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

Head-to-Head Benchmarks

The benchmark data tells a one-sided story in the direct comparison between the AMD Ryzen AI Max+ 392 and the Intel Core Ultra 9 290K Plus. Across the eleven Passmark tests recorded in the database, the Intel part wins every single contest. The closest margin comes in integer math, where the Intel chip scores 166194 against AMD's 152414, a lead of 8.3 percent. That is the only head-to-head result in single digits. Everywhere else, the gap widens substantially.

The most lopsided defeat for AMD is in floating point math. The Intel Core Ultra 9 290K Plus scores 214760, while the Ryzen AI Max+ 392 manages 99548. That works out to a 53.6 percent deficit for the AMD processor. The data encryption test is nearly as brutal: Intel posts 52563 against AMD's 27784, a 47.1 percent difference. These are not small edges; they are generational gaps in raw throughput.

Multi-threaded workloads also favor Intel heavily. The Passmark multithread score for the Core Ultra 9 290K Plus is 60860, versus 45231 for the Ryzen AI Max+ 392, a 25.7 percent advantage. Prime number finding shows a 36.4 percent Intel lead (503 versus 320), and random string sorting lands at 79744 against 59487, a 25.4 percent gap. Data compression is similarly one-sided: 698346 for Intel, 554760 for AMD, a 20.6 percent difference.

Single-thread performance is also an Intel win, though not by the same magnitude as the math tests. The Core Ultra 9 290K Plus scores 4823 in the Passmark single-thread test, against 3927 for the Ryzen AI Max+ 392, a 18.6 percent edge. Extended instructions show a 12.7 percent Intel advantage (52338 versus 45666), and the physics test rounds out the sweep with Intel ahead by 12.9 percent (3315 versus 2887).

The aggregate numbers reflect this dominance. The AMD part carries an average benchmark score of 90541, while the Intel part sits at 84003. That seems contradictory given the head-to-head sweep, but the average includes different test sets. The AMD Ryzen AI Max+ 392's nearest rivals in the database are the Intel Xeon 654 (0.2 percent lower average score), the AMD Ryzen 9 9955HX (0.7 percent lower), the Intel Xeon w7-2575X (2.7 percent higher), and the Intel Xeon 6736P (3 percent higher). The Intel Core Ultra 9 290K Plus, by contrast, sits within 1.3 percent of a cluster of EPYC and Core Ultra parts: the Core Ultra 9 285K is 0.2 percent lower, the EPYC 4584PX is 1.1 percent lower, the EPYC 9135 is 1.2 percent lower, and the EPYC 7F72 is 1.3 percent higher.

Both CPUs rank in the 96th percentile of all processors in the database. That means neither is a slouch in absolute terms. But in a direct matchup, the Intel chip is consistently faster, and often by a wide margin. The only question is whether the AMD part's feature set compensates for those losses in specific use cases.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI Max+ 392 is a mobile part on the Zen 5 architecture, codenamed Strix Halo, built on TSMC's 4 nm process. It uses a 12-core, 24-thread configuration with a base clock of 3.20 GHz and a boost clock of 5.00 GHz. The Intel Core Ultra 9 290K Plus is a desktop part from the Arrow Lake Refresh generation, built on a 3 nm TSMC process. It packs 24 cores and 24 threads, with a base clock of 3.70 GHz and a boost clock of 5.80 GHz. Note that Intel's core count doubles AMD's, but thread count is identical, which means the Intel part uses only one thread per core while AMD uses two.

Cache hierarchies differ sharply. AMD allocates 80 KB of L1 per core, 1 MB of L2 per core, and a large 64 MB shared L3 pool. Intel allocates 192 KB of L1 per core, 3 MB of L2 per core, and 36 MB of shared L3. Intel's per-core cache is larger, but AMD's total L3 is nearly double Intel's. That larger shared pool can benefit workloads that repeatedly access a working set that exceeds Intel's 36 MB.

Memory architecture is a major differentiator. AMD supports LPDDR5X over a quad-channel bus, delivering 256.0 GB/s of bandwidth. Intel supports DDR5 over a dual-channel bus, delivering 115.2 GB/s. That is a 2.2x bandwidth advantage for AMD, which likely explains why the gap in integer math is relatively small despite Intel's core count advantage. Both parts support ECC memory, which is unusual for consumer-oriented chips and points at workstation or server-adjacent use cases.

The process nodes are both TSMC, but different generations: AMD uses 4 nm, Intel uses 3 nm. Intel's die is larger at 243 mm² with 17,800 million transistors, while AMD uses a dual-die design with each die measuring 70.6 mm². AMD does not list a transistor count in the database. PCIe support also differs: AMD provides Gen 4 with 16 lanes (CPU only), while Intel provides Gen 5 with 20 lanes (CPU only). Intel's PCIe is newer and has more lanes, which matters for expansion and GPU bandwidth.

