AMD Ryzen AI Max+ 392 vs Intel Core Ultra 9 285H Comparison

Where Each One Wins

The AMD Ryzen AI Max+ 392 and Intel Core Ultra 9 285H split their benchmark victories along clear lines of workload character. AMD takes 7 of the 11 recorded head-to-head tests, while Intel secures 4. The pattern is consistent: AMD dominates in heavily parallelized, throughput-oriented tasks, while Intel wins in lighter, latency-sensitive single-thread workloads.

AMD’s largest wins come in integer math, where it scores 152414 against Intel’s 85922, a 77.4% advantage. That margin is the biggest in the entire comparison. Data compression also heavily favors AMD: 554760 versus 335859, a 65.2% gap. Extended instruction workloads show a 70.4% lead (45666 against 26794). These are compute-heavy operations that scale with thread count and memory bandwidth, both of which AMD supplies in abundance.

Intel’s wins are narrower but meaningful. The Core Ultra 9 285H leads in single-thread performance by 11.1%, scoring 4415 against AMD’s 3927. Floating-point math goes Intel’s way by 8.8% (109190 versus 99548). The prime number test shows a slim 3% edge for Intel (330 versus 320). These results reflect Intel’s higher boost clock and stronger per-core efficiency in workloads that resist parallel scaling.

The multithread test shows AMD ahead by 32.4% (45231 versus 34171), which aligns with its 24 threads versus Intel’s 16. Physics simulation also favors AMD by 14.9% (2887 versus 2513). Random string sorting goes to AMD by 45.3% (59487 versus 40931), a task that benefits from large cache capacity. Data encryption is nearly even, with AMD ahead by just 6.3% (27784 versus 26140).

Overall benchmark averages reflect the split. The Ryzen AI Max+ 392 sits at the 96th percentile of all CPUs with an average score of 90541. The Core Ultra 9 285H lands at the 86th percentile with an average of 38312. That gap is large, but the percentile difference shows both are high-end mobile parts; AMD simply occupies a higher performance tier.

Architecture Differences

The two processors come from different design philosophies. AMD uses a monolithic Zen 5 architecture on a 4 nm TSMC process, codenamed Strix Halo. Intel uses Arrow Lake-H on a 3 nm TSMC process, part of the Core Ultra Series 2. Both are mobile parts, but their internal layouts diverge sharply.

AMD configures 12 cores and 24 threads with a base clock of 3.20 GHz and a boost clock of 5.00 GHz. Intel offers 16 cores and 16 threads, with a base clock of 2.90 GHz and a boost of 5.40 GHz. Intel’s higher boost clock explains its single-thread wins. AMD’s thread advantage explains its multithread dominance.

Cache hierarchies differ considerably. AMD allocates 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of shared L3. Intel provides 192 KB of L1 per core, 3 MB of L2 per core, and only 24 MB of shared L3. AMD’s 64 MB L3 is a major asset for data-heavy workloads like compression and sorting. Intel’s larger per-core L1 and L2 help with latency-sensitive tasks but do not compensate for the L3 deficit in throughput tests.

Memory support also diverges. AMD uses LPDDR5X over a quad-channel bus with 256.0 GB/s of bandwidth. Intel supports both DDR5 and LPDDR5X over a dual-channel bus with 102.4 GB/s. AMD’s bandwidth advantage is exactly 2.5 times Intel’s, which directly feeds its leads in integer math and compression. Both support ECC memory, a notable feature for mobile workstation chips.

PCIe connectivity differs. AMD provides Gen 4 with 16 lanes, while Intel provides Gen 5 with 8 lanes. Intel’s newer standard offers higher per-lane throughput, but AMD’s greater lane count suits devices that need many simultaneous connections. The integrated graphics also differ: AMD ships the Radeon 8060S, while Intel ships Arc Graphics 140T.

The sockets and packaging are distinct. AMD uses AMD Socket FP11, Intel uses Intel BGA 2049. Neither supports overclocking, as both have locked multipliers. Release dates differ by roughly a year: Intel launched on 2026-01-05, while AMD launched on 2026-01-05 in the database timeline, but the recorded release for Intel is 2025-01-12. The production status for both is listed as Active.

Head-to-Head Benchmarks

The recorded head-to-head data reveals where each processor decisively outperforms the other. The largest single margin is AMD’s 77.4% lead in integer math. That test measures raw arithmetic throughput, and AMD’s 152414 score against Intel’s 85922 shows a fundamental advantage in parallel integer execution. This alone positions the Ryzen AI Max+ 392 as the stronger choice for compilation, simulation, and other integer-heavy workloads.

