AMD Ryzen AI 9 365 vs Intel Core Ultra 5 235H Comparison

Head-to-Head Benchmarks

The head-to-head results between the AMD Ryzen AI 9 365 and Intel Core Ultra 5 235H reveal a clear split: Intel wins most individual tests, but AMD secures a decisive victory in integer-heavy workloads. Of the 15 recorded comparisons, Intel takes 10 wins while AMD claims 5, yet the margin of AMD's victories in raw compute tasks is substantial.

The most lopsided result favors AMD in PassMark integer math. The Ryzen AI 9 365 scores 101831 against Intel's 74247, a 37.2% advantage. This is the largest delta in any test and suggests the Zen 5 architecture handles integer operations with exceptional efficiency. Data compression also goes AMD's way, with a score of 354510 versus 301979, a 17.4% lead. Random string sorting follows the same pattern, with AMD ahead at 39447 versus 36208, an 8.9% gain, and extended instructions show a 7.5% edge at 25113 against 23354.

Intel's wins are more numerous but often narrower. The Cinebench R23 multicore test delivers Intel's biggest victory: 25598 against AMD's 18698, a 27% margin. Single-core R23 is even more dramatic, with Intel scoring 3613 versus 1992, a 44.9% gap that represents the largest delta of any benchmark. Cinebench R15 single-core also favors Intel by 16.8% (364 versus 303), though R15 multicore goes to AMD by 10.2% (2842 against 2580).

PassMark floating point math heavily favors Intel at 93509 versus 62802, a 32.8% lead. Prime number finding shows Intel at 239 against AMD's 117, a 51% gap. Data encryption goes Intel's way by 20.9% (23121 versus 18297), while physics testing shows Intel ahead by 14.2% (1985 versus 1704). The multithread score is nearly even, with Intel winning 30091 to 29467, just 2.1% apart. Single-thread tests give Intel an 11.9% edge at 4359 versus 3841 in both PassMark single_thread and singlethread results.

The overall average benchmark score reflects this split: AMD posts 40048 against Intel's 37522. AMD sits at the 87th percentile among all CPUs, while Intel ranks at the 85th. AMD's nearest rival is the AMD Ryzen 7 7700 with an average score of 40081, a negligible 0.1% difference, and the Intel Core 5 221E at 40144, just 0.2% behind. Intel's closest competitor is the Intel Core i9-13900HK at 37425, which trails by 0.3%, while the Intel Core Ultra 5 225F sits 0.6% behind.

Architecture Differences

The two processors employ fundamentally different design philosophies. AMD's Ryzen AI 9 365 uses the Zen 5 architecture on a 4 nm TSMC process, with the Strix Point codename and a die size of 233 mm². It relies on a hybrid configuration of Zen 5 and Zen 5c cores, totaling 10 cores and 20 threads. Intel's Core Ultra 5 235H, by contrast, uses the Arrow Lake architecture on a 3 nm TSMC process with the Arrow Lake-H codename, delivering 14 cores but only 14 threads, meaning it lacks simultaneous multithreading.

Cache allocations differ sharply. AMD provides 80 KB of L1 per core and 1 MB of L2 per core, with 16 MB of shared L3. Intel offers 192 KB of L1 per core and 3 MB of L2 per core, with 18 MB of shared L3. Intel's larger per-core caches align with its higher single-thread results, while AMD's smaller per-core L3 still supports strong multithreaded throughput.

Memory bandwidth favors Intel at 102.4 GB/s versus AMD's 89.6 GB/s, though both support DDR5 and LPDDR5X over dual-channel buses. PCIe connectivity diverges: AMD provides Gen 4 with 16 CPU lanes, while Intel offers Gen 5 with 8 CPU lanes, a tradeoff between bandwidth per lane and total lane count.

Integrated graphics differ as well. AMD pairs the CPU with Radeon 880M, while Intel uses Arc Graphics 140T. Both processors run at a 28 W TDP and boost to 5.00 GHz, but base clocks differ: AMD starts at 2.00 GHz, Intel at 2.40 GHz. AMD uses the AMD Socket FP8, Intel uses Intel BGA 2049. Neither supports ECC memory, and both are unlocked multipliers are false, meaning they are not overclockable.