Integrated graphics are a notable split. AMD pairs the Ryzen AI Max+ 392 with a Radeon 8060S, while Intel uses Arc Xe-LPG Graphics with 64 execution units. The database does not include graphics benchmarks, so no direct comparison is possible from the recorded data. But the presence of a Radeon 8060S on a mobile part suggests AMD is targeting systems that may rely heavily on the iGPU.

The market segments differ: AMD is a mobile processor on Socket FP11, Intel is a desktop processor on Socket 1851. Intel also has an unlocked multiplier, while AMD does not. That means the Intel part can be overclocked, assuming the rest of the platform allows it. The TDP figures reflect the segment difference: AMD draws 55 W, Intel draws 125 W. That is a 2.3x thermal envelope for Intel, which helps explain its higher clock speeds and performance.

Where Each One Wins

The Intel Core Ultra 9 290K Plus wins every benchmark in the head-to-head set, so the straightforward answer is that it wins everywhere in the recorded tests. But the magnitude of those wins matters for real-world use. The 53.6 percent lead in floating point math and the 47.1 percent lead in data encryption point to workloads that are heavily dependent on raw compute throughput, such as scientific simulation, financial modeling, and cryptographic operations. The 25.7 percent lead in multithread and 36.4 percent lead in prime number finding reinforce that picture: the Intel chip is the stronger general-purpose compute engine.

The AMD Ryzen AI Max+ 392 still has clear advantages that are not captured in the Passmark suite. Its 256.0 GB/s memory bandwidth is more than double Intel's 115.2 GB/s, which suggests that memory-bandwidth-bound workloads, such as large in-memory databases or certain data analytics tasks, could close or even reverse the gap. The 64 MB L3 cache is also nearly double Intel's 36 MB, which can help with cache-resident datasets. And the 55 W TDP means AMD achieves its performance at less than half the power draw of Intel's 125 W part, which matters for laptops, compact systems, or any environment where thermals are constrained.

The release timing also matters. AMD's part has a recorded release date of January 5, 2026, while Intel's release date is not recorded in the database. That absence is notable, but not interpretable beyond stating that the database lacks that information. Both parts are marked as Active in production status, so neither is discontinued.

For workloads that fit within the Passmark suite, the choice is clear: Intel wins. For workloads that depend on memory bandwidth, cache capacity, or power efficiency, the AMD part has structural advantages that the benchmarks do not measure. The database's percentile ranking puts both at 96, meaning both are near the top of the overall pool, but the Intel part does it with more cores and higher clocks, while AMD does it with a more efficient design and faster memory subsystem.

FAQ

Q: Which processor wins more head-to-head benchmark tests?

A: The Intel Core Ultra 9 290K Plus wins all 11 head-to-head Passmark tests recorded in the database. The AMD Ryzen AI Max+ 392 does not win any.

Q: What is the largest performance gap between the two?

A: The largest gap is in floating point math, where Intel scores 214760 versus AMD's 99548, a 53.6 percent difference. The smallest gap is in integer math, where Intel leads by 8.3 percent (166194 versus 152414).

Q: How do their core and thread counts compare?

A: The AMD Ryzen AI Max+ 392 has 12 cores and 24 threads. The Intel Core Ultra 9 290K Plus has 24 cores and 24 threads. Intel has twice the cores but the same thread count, meaning AMD uses simultaneous multithreading while Intel does not.

Q: Which processor has more memory bandwidth?

A: The AMD Ryzen AI Max+ 392 supports LPDDR5X over a quad-channel bus with 256.0 GB/s bandwidth. The Intel Core Ultra 9 290K Plus supports DDR5 over a dual-channel bus with 115.2 GB/s. AMD offers more than double the bandwidth.

Q: Are both processors in the same performance percentile?

A: Yes, both are in the 96th percentile of all CPUs in the database. However, their average benchmark scores differ: AMD averages 90541, while Intel averages 84003.

Q: Do both processors support ECC memory?

A: Yes, both the AMD Ryzen AI Max+ 392 and the Intel Core Ultra 9 290K Plus support ECC memory, according to the database.

Specification Differences

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

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

| Series | Not listed | Core Ultra Series 2 |

| Cores | 12 | 24 |

| Threads | 24 | 24 |

| Base Clock | 3.20 GHz | 3.70 GHz |

| Boost Clock | 5.00 GHz | 5.80 GHz |

| TDP | 55 W | 125 W |

| Socket | AMD Socket FP11 | Intel Socket 1851 |

| Codename | Strix Halo | Arrow Lake Refresh |

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

| Process Node | 4 nm | 3 nm |

| Foundry | TSMC | TSMC |

| Transistors | Not listed | 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 | 115.2 GB/s |

| PCIe | Gen 4, 16 lanes (CPU only) | Gen 5, 20 lanes (CPU only) |

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

| Market Segment | Mobile | Desktop |

| Release Date | 2026-01-05 | Not listed |

| Multiplier Unlocked | No | Yes |

| Part Number | 100-000001979 | Unknown |