extended instructions, where the AMD part scores 24193 against Intel's 14642. That is a 65.2% advantage, the most lopsided result in the dataset. The AMD chip also leads substantially in data compression, scoring 327891 versus 247921, a 32.3% edge. These two tests alone establish the AMD processor as the stronger choice for workloads that rely on modern instruction set extensions and compression or decompression tasks.

Integer math shows a 19.6% gap in AMD's favor, with scores of 87940 against 73555. Floating point math is closer but still favors AMD, 55799 versus 51671, an 8% difference. Random string sorting goes to AMD by 22.4%, 34801 versus 28438. Multi-threaded performance, measured by the PassMark multi-thread test, gives AMD a solid 18.3% lead, 25889 versus 21884. The single-thread score follows the same pattern: 4029 for AMD versus 3405 for Intel, again an 18.3% gap, which is notable given that the Intel chip has a higher base clock.

Data encryption is the closest race in the whole set, with AMD scoring 16055 and Intel 15216, a margin of only 5.5%. Both chips handle encryption well, but the AMD part still comes out ahead.

The Intel Core 5 220H claims its first win in the prime number search test, scoring 82 versus AMD's 71, which translates to a 13.4% advantage. The other Intel win is in the physics test, where it scores 1478 against AMD's 1210, an 18.1% lead. These two results are the only measurable areas where Intel outperforms AMD, and both are relatively niche workloads rather than general-purpose indicators.

Average benchmark scores confirm the overall picture. The AMD Ryzen AI Embedded P164 posts an average score of 52901, sitting at the 91st percentile of all CPUs in the database. The Intel Core 5 220H averages 28574, which places it at the 80th percentile. The average score gap is substantial: AMD's average is roughly 85% higher than Intel's. The nearest rival lists reflect this positioning. AMD's closest competitors, the AMD Ryzen 5 9500F at 52873 (0.1% behind) and the Intel Xeon 634 at 52974 (0.1% ahead), are both in a completely different performance class from the Intel Core 5 220H. The Intel part's nearest rivals, such as the AMD EPYC 7203P at 28583 and the AMD Ryzen 7 PRO 6850HS at 28549, all sit within 0.3% of its average score.

Where Each One Wins

The AMD Ryzen AI Embedded P164 dominates the majority of compute-oriented workloads. Data compression, integer math, floating point math, encryption, extended instructions, random string sorting, multi-threaded throughput, and single-threaded performance all favor AMD. The 65.2% lead in extended instructions indicates that software written to use the latest SIMD and vector extensions will run markedly better on the AMD chip. The 32.3% compression lead suggests advantages for archive handling, database workloads, and any task that moves large volumes of in-memory data around. The 18.3% single-thread advantage means even lightly threaded applications, such as many office productivity tools and legacy software, will feel snappier on the AMD part.

The Intel Core 5 220H has a narrower but still real set of strengths. Its 13.4% win in prime number search points to an advantage in workloads dominated by simple integer iteration over small data sets, which is a common pattern in some cryptographic validation routines and mathematical sieve algorithms. The 18.1% win in the physics test, which typically stresses collision detection and rigid body simulation, indicates that certain game physics or lightweight simulation tasks run better on Intel. However, these wins are isolated and do not reflect a general pattern of Intel superiority in any broad category.

The multi-threaded comparison deserves attention because both chips have 16 threads, yet AMD still leads by 18.3%. Equal thread counts make the AMD advantage a direct result of per-thread efficiency rather than a matter of having more parallel resources. Intel's higher core count, 12 versus 8, does not translate into a multi-threaded win. The physics test is the only place where Intel's extra physical cores appear to help, and even there the margin is moderate.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI Embedded P164 is built on TSMC's 4 nm process and uses the Gorgon Point codename, part of the Ryzen AI Embedded generation based on the Zen 5 and Zen 5c core architectures. The Intel Core 5 220H uses Intel's 10 nm process, with the Raptor Lake architecture and the Raptor Lake-H codename, part of the Core 5 Raptor Lake Refresh generation. The process node difference is significant: the AMD chip is manufactured at a smaller geometry, which contributes to its higher efficiency per clock.

