AMD Ryzen AI 9 365 vs Intel Core Ultra 5 338H Comparison

The Verdict

The benchmark data splits these two mobile processors almost evenly, with Intel Core Ultra 5 338H taking 8 wins and AMD Ryzen AI 9 365 taking 7 in the recorded head-to-head tests. The AMD part is the multi-core throughput champion, leading Cinebench R23 multi-core by 14.5% and Cinebench R15 multi-core by 13.5%. The Intel part counters with decisive victories in single-threaded workloads and specialized math tasks, leading PassMark single-thread by 8.1% and floating point math by 25.3%.

For users whose primary workloads are rendering, video encoding, or any task that scales across cores, the Ryzen AI 9 365 delivers higher raw throughput in the Cinebench tests that matter most. For users running physics simulations, encryption tasks, or single-thread-bound applications, the Core Ultra 5 338H is the stronger choice. The AMD chip also holds a higher overall percentile ranking at 87 versus 84 for Intel, and its average benchmark score of 40048 exceeds the Intel part's 33989.

Architecture Differences

The two processors use fundamentally different manufacturing and core designs. AMD builds the Ryzen AI 9 365 on a 4 nm TSMC process with a Zen 5 architecture, codenamed Strix Point. It belongs to the Ryzen AI 300 generation, which combines Zen 5 and Zen 5c cores. The chip has 10 cores and 20 threads, with a base clock of 2.00 GHz and a boost clock of 5.00 GHz. Its thermal design power is 28 watts, and it uses the AMD Socket FP8. Cache is organized as 80 KB L1 per core, 1 MB L2 per core, and 16 MB of shared L3. The integrated graphics are Radeon 880M, and memory support covers DDR5 and LPDDR5X over a dual-channel bus with 89.6 GB/s bandwidth. PCIe connectivity is Gen 4 with 16 lanes from the CPU.

Intel builds the Core Ultra 5 338H on a 3 nm Intel process with the Panther Lake architecture, part of the Ultra 5 Panther Lake-H generation. It has 12 cores but only 12 threads, meaning no Hyper-Threading support, with a base clock of 1.90 GHz and a boost clock of 4.70 GHz. Its thermal design power is 25 watts, and it uses the Intel BGA 2540 socket. Cache is organized as 192 KB L1 per core, 2.5 MB L2 per core, and 18 MB of shared L3. The integrated graphics are Arc B370, and memory support is limited to LPDDR5X over a dual-channel bus with 136.5 GB/s bandwidth, which is 52.5% higher than the AMD part. PCIe connectivity is Gen 5 with 4 lanes from the CPU.

The Intel part uses a smaller process node at 3 nm versus 4 nm, but the AMD part has twice the thread count (20 versus 12). The Intel chip also carries a 2 MB larger L3 cache (18 MB versus 16 MB) and higher per-core L1 and L2 allocations. The AMD chip offers a higher boost clock at 5.00 GHz versus 4.70 GHz, and broader memory support with DDR5 in addition to LPDDR5X.

Head-to-Head Benchmarks

The largest single win for AMD comes in PassMark integer math, where the Ryzen AI 9 365 scores 101831 against 64934 for Intel, a 56.8% advantage. This is the widest margin in any recorded test. Data compression also favors AMD heavily, with a score of 354510 versus 276539, a 28.2% lead. Random string sorting goes to AMD at 39447 versus 34082, a 15.7% margin. Extended instructions favor AMD by 5%, with scores of 25113 versus 23906. The multi-thread PassMark test shows a narrower AMD win at 29467 versus 28717, a 2.6% edge.

Intel's largest win comes in PassMark find prime numbers, where it scores 304 against 117 for AMD, a 61.5% advantage. Floating point math also goes strongly to Intel at 84067 versus 62802, a 25.3% lead. Physics tests favor Intel at 2697 versus 1704, a 36.8% margin. Data encryption goes to Intel at 21367 versus 18297, a 14.4% lead. Single-thread tests show Intel ahead at 4180 versus 3841, an 8.1% margin. In Cinebench R23 single-core, Intel leads 2044 versus 1992, a 2.5% edge, and in Cinebench R15 single-core, Intel leads 305 versus 303, a 0.7% margin.

