AMD Ryzen AI Embedded P164 vs Intel Core 5 223PE Comparison

Head-to-Head Benchmarks

The benchmark data shows a decisive sweep for the Intel Core 5 223PE across all eleven recorded head-to-head tests. The AMD Ryzen AI Embedded P164 does not secure a single win, with the Intel part leading by margins ranging from a modest 1.9% to a dominant 55.3%. The most substantial gap appears in prime number finding, where Intel scores 159 against AMD's 71, a delta of -55.3% from AMD's perspective. This suggests the Intel architecture handles integer-heavy iterative workloads with significantly greater efficiency.

The physics test reveals another major divide. Intel posts 2493 versus AMD's 1210, a 51.5% advantage. Physics simulations often stress memory latency and branch prediction, and the data indicates the Intel design has a clear edge in that domain. Floating point math also favors Intel heavily: 76468 against 55799, a 27% lead. Integer math shows a narrower but still meaningful gap, with Intel at 99819 and AMD at 87940, an 11.9% difference.

Multithreaded performance follows the same pattern. Intel scores 31124 in the PassMark multithread test, while AMD manages 25889, a 16.8% deficit. Both processors have 8 cores and 16 threads, so this gap reflects per-core efficiency and memory subsystem behavior rather than a raw core count advantage. The single-thread test shows a smaller edge for Intel: 4219 versus 4029, a 4.5% difference. This indicates Intel's higher boost clock and architecture design translate into a modest but consistent lead in lightly threaded workloads.

Data compression and encryption also go to Intel. Compression scores 346623 against 327891, a 5.4% margin, while encryption shows Intel at 18448 versus AMD's 16055, a 13% advantage. Extended instructions favor Intel by 1.9% (24672 vs 24193), and random string sorting shows a 2.8% edge (35798 vs 34801). The two single-thread entries in the database, listed as both passmark_single_thread and passmark_singlethread, record identical scores of 4029 for AMD and 4219 for Intel, confirming the consistency of the measurement.

The average benchmark score tells a different story, however. AMD's average across all recorded tests is 52901, while Intel's is 40585. This discrepancy arises because the head-to-head list draws exclusively from PassMark tests, whereas the average score includes Cinebench results for Intel that are not present in AMD's benchmark set. The AMD part sits at the 91st percentile among all CPUs, while Intel sits at the 87th percentile. AMD's nearest rivals include the AMD Ryzen 9 7900X and AMD EPYC 7313P, with delta percentages of -0.7% and -0.6% respectively, meaning AMD's average score is marginally below those parts. Intel's nearest rivals include the Intel Core 7 253PE and Intel Core Ultra X7 368H, with deltas of 0.1% and 0.2%, placing Intel slightly above those parts.

Architecture Differences

The two processors come from fundamentally different design philosophies. AMD's Ryzen AI Embedded P164 uses the Gorgon Point codename, built on the Ryzen AI Embedded generation with a hybrid of Zen 5 and Zen 5c cores. It is fabricated on a 4 nm process at TSMC, with a die size of 233 mm². Intel's Core 5 223PE uses the Bartlett Lake codename, part of the Core 5 generation, and is fabricated on a 10 nm process at Intel's own foundry. The process node difference is stark: 4 nm versus 10 nm suggests AMD should have an efficiency advantage, though the benchmark results show Intel winning on raw performance.

Cache configurations differ substantially. Both parts have 80 KB of L1 cache per core, but AMD provides 1 MB of L2 per core while Intel provides 2 MB per core. The L3 cache shows an even larger divergence: AMD has 8 MB total, while Intel has 24 MB shared. This triple-sized L3 cache likely contributes to Intel's strong performance in data compression and encryption workloads, where larger working sets can reside on-chip. The L2 doubling also helps with random string sorting and prime number generation.

Memory support differs in kind. AMD supports DDR5 and LPDDR5X, while Intel supports DDR4 and DDR5. Both run dual-channel memory buses with identical peak bandwidth of 89.6 GB/s. Both support ECC memory, which matters for embedded and reliability-focused deployments. The PCIe interfaces diverge: AMD offers Gen 4 with 16 lanes from the CPU, while Intel offers Gen 5 with 16 lanes. This gives Intel double the per-lane bandwidth for storage or accelerator connectivity.

Clock speeds favor Intel. The base clock is 2.90 GHz for Intel versus 2.00 GHz for AMD, and the boost clock is 5.20 GHz versus 5.00 GHz. The higher base clock alone could explain some of the single-thread performance gap. Thermal design power differs markedly: AMD is rated at 28 watts, while Intel is rated at 65 watts. This 37-watt difference reflects Intel's willingness to trade power for performance, and it aligns with the market segments: AMD is a mobile-part, while Intel is a desktop part. The sockets also differ, with AMD using Socket FP8 and Intel using Socket 1700, making direct platform swaps impossible.