Core configurations differ notably. The AMD chip has 8 cores and 16 threads, while the Intel chip has 12 cores and 16 threads. Intel achieves equal thread counts with fewer threads per core, meaning its 12 cores are a mix of performance and efficiency cores typical of the Raptor Lake hybrid design, though the exact core type breakdown is not recorded in the data. The AMD chip uses its 8 full cores with simultaneous multithreading to reach 16 threads.

Cache hierarchies also diverge. Both chips have 80 KB of L1 cache per core. The L2 cache differs: AMD provides 1 MB per core, while Intel provides 2 MB per core, giving Intel a larger per-core L2 footprint. L3 cache is where the gap reverses significantly. AMD has 8 MB total L3, while Intel has 18 MB shared L3. The larger Intel L3 helps in some repeated-access workloads, but the benchmark data shows it does not overcome AMD's architectural efficiency.

Clock speeds favor Intel on paper. The Intel chip has a base clock of 2.70 GHz and a boost clock of 4.90 GHz. The AMD chip has a base clock of 2.00 GHz and a boost clock of 5.00 GHz. Despite the lower base clock, AMD achieves an 18.3% higher single-thread score, which indicates the Zen 5 architecture delivers more instructions per clock. The higher boost clock on AMD, 5.00 GHz versus 4.90 GHz, also helps in burst workloads.

Memory support differs as well. AMD supports DDR5 and LPDDR5X, while Intel supports DDR4 and DDR5. AMD's memory bandwidth is recorded at 89.6 GB/s, while Intel's bandwidth is not listed. AMD also supports ECC memory, which Intel does not. PCIe capabilities differ: AMD provides Gen 4 with 16 lanes, while Intel provides Gen 5 with 8 lanes. The integrated graphics also differ, with AMD using the Radeon 880M and Intel using Iris Xe Graphics with 80 execution units.

The thermal design power is another clear distinction. AMD is rated at 28 W, while Intel is rated at 45 W. The lower AMD TDP makes it the more power-efficient part on paper, and its performance wins come despite drawing less thermal headroom. The Intel chip uses a different socket, BGA 1744, while AMD uses the FP8 socket.

FAQ

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

A: The AMD Ryzen AI Embedded P164 scores 4029 in the PassMark single-thread test, which is 18.3% higher than the Intel Core 5 220H's score of 3405.

Q: Does the Intel chip's higher core count give it a multi-threaded advantage?

A: No. Despite having 12 cores versus AMD's 8, the Intel chip scores 21884 in the PassMark multi-thread test, while AMD scores 25889, an 18.3% lead for AMD.

Q: What is the biggest performance gap between the two chips?

A: The extended instructions test shows the largest gap. AMD scores 24193, which is 65.2% higher than Intel's 14642.

Q: Are there any tests where Intel wins?

A: Yes. Intel wins the prime number search test, 82 versus 71, a 13.4% advantage, and the physics test, 1478 versus 1210, an 18.1% advantage.

Q: How do the average benchmark scores compare?

A: The AMD chip has an average benchmark score of 52901, placing it at the 91st percentile. The Intel chip averages 28574, placing it at the 80th percentile.

Q: What are the cache sizes for each processor?

A: Both have 80 KB of L1 per core. AMD has 1 MB of L2 per core and 8 MB of L3. Intel has 2 MB of L2 per core and 18 MB of shared L3.

The Verdict

The benchmark data is unambiguous. The AMD Ryzen AI Embedded P164 outperforms the Intel Core 5 220H in the vast majority of recorded tests, including the most representative general-purpose metrics like single-thread performance, multi-threaded throughput, integer math, and floating point math. Its 91st percentile ranking versus Intel's 80th percentile, combined with an average score of 52901 versus 28574, places it in a clearly higher performance tier.

The Intel Core 5 220H is only the better choice for workloads specifically dominated by prime number iteration or physics simulation, where its 13.4% and 18.1% wins respectively give it a narrow edge. For any broader definition of compute work, including compression, encryption, sorting, and instruction-heavy code, the AMD part delivers between 5.5% and 65.2% higher performance.

The AMD chip also does so at a lower thermal design power of 28 W compared to Intel's 45 W, with a smaller process node and ECC memory support. The Intel chip's advantages are limited to a larger L3 cache, DDR4 compatibility, and Gen 5 PCIe lanes. For users selecting between these two mobile processors, the recorded data supports the AMD Ryzen AI Embedded P164 as the stronger performer in nearly every measurable category.