Where Each One Wins

The benchmark data points to distinct usage scenarios. AMD's Ryzen AI 9 365 excels in integer arithmetic and data manipulation tasks. Its 37.2% lead in integer math and 17.4% edge in data compression make it suited for database operations, spreadsheet calculations, and compression workloads. The 8.9% advantage in random string sorting further supports this profile, suggesting strong performance in text processing and sorting algorithms. Extended instructions at 7.5% ahead indicate solid SIMD and vector handling.

Intel's Core Ultra 5 235H dominates in floating-point and single-threaded scenarios. The 32.8% lead in floating point math positions it for scientific computing, 3D rendering, and simulation workloads. Single-core results, particularly the 44.9% advantage in Cinebench R23 single-core, show Intel's architecture excels in latency-sensitive tasks like legacy software, web browsing, and lightly threaded applications. The 51% edge in prime number finding reflects superior scalar integer throughput per core, while the 20.9% lead in encryption suggests faster cryptographic operations.

The PassMark multithread result is nearly identical, with Intel ahead by just 2.1%, implying both chips deliver comparable throughput for general parallel workloads. Physics testing gives Intel a 14.2% edge, which may matter for gaming physics simulation. AMD's Cinebench R15 multicore win by 10.2% contrasts with Intel's R23 multicore win by 27%, indicating that benchmark version differences can shift the picture. Overall, AMD wins 5 head-to-head tests, Intel wins 10, but the average benchmark score still favors AMD slightly due to the magnitude of its integer victories.

FAQ

Q: Which processor has the higher single-core performance?

A: Intel's Core Ultra 5 235H wins every single-core test. It leads by 16.8% in Cinebench R15 single-core, 44.9% in Cinebench R23 single-core, and 11.9% in PassMark single-thread tests.

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

A: Results vary by test. AMD wins Cinebench R15 multicore by 10.2% (2842 versus 2580), but Intel wins Cinebench R23 multicore by 27% (25598 versus 18698). PassMark multithread is nearly even, with Intel ahead by just 2.1%.

Q: What is the core and thread count difference?

A: AMD has 10 cores and 20 threads. Intel has 14 cores and 14 threads. Intel lacks simultaneous multithreading, so its thread count equals its core count.

Q: Which CPU has better memory bandwidth?

A: Intel provides higher memory bandwidth at 102.4 GB/s, while AMD offers 89.6 GB/s. Both support DDR5 and LPDDR5X memory over dual-channel buses.

Q: What process node does each processor use?

A: AMD uses a 4 nm TSMC process, while Intel uses a 3 nm TSMC process. Both are fabricated by TSMC.

Q: Which processor has the higher average benchmark score?

A: AMD achieves an average benchmark score of 40048, placing it at the 87th percentile. Intel averages 37522, at the 85th percentile, meaning AMD edges ahead in overall performance.

Specification Differences

| Specification | AMD Ryzen AI 9 365 | Intel Core Ultra 5 235H |

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

| Cores | 10 | 14 |

| Threads | 20 | 14 |

| Base clock | 2.00 GHz | 2.40 GHz |

| Socket | AMD Socket FP8 | Intel BGA 2049 |

| Architecture | Zen 5 | Arrow Lake |

| Codename | Strix Point | Arrow Lake-H |

| Generation | Ryzen AI 300 (Zen 5 / Zen 5c) | Ultra 5 (Arrow Lake-H) |

| Process node | 4 nm | 3 nm |

| Die size | 233 mm² | N/A |

| L1 cache | 80 KB per core | 192 KB per core |

| L2 cache | 1 MB per core | 3 MB per core |

| L3 cache | 16 MB | 18 MB shared |

| Memory bandwidth | 89.6 GB/s | 102.4 GB/s |

| PCIe | Gen 4, 16 lanes | Gen 5, 8 lanes |

| Integrated graphics | Radeon 880M | Arc Graphics 140T |

| Release date | 2024-06-30 | 2025-01-12 |

| Part number | 100-000001530 | SRQAP |

Both processors share a 28 W TDP, a 5.00 GHz boost clock, dual-channel memory support, and both lack ECC memory support. Neither has an unlocked multiplier, and both are active production parts for the mobile segment.