Integrated graphics provide another contrast. AMD includes Radeon 880M graphics, while Intel includes UHD Graphics 730. Neither is a focus of the benchmark data, but the Radeon 880M is generally positioned for more demanding visual tasks. The release dates are identical in the database: both parts entered production on the same date, and both are marked as Active. Intel lists a launch MSRP of $232, while AMD has no recorded launch MSRP. Intel also has a recorded part number, SA4QF, while AMD's part number is unknown. Neither processor has an unlocked multiplier, so overclocking is not supported on either platform.

The Verdict

The recorded data indicates that the Intel Core 5 223PE is the stronger performer in every head-to-head PassMark benchmark. For workloads that prioritize raw throughput in compression, encryption, physics simulation, and floating point math, Intel holds a decisive advantage. The 55.3% lead in prime number finding and the 51.5% lead in physics are particularly large, and these are not marginal wins. The Intel part also carries a higher base clock, a higher boost clock, twice the L2 cache per core, and triple the L3 cache. The 65 watt TDP, however, means it requires more power and cooling than the 28 watt AMD part.

The AMD Ryzen AI Embedded P164 is a mobile-oriented processor with a 4 nm process, a much smaller power envelope, and LPDDR5X memory support. Its average benchmark score of 52901 places it at the 91st percentile, above Intel's 40585 average at the 87th percentile, because the average includes different test sets. The recorded data does not show AMD winning any head-to-head test, but the power efficiency and embedded positioning may suit different deployment scenarios. The 28 watt TDP suggests the AMD part can operate in thermally constrained systems where the 65 watt Intel part would not fit.

The database shows Intel's nearest rivals are all around the same average score, with deltas of 0.1%, -0.1%, and 0.2%. AMD's nearest rivals show deltas of 0.1%, -0.1%, -0.6%, and -0.7%, indicating that AMD's average score is close to several high-end desktop and server parts. For a system builder who needs maximum compute per watt, the AMD part appears more suitable. For a system builder who needs maximum compute per socket, the Intel part appears more suitable. The choice hinges on whether the 65 watt envelope is acceptable and whether the specific benchmark wins matter more than the overall average.

FAQ

Q: Which processor has the higher boost clock?

A: The Intel Core 5 223PE boosts to 5.20 GHz, while the AMD Ryzen AI Embedded P164 boosts to 5.00 GHz.

Q: How much L3 cache does each processor have?

A: The AMD part has 8 MB of L3 cache, while the Intel part has 24 MB shared L3 cache.

Q: What is the largest performance gap in the head-to-head results?

A: The prime number finding test shows the largest gap, with Intel scoring 159 versus AMD's 71, a 55.3% difference.

Q: Do both processors support ECC memory?

A: Yes, both the AMD Ryzen AI Embedded P164 and the Intel Core 5 223PE support ECC memory.

Q: What is the thermal design power difference?

A: The AMD part is rated at 28 watts, while the Intel part is rated at 65 watts.

Q: Which processor has a higher average benchmark score?

A: The AMD part has an average benchmark score of 52901, while the Intel part has an average of 40585.

Where Each One Wins

The Intel Core 5 223PE wins every single head-to-head benchmark in the database. The largest margins come in compute-heavy tests: prime number finding (159 vs 71), physics (2493 vs 1210), and floating point math (76468 vs 55799). These are workloads that benefit from large caches and high clock speeds, and Intel's 2 MB L2 per core and 24 MB L3 provide a substantial advantage. Data encryption also favors Intel strongly at 18448 versus 16055, a 13% gap, which matters for security-focused applications. Integer math and multithreaded performance show 11.9% and 16.8% leads respectively, meaning Intel is the pick for general productivity and parallel compute.

The AMD Ryzen AI Embedded P164 does not win any recorded head-to-head test, but it has structural advantages that the benchmarks do not capture. The 4 nm process node and 28 watt TDP indicate significantly lower power draw, which is critical for embedded systems, fanless designs, and battery-powered equipment. The LPDDR5X memory support allows for low-power memory configurations, and the mobile socket (FP8) targets compact form factors. The Radeon 880M integrated graphics likely provide better visual performance than Intel's UHD Graphics 730, though no graphics benchmarks appear in the database. The higher average benchmark score and 91st percentile ranking suggest that, across a broader test suite, the AMD part holds its own against much higher-power competitors.

For workloads that are latency-sensitive and cache-hungry, such as physics simulation and prime number sieving, the Intel part is the clear choice. For workloads that run in power-constrained environments, such as embedded controllers or portable devices, the AMD part is the only viable option given its 28 watt TDP. The PCIe Gen 5 support on Intel also makes it preferable for high-bandwidth storage arrays, while AMD's Gen 4 support is sufficient for most embedded use cases. The data does not show any scenario where the AMD part outperforms Intel in raw compute, but the power and platform characteristics give it a distinct role.