AMD Ryzen AI Embedded P185 vs Intel Core Ultra 5 235H Comparison

Head-to-Head Benchmarks

The recorded benchmark data splits cleanly between the two processors, with the AMD Ryzen AI Embedded P185 taking 5 wins and the Intel Core Ultra 5 235H taking 6. The margins, however, tell a more revealing story than the raw win count.

The AMD part's largest victory comes in integer math, where it scores 117832 against Intel's 74247, a 58.7% advantage. This is the single biggest delta across all shared tests. Data compression also favors AMD heavily, with 374429 versus 301979, a 24% lead. Extended instructions go to AMD by 13.7% (26544 against 23354), and random string sorting shows a 12% edge (40557 versus 36208). The multithread score, often the headline metric for overall throughput, goes to AMD by a narrower 5.7% margin, 31817 to 30091.

Intel's wins are concentrated in different workload families. Floating point math is decisively Intel's, with 93509 versus 70587, a 24.5% advantage. Find prime numbers shows Intel at 239 against AMD's 129, a 46% gap. Data encryption favors Intel by 15.2% (23121 versus 19612), and physics simulation goes to Intel by 10.7% (1985 versus 1772). Single-thread performance, a critical metric for legacy and lightly threaded applications, goes to Intel by 8.8%, with 4359 against 3977.

The pattern suggests a division of labor: AMD dominates integer-heavy, compression-oriented, and multi-threaded workloads, while Intel leads in floating-point math, encryption, and single-thread responsiveness. The multithread win for AMD, despite Intel having 14 cores to AMD's 12, indicates that AMD's simultaneous multithreading (24 threads versus 14) provides a meaningful throughput advantage in parallel tasks. Yet the single-thread deficit of 8.8% means AMD relies on core count and thread count to overcome Intel's faster per-core execution.

FAQ

Q: Which processor has the higher overall average benchmark score?

A: The AMD Ryzen AI Embedded P185 records an average benchmark score of 62839, placing it in the 93rd percentile of all CPUs. The Intel Core Ultra 5 235H averages 37522, placing it in the 85th percentile.

Q: How do the two compare in multi-threaded performance?

A: The AMD Ryzen AI Embedded P185 scores 31817 in PassMark multithread, which is 5.7% higher than Intel's 30091. AMD achieves this with 12 cores and 24 threads, while Intel has 14 cores and 14 threads.

Q: Which processor wins in single-thread performance?

A: The Intel Core Ultra 5 235H scores 4359 in PassMark single-thread, which is 8.8% higher than AMD's 3977. This is one of Intel's larger wins in the head-to-head data.

Q: What is the difference in cache capacity?

A: The AMD Ryzen AI Embedded P185 has 16 MB of L3 cache, while the Intel Core Ultra 5 235H has 18 MB of shared L3 cache. Intel also has larger per-core L1 (192 KB versus 80 KB) and L2 (3 MB versus 1 MB) allocations.

Q: Do both processors support the same memory types?

A: Both support DDR5 and LPDDR5X memory. However, Intel's memory bandwidth is 102.4 GB/s, while AMD's is 89.6 GB/s. Intel does not support ECC memory, whereas AMD does.

Q: Which processor has a higher boost clock?

A: The AMD Ryzen AI Embedded P185 boosts to 5.10 GHz, while the Intel Core Ultra 5 235H boosts to 5.00 GHz. AMD's base clock is 2.00 GHz, while Intel's is 2.40 GHz.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI Embedded P185 uses the Gorgon Point codename, built on the Zen 5 and Zen 5c hybrid core architecture. It is fabricated on a 4 nm process at TSMC, with a die size of 233 mm². The Intel Core Ultra 5 235H uses the Arrow Lake-H codename, part of the Core Ultra Series 2, built on a 3 nm process, also at TSMC. Intel's die size is not recorded in the database.

The core configurations diverge sharply. AMD fields 12 cores and 24 threads, indicating simultaneous multithreading is enabled. Intel has 14 cores but only 14 threads, meaning it lacks hyper-threading on this part. This explains the multithread score gap: AMD's 24 threads can be scheduled across more logical processors, even if Intel has more physical cores.

Cache hierarchies also differ. AMD allocates 80 KB of L1 per core and 1 MB of L2 per core, with a shared 16 MB L3. Intel allocates 192 KB of L1 per core and 3 MB of L2 per core, with an 18 MB shared L3. The larger per-core caches on Intel may contribute to its single-thread and floating-point advantages, as more data can reside closer to the execution units.

The integrated graphics differ as well. AMD uses the Radeon 890M, while Intel uses Arc Graphics 140T. Neither benchmark data in this comparison covers GPU performance, so the analysis must remain CPU-centric. The memory controller also differs: AMD's memory bandwidth is 89.6 GB/s, while Intel's is 102.4 GB/s. Intel supports PCIe Gen 5 with 8 CPU lanes, whereas AMD uses PCIe Gen 4 with 16 CPU lanes.

Release timing shows Intel shipped earlier, with a release date of 2025-01-12, while AMD's part came later on 2026-02-28. Both are listed as active production parts for the mobile market segment. Neither has a recorded launch MSRP in the database.

