AMD Ryzen 9 8940HX vs AMD Ryzen AI 9 365 Comparison

FAQ

Q: Which processor delivers the higher average benchmark score?

A: The AMD Ryzen 9 8940HX records an average benchmark score of 81103, while the AMD Ryzen AI 9 365 records 40048. The 8940HX sits at the 95th percentile among all CPUs, whereas the AI 9 365 sits at the 87th percentile.

Q: How do the two processors compare in Cinebench R23 multi-core performance?

A: The Ryzen 9 8940HX scores 32521 in Cinebench R23 multi-core, compared to 18698 for the Ryzen AI 9 365. This gives the 8940HX a 73.9% advantage in that test.

Q: Which processor wins in single-core performance?

A: The Ryzen AI 9 365 wins in both Cinebench R15 single-core and Cinebench R23 single-core. It scores 303 versus 292 in R15, a 3.6% lead, and 1992 versus 1917 in R23, a 3.8% lead.

Q: What are the core and thread counts for each chip?

A: The Ryzen 9 8940HX has 16 cores and 32 threads. The Ryzen AI 9 365 has 10 cores and 20 threads.

Q: Which processor has the higher boost clock?

A: The Ryzen 9 8940HX boosts to 5.30 GHz, while the Ryzen AI 9 365 boosts to 5.00 GHz.

Q: How does the Ryzen 9 8940HX compare to its nearest rivals in average score?

A: The 8940HX is essentially tied with the Intel Xeon w5-3535X at 81115, the Intel Core i9-14900KS at 81127, and the AMD Ryzen AI Max+ PRO 395 at 80762. Its delta against those rivals is 0%, 0%, and 0.4% respectively.

Architecture Differences

The Ryzen 9 8940HX uses the Zen 4 architecture under the Dragon Range codename, while the Ryzen AI 9 365 uses the Zen 5 architecture under the Strix Point codename. This architectural gap is significant: Zen 5 represents a newer microarchitecture generation with different execution resources, branch prediction, and instruction handling compared to Zen 4.

The process nodes also differ. The 8940HX is built on a 5 nm process at TSMC, while the AI 9 365 uses a 4 nm process, also at TSMC. The die sizes diverge substantially: the 8940HX uses a dual-die configuration with 2x 71 mm² dies, totaling roughly 142 mm² of silicon, while the AI 9 365 uses a monolithic 233 mm² die. The transistor count for the 8940HX is listed at 13,140 million, whereas the AI 9 365 does not have a recorded transistor count in the database.

Cache hierarchies differ in structure. The 8940HX has 64 KB of L1 cache per core, 1 MB of L2 per core, and 64 MB of L3 cache. The AI 9 365 has 80 KB of L1 per core, 1 MB of L2 per core, but only 16 MB of L3 cache. The larger L3 on the 8940HX is a direct consequence of its dual-die design, where each die contributes to the shared pool.

The integrated graphics also differ. The 8940HX pairs with a Radeon 610M, while the AI 9 365 includes a Radeon 880M. The product generations reflect this split: the 8940HX belongs to the 8000 series, while the AI 9 365 belongs to the Ryzen AI 300 series. The socket types are incompatible: the 8940HX uses AMD Socket FL1, and the AI 9 365 uses AMD Socket FP8.

Power delivery differs sharply. The 8940HX has a 55 W TDP, while the AI 9 365 has a 28 W TDP. This 27 W gap indicates different thermal envelopes and likely different chassis requirements. The 8940HX also has an unlocked multiplier, whereas the AI 9 365 does not.

PCIe support differs. The 8940HX supports PCIe Gen 5 with 28 lanes (CPU only), while the AI 9 365 supports PCIe Gen 4 with 16 lanes (CPU only). Memory support is broader on the AI 9 365: it supports both DDR5 and LPDDR5X, while the 8940HX supports only DDR5. Both use dual-channel memory buses, but the AI 9 365 has a higher memory bandwidth at 89.6 GB/s versus 83.2 GB/s for the 8940HX. Neither chip supports ECC memory.

