Intel Core Ultra 5 235A vs Intel Core Ultra 9 386H Comparison

Head-to-Head Benchmarks

The benchmark database records a decisive overall result: the Intel Core Ultra 5 235A wins 16 of the 17 head-to-head comparisons against the Intel Core Ultra 9 386H. The only recorded win for the Ultra 9 386H comes in PassMark physics, where it scores 3028 against 2437 for the Ultra 5 235A, a delta of 19.5% in favor of the mobile part.

The most striking separation appears in single-core Cinebench tests. In Cinebench R23 single-core, the Ultra 5 235A scores 4607 against 2071.5 for the Ultra 9 386H, a 122.4% advantage. Cinebench R15 single-core shows a similar pattern, with 464 versus 303.5, a 52.9% delta. The multi-core Cinebench results also favor the Ultra 5 235A, though the margin varies. Cinebench R23 multi-core delivers the largest multi-threaded gap: 32633 versus 20547, a 58.8% lead. Cinebench R20 multi-core shows a 6.9% edge (13705 versus 12820), while Cinebench R15 multi-core is nearly flat at 2% (3289 versus 3223).

PassMark results reinforce the same direction. The Ultra 5 235A leads in data compression by 11.8% (393800 versus 352365), in data encryption by 11% (30136 versus 27150), in extended instructions by 8.5% (31625 versus 29138), and in floating point math by 9.4% (118778 versus 108527). Integer math is a narrow 1.5% margin (88626 versus 87284). Random string sorting shows a 17.5% advantage (49489 versus 42135), and multithread performance sits 8.5% higher (38392 versus 35399). Single-thread PassMark scores are 4557 versus 4218, an 8% edge. Find prime numbers favors the Ultra 5 235A by 15% (392 versus 341).

The overall average benchmark scores in the database place the Ultra 5 235A at 48201 against 43210 for the Ultra 9 386H. The percentile rankings confirm the gap: the Ultra 5 235A sits in the 90th percentile of all CPUs, while the Ultra 9 386H sits in the 88th.

Where Each One Wins

The data points to a clear split. The Ultra 5 235A wins every Cinebench test, including both single-core and multi-core variants across R15, R20, and R23. It also dominates most PassMark workloads involving arithmetic, data handling, and string operations. This includes integer math, floating point math, data compression, encryption, extended instructions, prime number finding, random string sorting, and multithreaded throughput.

The Ultra 9 386H wins only the PassMark physics test. The physics benchmark measures a specific simulation workload, and the 19.5% margin there is notable. For use cases that rely on physics simulation, the mobile part delivers a measurable advantage. Everywhere else in the recorded data, the desktop Ultra 5 235A holds the lead.

The magnitude of the single-core gap matters for lightly threaded applications. A 122.4% difference in Cinebench R23 single-core means the Ultra 5 235A will be substantially faster for tasks that run on a single thread. The multi-core gaps range from 2% to 58.8% across Cinebench versions, so heavily threaded workloads will see a more variable but still consistent advantage.

Architecture Differences

The two processors come from different Intel lineups. The Ultra 5 235A belongs to the Core Ultra Series 2, uses the Arrow Lake architecture with the Arrow Lake-S codename, and is fabricated on a 3 nm process at TSMC. The Ultra 9 386H belongs to Core Ultra Series 3, uses Panther Lake architecture with the Panther Lake-H codename, and is fabricated on a 3 nm process at Intel foundries.

The core configurations differ. The Ultra 5 235A has 14 cores and 14 threads. The Ultra 9 386H has 16 cores and 16 threads. Neither processor enables simultaneous multithreading, as both have equal core and thread counts.

Cache hierarchies show structural differences. Both use a 192 KB L1 per core. The L2 cache is 3 MB per core on the Ultra 5 235A versus 2.5 MB per core on the Ultra 9 386H. The L3 cache is 24 MB shared on the Ultra 5 235A versus 18 MB shared on the Ultra 9 386H. The larger L2 and L3 on the desktop part align with its higher benchmark results.

Integrated graphics differ as well. The Ultra 5 235A uses Arc Xe-LPG Graphics with 24 execution units. The Ultra 9 386H uses Intel Xe3 Graphics. The database does not record comparative graphics benchmarks for these two parts, so the analysis here is limited to architectural specifications.