Specification Differences

The following fields differ between the two processors:

• Cores: AMD has 12, Intel has 14.
• Threads: AMD has 24, Intel has 14.
• Base clock: AMD runs at 2.00 GHz, Intel at 2.40 GHz.
• Boost clock: AMD boosts to 5.10 GHz, Intel to 5.00 GHz.
• Socket: AMD uses Socket FP8, Intel uses BGA 2049.
• Codename: AMD is Gorgon Point, Intel is Arrow Lake-H.
• Generation: AMD is Ryzen AI Embedded (Zen 5 / Zen 5c), Intel is Ultra 5 (Arrow Lake-H).
• Process node: AMD uses 4 nm, Intel uses 3 nm.
• Die size: AMD measures 233 mm², Intel is not recorded.
• L1 cache: AMD has 80 KB per core, Intel has 192 KB per core.
• L2 cache: AMD has 1 MB per core, Intel has 3 MB per core.
• L3 cache: AMD has 16 MB, Intel has 18 MB shared.
• Memory bandwidth: AMD at 89.6 GB/s, Intel at 102.4 GB/s.
• ECC memory: Supported on AMD, not on Intel.
• PCIe: AMD uses Gen 4 with 16 lanes, Intel uses Gen 5 with 8 lanes.
• Integrated graphics: Radeon 890M versus Arc Graphics 140T.
• Part number: AMD is unknown, Intel is SRQAP.
• Release date: AMD on 2026-02-28, Intel on 2025-01-12.

Both share a TDP of 28 watts, dual-channel memory buses, DDR5 and LPDDR5X support, and locked multipliers. Both target the mobile market segment and are currently in active production.

The Verdict

The benchmark data indicates two different optimization targets. The AMD Ryzen AI Embedded P185 is the stronger choice for integer math, data compression, and multithreaded throughput. Its 24 threads provide a clear advantage in parallel workloads, and the 58.7% integer math lead suggests strong performance in database operations, compression algorithms, and general compute tasks that rely on integer arithmetic. The 93rd percentile ranking places it among the higher-performing mobile processors in the database.

The Intel Core Ultra 5 235H, with a 85th percentile ranking, is the better option for floating-point math, encryption, and single-threaded tasks. The 24.5% floating-point lead and the 46% prime number advantage indicate a processor well-suited for scientific calculations, financial modeling, or any workload with heavy transcendental functions. The 8.8% single-thread lead means faster response in legacy applications or workflows that cannot leverage multiple cores. Intel also delivers higher memory bandwidth, 102.4 GB/s versus 89.6 GB/s, which may benefit memory-bound workloads.

The overall average benchmark score strongly favors AMD, 62839 versus 37522, but the per-test breakdown shows that neither processor dominates universally. The data suggests a system designer should choose AMD for mixed multi-threaded server-like workloads and Intel for single-thread-sensitive or floating-point-heavy applications. The fact that AMD's nearest rivals include the Intel Core Ultra 7 255HX (0.2% higher average score) and the Intel Core Ultra 7 265HX (0.5% higher), while Intel's nearest rivals include the Intel Core i9-13900HK (0.3% higher) and the AMD Ryzen AI 5 PRO 435 (0.6% lower), further contextualizes their competitive positions.

Where Each One Wins

AMD Ryzen AI Embedded P185 wins in:

• Integer math: 117832 versus 74247, a 58.7% lead. This is the largest margin in any test.
• Data compression: 374429 versus 301979, a 24% lead.
• Extended instructions: 26544 versus 23354, a 13.7% lead.
• Random string sorting: 40557 versus 36208, a 12% lead.
• Multithread: 31817 versus 30091, a 5.7% lead.

These wins point toward workloads involving data processing, compression pipelines, and parallel integer computation. The multithread advantage, paired with 24 threads, makes this processor suitable for heavily threaded applications such as rendering, compilation, or virtualization environments where thread count matters more than per-core speed.

Intel Core Ultra 5 235H wins in:

• Find prime numbers: 239 versus 129, a 46% lead.
• Floating point math: 93509 versus 70587, a 24.5% lead.
• Data encryption: 23121 versus 19612, a 15.2% lead.
• Physics simulation: 1985 versus 1772, a 10.7% lead.
• Single-thread: 4359 versus 3977, an 8.8% lead.

These wins favor workloads with heavy floating-point arithmetic, cryptographic operations, and physics calculations. The single-thread lead also benefits interactive applications, scripting environments, and lightly threaded legacy software. The higher memory bandwidth of 102.4 GB/s may further assist these workloads, though the recorded benchmarks do not isolate memory bandwidth effects.

The split is nearly even, 5 wins for AMD and 6 for Intel, but the magnitude of AMD's integer and compression wins is substantially larger than most of Intel's wins. The floating-point and prime number wins for Intel are also large, but they apply to a narrower set of real-world applications. The data implies that for general-purpose mobile computing with a mix of workloads, AMD's multithread and integer advantages may provide broader benefit, while Intel's part is more specialized toward numeric and single-threaded tasks.