AMD Ryzen AI Embedded P174i vs Intel Core 5 120HL Comparison

Head-to-Head Benchmarks

The database contains no recorded benchmark scores for either processor, so a direct performance comparison cannot be expressed in numerical terms. Both CPUs hold a percentile ranking of 50 against all CPUs in the database, which places them at the median of the recorded population. Without head-to-head benchmark entries, the largest wins in either direction remain unmeasured. What can be established from the recorded data is the structural capacity of each design, which suggests where performance differences would likely appear once workloads are applied.

The AMD Ryzen AI Embedded P174i uses a 10 core, 20 thread configuration. The Intel Core 5 120HL uses a 12 core, 16 thread setup. The Intel part has two more physical cores, but the AMD part has four more threads. In heavily threaded workloads, the AMD processor has the theoretical advantage of additional logical processors. In tasks that scale with physical core count, the Intel processor has the edge. The AMD chip boosts to 5.00 GHz, while the Intel chip boosts to 4.70 GHz. That 0.30 GHz difference in maximum boost clock favors the AMD part in lightly threaded tasks where single-core frequency dominates. The Intel part has a higher base clock at 2.60 GHz compared to 2.00 GHz on the AMD, which can help sustain moderate workloads without reaching boost states.

The absence of benchmark scores means the database cannot confirm how these specifications translate into real-world performance. The data does show that the AMD processor operates within a 28 watt TDP, while the Intel processor is rated at 45 watts. That power gap suggests the AMD part will sustain its boost behavior for longer in thermally constrained environments, while the Intel part draws more power to feed its higher base clock and additional physical cores.

Where Each One Wins

The use-case split follows directly from the specification differences recorded in the database. The AMD Ryzen AI Embedded P174i targets mobile integration, as indicated by its market segment field. Its 28 watt TDP and support for LPDDR5X memory make it suited for compact, power-limited systems where sustained throughput matters more than raw peak performance. The 10 core, 20 thread layout with a 5.00 GHz boost clock gives it a profile that serves both parallel workloads and bursty single-threaded tasks. The integrated Radeon 880M graphics provide a feature set that can handle general visual output without a discrete GPU.

The Intel Core 5 120HL is classified as a desktop part. Its 45 watt TDP and Intel Socket 1700 compatibility point toward systems with more robust cooling and power delivery. The 12 core, 16 thread configuration with a 4.70 GHz boost clock suits workloads that use many physical cores, such as video encoding, compilation, and simulation tasks. The Iris Xe Graphics 80EU integrated GPU offers a different graphics feature set, and the processor supports both DDR4 and DDR5 memory, which gives platform builders flexibility in memory selection.

The recorded data shows the AMD part supports ECC memory, while the Intel part does not. That places the AMD processor in reliability-focused roles where memory corruption cannot be tolerated, such as embedded controllers, network appliances, or industrial systems. The Intel processor, without ECC, aligns with consumer desktop use where error-correcting memory is not a requirement.

Architecture Differences

The AMD Ryzen AI Embedded P174i is built on the Gorgon Point codename and belongs to the Ryzen AI Embedded generation, which uses a hybrid arrangement of Zen 5 and Zen 5c cores. It is fabricated on a 4 nm process at TSMC. The Intel Core 5 120HL uses the Raptor Lake architecture, specifically the Raptor Lake-PS codename, and is built on a 10 nm process at Intel. The process node difference is significant: the AMD chip uses a denser manufacturing process, which contributes to its lower 28 watt TDP despite a 5.00 GHz boost clock.

Cache layouts differ between the two designs. The AMD processor has 80 KB of L1 cache per core and 1 MB of L2 per core, with 16 MB of L3 cache. The Intel processor also has 80 KB of L1 per core but doubles the L2 allocation to 2 MB per core, with 18 MB of shared L3 cache. The larger L2 and L3 allocations on the Intel part provide more on-die storage for frequently accessed data, which can reduce memory traffic in workloads with moderate working sets. The AMD part compensates with a smaller overall cache footprint but a higher boost clock and more threads.

Memory support diverges clearly. The AMD processor supports DDR5 and LPDDR5X, with a recorded memory bandwidth of 89.6 GB/s over a dual-channel bus. The Intel processor supports DDR4 and DDR5, also over a dual-channel bus, but no memory bandwidth figure is recorded in the database. ECC memory support is present on the AMD part and absent on the Intel part. The Intel processor offers 8 PCIe Gen 4 lanes from the CPU, while the AMD processor offers 16 PCIe Gen 4 lanes from the CPU. That doubles the direct CPU-attached expansion capacity on the AMD side, which matters for systems needing fast NVMe storage or add-in cards without going through the chipset.

The Intel part's Raptor Lake architecture uses a hybrid core design, which is a different implementation approach from the Zen 5 and Zen 5c combination in the AMD part. The database records the Intel generation as Core 5 (Raptor Lake-PS) and does not list a hybrid core breakdown, so the exact performance-core and efficiency-core split is not specified. The AMD generation is listed as Ryzen AI Embedded (Zen 5 / Zen 5c), confirming a mix of full-performance and compact cores.