Where Each One Wins

The Ryzen 9 8940HX dominates heavily in multi-threaded workloads. Out of 15 head-to-head benchmark comparisons, it wins 13. The wins span Cinebench R15 multi-core, Cinebench R23 multi-core, and the entire Passmark suite except for the physics test where it still wins but by a smaller margin. The largest wins come in data encryption at 114.9% ahead, find prime numbers at 114.5% ahead, and random string sorting at 91.1% ahead. These are compute-intensive, parallel-friendly tasks where the 16-core, 32-thread configuration provides a massive throughput advantage over the 10-core, 20-thread part.

The Ryzen AI 9 365 wins only two benchmarks: Cinebench R15 single-core and Cinebench R23 single-core. Its wins are modest but consistent, with a 3.6% lead in R15 and a 3.8% lead in R23. This indicates that the Zen 5 architecture delivers better per-thread performance than Zen 4, at least in the single-core rendering workloads captured by Cinebench. The AI 9 365 also shows a slight edge in Passmark single-thread? No, the data shows the 8940HX wins Passmark single-thread by 0.9%, so the AI 9 365's single-core advantage is specific to Cinebench, not universal.

For workloads that rely on single-threaded rendering, such as certain CAD tools or older software that does not scale across cores, the AI 9 365 offers a small but measurable improvement. For everything else, the 8940HX is the stronger choice. The 8940HX also wins in Passmark physics by 23.5%, which is the smallest margin among its multi-threaded victories. That suggests the AI 9 365's cores are more efficient individually, but the sheer core count of the 8940HX overwhelms that efficiency in most parallel tasks.

The AI 9 365's lower TDP of 28 W versus 55 W makes it more suitable for thinner, lighter laptops where sustained multi-core load is less common. The 8940HX, with its higher TDP and unlocked multiplier, targets performance-oriented mobile workstations or gaming laptops that can accommodate the additional heat.

Specification Differences

The specification table below lists only the fields where the two processors differ.

| Specification | AMD Ryzen 9 8940HX | AMD Ryzen AI 9 365 |

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

| Series | 8000 series | Not specified |

| Cores | 16 | 10 |

| Threads | 32 | 20 |

| Base clock | 2.40 GHz | 2.00 GHz |

| Boost clock | 5.30 GHz | 5.00 GHz |

| TDP | 55 W | 28 W |

| Socket | AMD Socket FL1 | AMD Socket FP8 |

| Architecture | Zen 4 | Zen 5 |

| Codename | Dragon Range | Strix Point |

| Generation | Ryzen 9 (Zen 4, Dragon Range) | Ryzen AI 300 (Zen 5 / Zen 5c) |

| Process node | 5 nm | 4 nm |

| Transistors | 13,140 million | Not specified |

| Die size | 2x 71 mm² | 233 mm² |

| L1 cache | 64 KB per core | 80 KB per core |

| L3 cache | 64 MB | 16 MB |

| Memory support | DDR5 | DDR5, LPDDR5X |

| Memory bandwidth | 83.2 GB/s | 89.6 GB/s |

| PCIe | Gen 5, 28 lanes (CPU only) | Gen 4, 16 lanes (CPU only) |

| Integrated graphics | Radeon 610M | Radeon 880M |

| Release date | 2025-04-22 | 2024-06-30 |

| Multiplier unlocked | Yes | No |

| Part number | 100-000001849 | 100-000001530 |

Both processors share the same manufacturer, use dual-channel memory, lack ECC support, target the mobile market segment, and have active production status. The L2 cache is identical at 1 MB per core. Neither has a listed launch MSRP.

Head-to-Head Benchmarks

The benchmark results show a clear pattern: the Ryzen 9 8940HX wins almost every multi-threaded test, often by substantial margins, while the Ryzen AI 9 365 takes the two Cinebench single-core tests.

