AMD Ryzen AI Max+ 392 vs Intel Core Ultra 9 290HX Plus Comparison

FAQ

Q: Which processor delivers higher single-thread performance?

A: The Intel Core Ultra 9 290HX Plus records a PassMark single-thread score of 4951, which is 20.7% ahead of the AMD Ryzen AI Max+ 392's score of 3927.

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

A: The Intel Core Ultra 9 290HX Plus scores 59439 in PassMark multithread, while the AMD Ryzen AI Max+ 392 scores 45231. The Intel part leads by 23.9% in this test.

Q: What is the core and thread configuration for each processor?

A: The AMD Ryzen AI Max+ 392 uses 12 cores and 24 threads, while the Intel Core Ultra 9 290HX Plus uses 24 cores and 24 threads. Both have the same 55 TDP rating.

Q: Which processor has the larger shared cache?

A: The AMD Ryzen AI Max+ 392 has 64 MB of shared L3 cache, compared to 36 MB on the Intel Core Ultra 9 290HX Plus.

Q: How does memory bandwidth differ between the two?

A: The AMD Ryzen AI Max+ 392 uses quad-channel LPDDR5X memory with 256.0 GB/s bandwidth. The Intel Core Ultra 9 290HX Plus uses dual-channel DDR5 memory with 102.4 GB/s bandwidth.

Q: What are the average benchmark scores and percentile rankings?

A: The AMD Ryzen AI Max+ 392 has an average benchmark score of 90541 and sits in the 96th percentile. The Intel Core Ultra 9 290HX Plus has an average score of 79574 and sits in the 95th percentile.

Architecture Differences

The two processors take fundamentally different approaches to mobile computing. The AMD Ryzen AI Max+ 392 uses the Zen 5 architecture under the Strix Halo codename, fabricated on a 4 nm process by TSMC. The Intel Core Ultra 9 290HX Plus belongs to the Core Ultra Series 2, uses the Arrow Lake-HX Refresh codename, and is built on a 3 nm process, also by TSMC. The process node difference gives Intel a slight density advantage, though the architectural designs diverge substantially.

Core counts differ significantly. AMD provides 12 cores with 24 threads, while Intel provides 24 cores with 24 threads. This means AMD relies on simultaneous multithreading to reach 24 threads, whereas Intel achieves 24 threads through physical cores alone. The Intel part has double the physical core count, which directly influences its strong showing in parallel workloads.

Cache hierarchies are distinct. AMD implements 80 KB L1 per core, 1 MB L2 per core, and 64 MB shared L3. Intel implements 192 KB L1 per core, 3 MB L2 per core, and 36 MB shared L3. AMD's larger L3 pool helps with data reuse across cores, while Intel's larger per-core L1 and L2 caches benefit individual thread performance.

Memory support diverges sharply. AMD uses LPDDR5X across a quad-channel bus, delivering 256.0 GB/s of memory bandwidth. Intel uses DDR5 across a dual-channel bus, delivering 102.4 GB/s. The AMD platform provides 2.5 times the raw memory bandwidth, which matters for integrated graphics and memory-intensive compute.

PCIe connectivity differs. AMD provides Gen 4 with 16 lanes, while Intel provides Gen 5 with 20 lanes. Intel's newer PCIe generation and higher lane count allow more expansion headroom for discrete GPUs and storage.

Integrated graphics also differ. AMD pairs with Radeon 8060S graphics, while Intel pairs with Arc Xe-LPG Graphics with 64 execution units. The AMD memory bandwidth advantage likely benefits its integrated GPU more heavily.

The die sizes reflect different integration strategies. AMD uses a dual-chip design with each die measuring 70.6 mm², totaling roughly 141.2 mm². Intel uses a single 243 mm² die with 17,800 million transistors. AMD's smaller dies may improve yield, while Intel's larger monolithic die concentrates more logic in one place.

Both processors support ECC memory and use a mobile socket, AMD on AMD Socket FP11 and Intel on Intel BGA 2114. The Intel part has an unlocked multiplier, while the AMD part does not. Both are listed as Active in production status.

Head-to-Head Benchmarks

The recorded data shows a clean sweep. The Intel Core Ultra 9 290HX Plus wins all 11 benchmark comparisons, with margins ranging from 7.5% to 50.7%. No single test favors the AMD Ryzen AI Max+ 392.

The largest gap appears in floating-point math. Intel scores 201773 against AMD's 99548, a 50.7% advantage. This indicates substantially higher throughput for scientific and engineering workloads that rely on FP arithmetic.

Data encryption shows a 44.4% gap. Intel scores 50008, AMD scores 27784. The Intel part handles cryptographic workloads with significantly more headroom.

Prime number finding favors Intel by 38.3%. Intel scores 519, AMD scores 320. This workload stresses integer throughput and branch prediction, where Intel's higher core count provides a decisive edge.

Random string sorting shows a 25.9% gap. Intel scores 80327, AMD scores 59487. Memory access patterns and cache efficiency drive this result, and Intel's larger L1 and L2 per core likely contribute.

Multi-threaded performance shows Intel ahead by 23.9%. Intel scores 59439, AMD scores 45231. With identical thread counts but differing core layouts, Intel's physical cores deliver more aggregate work.