Cinebench multi-core results are the defining differentiator. The Ryzen AI 9 365 scores 18698 in R23 multi-core versus 16331 for Intel, a 14.5% lead. In R15 multi-core, AMD scores 2842 versus 2504, a 13.5% lead. These results show that despite having fewer cores, the AMD chip's thread count advantage of 20 versus 12 translates directly into rendering performance.

The Intel part also shows a higher single-core score in PassMark at 4180 versus 3841, and a higher Cinebench R23 single-core score at 2044 versus 1992. The AMD chip's higher boost clock does not overcome Intel's architectural efficiency in single-threaded tasks.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 5 338H has 12 cores, but the AMD Ryzen AI 9 365 has 10 cores and 20 threads. The AMD part has 20 threads versus 12 for Intel, giving it a substantial thread-count advantage.

Q: What is the largest benchmark margin between the two?

A: The largest margin is in PassMark find prime numbers, where Intel scores 304 versus 117 for AMD, a 61.5% advantage. The next largest is PassMark integer math, where AMD leads by 56.8%.

Q: Which chip has higher memory bandwidth?

A: The Intel Core Ultra 5 338H has 136.5 GB/s memory bandwidth versus 89.6 GB/s for the AMD Ryzen AI 9 365. Intel also supports only LPDDR5X memory, while AMD supports both DDR5 and LPDDR5X.

Q: How do the integrated graphics compare?

A: The AMD Ryzen AI 9 365 uses Radeon 880M graphics, while the Intel Core Ultra 5 338H uses Arc B370 graphics. The database does not include graphics-specific benchmark scores for either part.

Q: Which processor has a higher overall benchmark percentile?

A: The AMD Ryzen AI 9 365 is in the 87th percentile among all CPUs, while the Intel Core Ultra 5 338H is in the 84th percentile. The AMD part also has a higher average benchmark score at 40048 versus 33989.

Q: Which chip wins more head-to-head tests?

A: The Intel Core Ultra 5 338H wins 8 of the 15 recorded head-to-head tests, while the AMD Ryzen AI 9 365 wins 7. Intel's wins are concentrated in single-thread and math-heavy workloads, while AMD's wins are in multi-thread and integer-heavy workloads.

Where Each One Wins

The AMD Ryzen AI 9 365 is the clear choice for multi-core rendering workloads. Its Cinebench R23 multi-core score of 18698 puts it 14.5% ahead of Intel, and its Cinebench R15 multi-core score of 2842 is 13.5% ahead. For integer math, the AMD part dominates with a 56.8% lead, making it better suited for tasks like compression, sorting, and general data processing. The data compression score of 354510 versus 276539 reinforces this strength, as does the random string sorting score of 39447 versus 34082. The extended instructions score of 25113 versus 23906 shows a 5% edge in SIMD-heavy workloads.

The Intel Core Ultra 5 338H is the better option for single-threaded performance. Its PassMark single-thread score of 4180 is 8.1% ahead, and its Cinebench R23 single-core score of 2044 is 2.5% ahead. For floating point math, Intel leads by 25.3% with a score of 84067 versus 62802, which matters for scientific computing and financial modeling. The physics score of 2697 versus 1704 represents a 36.8% advantage, making Intel the better fit for physics simulation workloads. Data encryption also favors Intel at 21367 versus 18297, a 14.4% margin. The find prime numbers result of 304 versus 117 shows Intel's strength in sequential number-crunching tasks.

In practical terms, the AMD Ryzen AI 9 365 fits workloads that use all available threads: video rendering, code compilation, batch image processing, and database operations. The Intel Core Ultra 5 338H fits workloads that rely on single-core speed or floating point throughput: physics engines, encryption, spreadsheet calculations, and legacy single-threaded applications. The overall win split of 8 to 7 means neither part is universally superior, and the choice depends entirely on the specific task mix. The AMD part's higher percentile ranking at 87 versus 84 suggests it performs well across a broader range of benchmarks, but the Intel part's wins in specialized math tests make it the stronger specialist in those domains.