Starting with Cinebench R15 multi-core, the 8940HX scores 5355 against 2842 for the AI 9 365. That is an 88.4% advantage. The R23 multi-core test shows a 73.9% lead, with scores of 32521 and 18698. These are the headline rendering numbers, and they reflect the core-count disparity: 16 cores versus 10, and 32 threads versus 20.

In single-core tests, the AI 9 365 takes narrow wins. Cinebench R15 single-core shows 303 for the AI 9 365 and 292 for the 8940HX, a 3.6% margin. Cinebench R23 single-core shows 1992 versus 1917, a 3.8% margin. These wins indicate that Zen 5's per-thread IPC is higher than Zen 4's in rendering workloads, even at a lower boost clock (5.00 GHz versus 5.30 GHz). The AI 9 365 achieves this despite a 0.30 GHz clock disadvantage, which points to architectural efficiency rather than clock speed.

Passmark data compression favors the 8940HX at 650984 versus 354510, a 83.6% win. Data encryption shows 39318 versus 18297, a 114.9% win, the largest margin in the entire comparison. Extended instructions score 47802 versus 25113, a 90.3% win. Find prime numbers yields 251 versus 117, a 114.5% win. Floating point math shows 113921 versus 62802, an 81.4% win. Integer math records 189221 versus 101831, an 85.8% win. Multithread score is 49731 versus 29467, a 68.8% win. Physics is the closest multi-threaded result: 2104 versus 1704, a 23.5% win. Random string sorting shows 75381 versus 39447, a 91.1% win.

The only Passmark test where the 8940HX does not win by a wide margin is the single-thread test, where it scores 3874 against 3841, a 0.9% win. This is a narrow victory, but it contradicts the Cinebench single-core results. The discrepancy suggests that the AI 9 365's single-core advantage is workload-specific, not a general property. In Passmark's single-thread workload, the 8940HX's higher boost clock compensates for the older architecture.

The average benchmark scores reflect the overall dominance: 81103 for the 8940HX versus 40048 for the AI 9 365. That is roughly a twofold difference, consistent with the 88.4% and 73.9% Cinebench multi-core margins. The 8940HX's nearest rivals, the Intel Xeon w5-3535X and Intel Core i9-14900KS, both score within 0% of it, which places it in desktop-class performance territory despite being a mobile part. The AI 9 365's nearest rival, the AMD Ryzen 7 7700, scores 40081, within 0.1% of it, indicating that the AI 9 365 performs like a mid-range desktop processor in aggregate despite its mobile positioning.

The 23.5% physics win for the 8940HX is notable because it is the smallest margin among its 13 wins. Physics workloads often depend on a mix of single-threaded and multi-threaded behavior, and the AI 9 365's superior single-core performance narrows the gap. However, the 8940HX still prevails.

The data encryption and find prime numbers tests show the largest gaps at 114.9% and 114.5% respectively. These are highly parallel workloads that scale nearly linearly with core count, so the 60% core advantage for the 8940HX translates into more than double the score in some cases. The 8940HX's 64 MB L3 cache likely helps in these workloads by reducing memory traffic stalls, compared to the AI 9 365's 16 MB L3.

For single-core rendering, the AI 9 365's wins are real but small. A 3.8% lead in Cinebench R23 single-core is meaningful for applications that are latency-bound, but it does not compensate for the multi-core deficit. The AI 9 365 also has higher memory bandwidth at 89.6 GB/s versus 83.2 GB/s, which may benefit certain memory-intensive single-threaded tasks, but the benchmark data does not show a corresponding advantage outside the Cinebench single-core tests.

In summary, the head-to-head data positions the 8940HX as a high-core-count performer for parallel workloads, with wins across every Passmark test and both Cinebench multi-core tests. The AI 9 365 is a lower-power part with a newer architecture that wins only the two Cinebench single-core tests, and by small margins. The 8940HX's average score is more than double that of the AI 9 365, and its percentile ranking at 95 versus 87 confirms a wide performance gap. The AI 9 365's redeeming feature is its power efficiency and single-thread rendering capability, but the recorded benchmarks show a processor that is decisively outmatched in overall throughput.