Single-thread performance shows Intel ahead by 20.7%. Intel scores 4951, AMD scores 3927. This result appears twice in the data set under slightly different labels, confirming consistency.

Data compression favors Intel by 15.8%. Intel scores 658724, AMD scores 554760. The compression workload benefits from Intel's higher integer math score.

Physics simulation shows Intel ahead by 14.8%. Intel scores 3387, AMD scores 2887. This test often correlates with gaming physics and simulation workloads.

Extended instruction sets favor Intel by 11%. Intel scores 51290, AMD scores 45666. AVX and other extended instruction throughput is higher on the Intel part.

Integer math shows the narrowest gap at 7.5%. Intel scores 164839, AMD scores 152414. While Intel still wins, AMD's Zen 5 integer performance per core remains competitive.

The pattern is consistent. Intel wins every category, with the largest margins in floating-point, encryption, and prime number workloads. AMD's closest result comes in integer math, where the architecture's per-core efficiency narrows the gap.

Specification Differences

| Specification | AMD Ryzen AI Max+ 392 | Intel Core Ultra 9 290HX Plus |

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

| Cores | 12 | 24 |

| Threads | 24 | 24 |

| Base Clock | 3.20 GHz | 2.70 GHz |

| Boost Clock | 5.00 GHz | 5.50 GHz |

| Socket | AMD Socket FP11 | Intel BGA 2114 |

| Architecture | Zen 5 | Arrow Lake-HX Refresh |

| Codename | Strix Halo | Arrow Lake-HX Refresh |

| Generation | Ryzen AI Max (Zen 5, Strix Halo) | Ultra 9 (Arrow Lake-HX) |

| Process Node | 4 nm | 3 nm |

| Transistors | Not recorded | 17,800 million |

| Die Size | 2x 70.6 mm² | 243 mm² |

| L1 Cache | 80 KB per core | 192 KB per core |

| L2 Cache | 1 MB per core | 3 MB per core |

| L3 Cache | 64 MB shared | 36 MB shared |

| Memory Support | LPDDR5X | DDR5 |

| Memory Bus | Quad-channel | Dual-channel |

| Memory Bandwidth | 256.0 GB/s | 102.4 GB/s |

| PCIe | Gen 4, 16 Lanes | Gen 5, 20 Lanes |

| Integrated Graphics | Radeon 8060S | Arc Xe-LPG Graphics 64EU |

| Multiplier Unlocked | No | Yes |

| Release Date | 2026-01-05 | 2026-03-16 |

| Part Number | 100-000001979 | SADSS |

The specification table highlights the trade-offs. AMD offers higher base and boost clocks, much larger L3 cache, quad-channel memory with higher bandwidth, and a smaller dual-die package. Intel offers more cores, a newer process node, more per-core cache, PCIe Gen 5, an unlocked multiplier, and a later release date. Both parts share the same 55 TDP and ECC support.

Where Each One Wins

The Intel Core Ultra 9 290HX Plus wins every benchmark recorded in the head-to-head comparison. The data shows no test where AMD takes the lead. Intel's strongest categories are floating-point math, where it leads by 50.7%, and data encryption, where it leads by 44.4%. These workloads benefit from the 24 physical cores and the newer 3 nm process.

For integer math, the margin narrows to 7.5%, yet Intel still wins. The AMD Ryzen AI Max+ 392 does not win any measured category, so any use-case advantage must come from platform features rather than benchmark scores.

AMD's platform advantages appear in memory bandwidth and cache capacity. The quad-channel LPDDR5X configuration with 256.0 GB/s bandwidth is 2.5 times the Intel platform's rate. The 64 MB shared L3 cache is nearly double Intel's 36 MB. These characteristics matter for workloads that stream large data sets or require frequent cache hits across many threads.

The integrated graphics differ. AMD pairs with Radeon 8060S, while Intel pairs with Arc Xe-LPG Graphics with 64 execution units. The higher memory bandwidth on AMD's platform directly feeds the integrated GPU, which may benefit graphics and compute workloads that stay on the CPU package.

Intel's advantages include more cores, a newer process node, more per-core L1 and L2 cache, PCIe Gen 5 connectivity, and an unlocked multiplier. The 24 physical cores provide strong parallel throughput. The unlocked multiplier allows frequency tuning for users who adjust clock settings.

The average benchmark scores place AMD at 90541 versus Intel at 79574. AMD sits in the 96th percentile, Intel in the 95th. AMD's nearest rivals include the Intel Xeon 654 at 90717, just 0.2% ahead, and the AMD Ryzen 9 9955HX at 91199, 0.7% ahead. Intel's nearest rivals include the Intel Core i9-14900KF at 79371, 0.3% behind, and the AMD EPYC 7413 at 80041, 0.6% ahead.

For workloads that stress floating-point throughput, encryption, or prime calculation, the Intel part is the stronger choice based on recorded data. For workloads that rely on memory bandwidth, large shared cache, or integrated graphics with high-bandwidth memory, the AMD platform offers structural advantages even though the benchmark sweep favors Intel.