Memory support also separates them. The Ultra 5 235A supports DDR5 only, while the Ultra 9 386H supports both DDR5 and LPDDR5X. The memory bus is dual-channel for both. Memory bandwidth is recorded at 102.4 GB/s for the Ultra 5 235A and 115.2 GB/s for the Ultra 9 386H.

PCIe connectivity differs: the Ultra 5 235A offers Gen 5 with 20 CPU lanes, while the Ultra 9 386H offers Gen 5 with 12 CPU lanes. Neither supports ECC memory, and neither has an unlocked multiplier.

Specification Differences

The following fields differ between the two processors in the database:

• Cores: 14 versus 16
• Threads: 14 versus 16
• Base clock: 3.40 GHz versus 2.10 GHz
• Boost clock: 5.00 GHz versus 4.90 GHz
• TDP: 65 W versus 25 W
• Socket: Intel Socket 1851 versus Intel BGA 2540
• Architecture: Arrow Lake versus Panther Lake
• Codename: Arrow Lake-S versus Panther Lake
• Generation: Ultra 5 (Arrow Lake) versus Ultra 9 (Panther Lake-H)
• Foundry: TSMC versus Intel
• Transistors: 17,800 million versus not recorded
• Die size: 243 mm² versus not recorded
• L2 cache: 3 MB per core versus 2.5 MB per core
• L3 cache: 24 MB shared versus 18 MB shared
• Memory support: DDR5 versus DDR5, LPDDR5X
• Memory bandwidth: 102.4 GB/s versus 115.2 GB/s
• PCIe lanes: 20 CPU lanes versus 12 CPU lanes
• Integrated graphics: Arc Xe-LPG Graphics 24EU versus Intel Xe3 Graphics
• Market segment: Desktop versus Mobile
• Release date: 2025-07-28 versus 2026-01-04
• Launch MSRP: $269 versus not recorded
• Part number: SRWPN versus SA4R5Q9EH

FAQ

Q: Which processor has a higher boost clock?

A: The Intel Core Ultra 5 235A boosts to 5.00 GHz, while the Intel Core Ultra 9 386H boosts to 4.90 GHz.

Q: How large is the L3 cache on each processor?

A: The Ultra 5 235A has 24 MB of shared L3 cache. The Ultra 9 386H has 18 MB of shared L3 cache.

Q: Which processor supports LPDDR5X memory?

A: Only the Intel Core Ultra 9 386H supports LPDDR5X memory. The Ultra 5 235A supports DDR5 only.

Q: What is the TDP of each processor?

A: The Ultra 5 235A has a TDP of 65 W. The Ultra 9 386H has a TDP of 25 W.

Q: Which processor has more cores?

A: The Intel Core Ultra 9 386H has 16 cores, while the Intel Core Ultra 5 235A has 14 cores. Both have matching thread counts.

Q: What is the average benchmark score for each processor?

A: The Ultra 5 235A has an average benchmark score of 48201. The Ultra 9 386H has an average benchmark score of 43210.

The Verdict

The recorded data consistently favors the Intel Core Ultra 5 235A for general computation. It wins 16 of 17 head-to-head benchmarks, including every Cinebench test and nearly all PassMark workloads. Its average benchmark score is 11.5% higher than the Ultra 9 386H, and its 90th percentile ranking beats the 88th percentile of the Ultra 9 386H. For desktop users running multi-threaded applications, Cinebench R23 multi-core shows a 58.8% advantage, and single-thread performance is more than double in the same test version.

The Intel Core Ultra 9 386H serves a different role. As a mobile part with a 25 W TDP, it is built for power-constrained environments. Its 16 cores and 16 threads outnumber the desktop part, yet it trails in most measured workloads. The single exception is PassMark physics, where it leads by 19.5%. The Ultra 9 386H also supports LPDDR5X memory and has a higher memory bandwidth rating at 115.2 GB/s, despite the Ultra 5 235A having a larger cache hierarchy.

Users who need maximum processing throughput, whether single-threaded or multi-threaded, should look to the Ultra 5 235A based on the benchmark records. Users who require a mobile form factor with low power draw and who prioritize physics simulation workloads may find the Ultra 9 386H more appropriate. The data does not support any other use case for the Ultra 9 386H over the Ultra 5 235A.