Specification Differences

The two processors differ across several recorded fields. Core count: 10 on the AMD, 12 on the Intel. Thread count: 20 on the AMD, 16 on the Intel. Base clock: 2.00 GHz on the AMD, 2.60 GHz on the Intel. Boost clock: 5.00 GHz on the AMD, 4.70 GHz on the Intel. TDP: 28 watts on the AMD, 45 watts on the Intel. Socket: AMD Socket FP8 on the AMD, Intel Socket 1700 on the Intel. Process node: 4 nm on the AMD, 10 nm on the Intel. Foundry: TSMC on the AMD, Intel on the Intel. Codename: Gorgon Point on the AMD, Raptor Lake-PS on the Intel. Generation: Ryzen AI Embedded (Zen 5 / Zen 5c) on the AMD, Core 5 (Raptor Lake-PS) on the Intel. Die size: 233 mm² on the AMD, not recorded on the Intel. L2 cache: 1 MB per core on the AMD, 2 MB per core on the Intel. L3 cache: 16 MB on the AMD, 18 MB shared on the Intel. Memory support: DDR5 and LPDDR5X on the AMD, DDR4 and DDR5 on the Intel. Memory bandwidth: 89.6 GB/s on the AMD, not recorded on the Intel. ECC memory: supported on the AMD, not supported on the Intel. PCIe lanes: 16 Gen 4 lanes on the AMD, 8 Gen 4 lanes on the Intel. Integrated graphics: Radeon 880M on the AMD, Iris Xe Graphics 80EU on the Intel. Market segment: Mobile on the AMD, Desktop on the Intel. Release date: 2026-02-28 for the AMD, 2024-04-07 for the Intel. Launch MSRP: not recorded for the AMD, $279 for the Intel. The Intel part has a recorded part number of SRPFR; the AMD part number is listed as unknown. Neither processor has an unlocked multiplier.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core 5 120HL has 12 cores and 16 threads. The AMD Ryzen AI Embedded P174i has 10 cores and 20 threads. The Intel part has two more physical cores, while the AMD part has four more threads.

Q: What are the maximum boost clocks of each processor?

A: The AMD Ryzen AI Embedded P174i boosts to 5.00 GHz. The Intel Core 5 120HL boosts to 4.70 GHz. The AMD part has a 0.30 GHz higher boost clock.

Q: Do both processors support ECC memory?

A: No. The AMD Ryzen AI Embedded P174i supports ECC memory. The Intel Core 5 120HL does not support ECC memory.

Q: What memory types does each processor support?

A: The AMD Ryzen AI Embedded P174i supports DDR5 and LPDDR5X. The Intel Core 5 120HL supports DDR4 and DDR5. The AMD part also has a recorded memory bandwidth of 89.6 GB/s, while the Intel part has no recorded memory bandwidth figure.

Q: How many PCIe Gen 4 lanes does each CPU provide?

A: The AMD Ryzen AI Embedded P174i provides 16 PCIe Gen 4 lanes from the CPU. The Intel Core 5 120HL provides 8 PCIe Gen 4 lanes from the CPU.

Q: What is the TDP of each processor?

A: The AMD Ryzen AI Embedded P174i has a 28 watt TDP. The Intel Core 5 120HL has a 45 watt TDP.

The Verdict

The database shows two processors designed for different roles despite some surface-level similarities. The AMD Ryzen AI Embedded P174i belongs in the mobile segment with a 28 watt TDP, ECC memory support, LPDDR5X compatibility, and 16 PCIe Gen 4 lanes. Those features point toward embedded systems, compact industrial PCs, and mobile workstations where power efficiency, memory reliability, and expansion capacity are priorities. The 10 core, 20 thread configuration with a 5.00 GHz boost clock gives it a strong mix of parallel throughput and single-threaded responsiveness within a tight power envelope. The 4 nm TSMC process and 89.6 GB/s memory bandwidth support this positioning.

The Intel Core 5 120HL targets the desktop segment with a 45 watt TDP, Intel Socket 1700, and support for both DDR4 and DDR5 memory. The 12 core, 16 thread layout favors workloads that use every physical core rather than relying on extra threads. The 2 MB per core L2 cache and 18 MB shared L3 cache provide a larger on-die data footprint than the AMD part. The lack of ECC support and the smaller 8 lane PCIe Gen 4 allocation make it less suited to reliability-focused or expansion-heavy embedded roles, but the higher base clock of 2.60 GHz helps with sustained moderate workloads.

The choice between the two comes down to the platform requirements. Systems that need ECC memory, LPDDR5X, lower power draw, more PCIe lanes, and a 2026 release date should use the AMD part. Systems that need more physical cores, DDR4 compatibility, a larger cache pool, and an established desktop socket should use the Intel part. The Intel processor has a recorded launch MSRP of $279, while no launch MSRP is recorded for the AMD processor. Neither processor shows benchmark scores in the database, so the final decision should be guided by the documented specification differences and the specific workload demands of the target system.