Extended instructions show a 70.4% AMD advantage (45666 versus 26794). This test covers specialized instruction sets such as AES and AVX, where AMD’s larger cache and higher memory bandwidth amplify performance. Data compression follows at 65.2% (554760 versus 335859). Compression algorithms are notoriously bandwidth-hungry, and AMD’s 256.0 GB/s versus Intel’s 102.4 GB/s explains the gap.

Random string sorting gives AMD a 45.3% edge (59487 versus 40931). Sorting touches large data sets repeatedly, and AMD’s 64 MB L3 cache keeps more working data on-die. The multithread test shows a 32.4% AMD lead (45231 versus 34171), consistent with its 24-thread count. Physics simulation favors AMD by 14.9% (2887 versus 2513), a workload that benefits from both threading and cache.

Data encryption is the closest contest. AMD wins by just 6.3% (27784 versus 26140). Both processors handle encryption well, but AMD’s margin comes from its thread advantage rather than any architectural superiority in cryptographic instructions.

Intel’s wins are concentrated in single-thread and floating-point tests. The single-thread test shows Intel ahead by 11.1% (4415 versus 3927). This reflects Intel’s 5.40 GHz boost clock against AMD’s 5.00 GHz. Floating-point math gives Intel an 8.8% edge (109190 versus 99548). Prime number finding is nearly tied, with Intel ahead by 3% (330 versus 320). These results indicate that Intel retains a per-core efficiency advantage, but it only matters when workloads cannot use AMD’s extra threads.

The Cinebench and Geekbench results for Intel provide additional context, though no direct AMD equivalents exist in the database. Intel scores 20781.5 in Cinebench R23 multicore and 2129.5 in single-core. Geekbench shows 14743 multicore and 2178 single-core. These are strong numbers for a 45 W mobile part, but the PassMark comparisons remain the primary basis for head-to-head analysis.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen AI Max+ 392 has 12 cores and 24 threads. The Intel Core Ultra 9 285H has 16 cores and 16 threads. AMD has more threads, while Intel has more physical cores.

Q: Why does AMD win the multithread test by 32.4%?

A: AMD’s 24 threads versus Intel’s 16 threads, combined with a 64 MB shared L3 cache and 256.0 GB/s memory bandwidth, allows it to sustain higher throughput in parallel workloads. Intel’s higher boost clock cannot compensate for the thread deficit.

Q: Why does Intel win the single-thread test by 11.1%?

A: Intel’s boost clock reaches 5.40 GHz, above AMD’s 5.00 GHz. Additionally, Intel provides 192 KB of L1 and 3 MB of L2 per core, which reduces latency for single-threaded code. The recorded scores are 4415 for Intel and 3927 for AMD.

Q: What are the memory bandwidth differences?

A: AMD uses quad-channel LPDDR5X with 256.0 GB/s. Intel uses dual-channel DDR5 or LPDDR5X with 102.4 GB/s. AMD’s bandwidth is exactly 2.5 times higher, which drives its leads in compression, sorting, and integer math.

Q: Do both processors support ECC memory?

A: Yes. Both the AMD Ryzen AI Max+ 392 and the Intel Core Ultra 9 285H list ECC memory support in the database.

Q: How do their overall benchmark averages compare?

A: AMD has an average benchmark score of 90541 and sits at the 96th percentile of all CPUs. Intel has an average of 38312 and sits at the 86th percentile. AMD’s nearest rivals include the Intel Xeon 654 at a 0.2% higher score and the AMD Ryzen 9 9955HX at 0.7% higher.

Specification Differences

| Specification | AMD Ryzen AI Max+ 392 | Intel Core Ultra 9 285H |

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

| Cores | 12 | 16 |

| Threads | 24 | 16 |

| Base clock | 3.20 GHz | 2.90 GHz |

| Boost clock | 5.00 GHz | 5.40 GHz |

| TDP | 55 W | 45 W |

| Socket | AMD Socket FP11 | Intel BGA 2049 |

| Architecture | Zen 5 | Arrow Lake |

| Codename | Strix Halo | Arrow Lake-H |

| Process node | 4 nm | 3 nm |

| 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) | 24 MB (shared) |

| Memory support | LPDDR5X | DDR5, LPDDR5X |

| Memory bus | Quad-channel | Dual-channel |

| Memory bandwidth | 256.0 GB/s | 102.4 GB/s |

| PCIe | Gen 4, 16 lanes | Gen 5, 8 lanes |

| Integrated graphics | Radeon 8060S | Arc Graphics 140T |

| Part number | 100-000001979 | SRQAL |

| Release date | 2026-01-05 | 2025-01-12 |

| Launch MSRP | None listed | $651 |