AMD Ryzen AI 9 HX 370 vs Intel Core Ultra 9 285T Comparison

Head-to-Head Benchmarks

The benchmark comparison between the AMD Ryzen AI 9 HX 370 and the Intel Core Ultra 9 285T reveals a clear split in workload types. Across the 15 shared tests, Intel wins 11 while AMD takes 4, but the margins tell a more nuanced story than the raw win count.

The most dramatic Intel victories come in single-threaded and lightly threaded workloads. In Cinebench R23 single-core, the Core Ultra 9 285T scores 4739 against the Ryzen AI 9 HX 370's 2018, a 57.4% advantage. That is the largest delta in the entire comparison. Cinebench R15 single-core follows a similar pattern: Intel's 477 versus AMD's 306.5, a 35.7% gap. The 3DMark single-thread result, while not part of the shared head-to-head set, reinforces this trend in the broader database: Intel posts 4576 in PassMark single-thread versus AMD's 3973, a 13.2% edge.

Multi-core results paint a different picture depending on the test. In Cinebench R23 multicore, Intel's 33573 crushes AMD's 21761, a 35.2% margin. That is a substantial lead for a 24-core processor against a 12-core part. However, Cinebench R15 multicore tells nearly the opposite story: AMD's 3409.5 edges out Intel's 3384 by just 0.8%. The R15 result suggests that at lower thread counts with shorter render times, the AMD part holds its own despite having half the cores.

PassMark multithread shows Intel ahead at 39931 versus 35148, a 12% gap. PassMark physics heavily favors Intel at 2842 versus 1881, a 33.8% difference, indicating stronger simulation and physics throughput. Floating-point math also goes decisively Intel's way: 137923 versus 76892, a 44.3% margin. Integer math is closer, with Intel's 132433 beating AMD's 122933 by 7.2%.

AMD's four wins are concentrated in specific data-processing tasks. PassMark data compression is AMD's best result: 445719 versus 384140, a 16% advantage. Extended instructions go AMD's way at 31653 versus 27477, a 15.2% lead. Random string sorting shows a narrow AMD win at 48723 versus 47695, only 2.2% ahead. The fourth win is the Cinebench R15 multicore result already noted.

Intel also dominates in prime number finding, scoring 345 versus AMD's 126, a 63.5% margin, the second-largest delta in the comparison. Data encryption goes to Intel at 32061 versus 22252, a 30.6% lead.

Where Each One Wins

The data indicates two distinct usage profiles. The Intel Core Ultra 9 285T is the stronger choice for compute-heavy, multi-threaded rendering, physics simulation, and floating-point workloads. Its Cinebench R23 multicore lead of 35.2% and PassMark physics advantage of 33.8% point to sustained all-core performance that benefits from the 24-core configuration. The 44.3% floating-point math margin reinforces this positioning for scientific computing and numerical analysis. Prime number finding, with a 63.5% lead, further cements Intel's role in integer-heavy algorithmic tasks.

The AMD Ryzen AI 9 HX 370 wins where data throughput and instruction efficiency matter more than raw core count. Its 16% data compression advantage suggests better memory subsystem efficiency for archive and database workloads. The 15.2% extended instructions lead indicates stronger SIMD or specialized instruction execution. The near-tie in Cinebench R15 multicore, despite AMD having half the cores, shows that the Zen 5 architecture delivers competitive short-burst multi-threaded performance at a much lower thread count. The 2.2% random string sorting win is narrow but consistent with AMD's pattern of winning memory-access-heavy tasks.

In single-threaded performance, Intel is the clear victor across every shared metric. The 57.4% Cinebench R23 single-core gap and 35.7% R15 single-core gap are too large to ignore. For users running lightly threaded applications such as legacy software, spreadsheet recalculation, or single-core CAD tools, the Intel part provides substantially more responsiveness.

FAQ

Q: Which processor has the higher average benchmark score?

A: The Intel Core Ultra 9 285T has an average benchmark score of 51310, placing it in the 91st percentile of all CPUs. The AMD Ryzen AI 9 HX 370 scores 37904 on average, which puts it in the 86th percentile.

Q: How do the two compare in Cinebench R23 multicore?

A: Intel wins decisively with a score of 33573 versus AMD's 21761. That is a 35.2% advantage for the Core Ultra 9 285T.

Q: Are there any benchmarks where the AMD processor wins?

A: Yes, AMD wins four of the 15 shared tests. The largest AMD victory is PassMark data compression at 445719 versus 384140, a 16% margin. AMD also wins extended instructions by 15.2%, random string sorting by 2.2%, and Cinebench R15 multicore by 0.8%.

Q: What is the single biggest performance gap between the two?

A: The largest delta is in Cinebench R23 single-core, where Intel's 4739 beats AMD's 2018 by 57.4%. PassMark find prime numbers is close behind, with Intel leading 345 versus 126, a 63.5% margin relative to AMD's score.

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 9 285T has 24 cores and 24 threads. The AMD Ryzen AI 9 HX 370 has 12 cores and 24 threads, meaning AMD uses simultaneous multithreading while Intel does not.

Q: How does the Intel processor compare to its own nearest rivals?

A: The Intel Core Ultra 9 285T's average score of 51310 sits 0.6% above the Intel Core i9-14900T (51015), 0.8% below the Intel Core i9-13900F (51730), 1.1% above the Intel Core i7-13850HX (50761), and 1.2% above the AMD Ryzen 9 5900XT (50718).

Specification Differences

The two processors differ substantially in core configuration. The AMD Ryzen AI 9 HX 370 packs 12 cores and 24 threads, while the Intel Core Ultra 9 285T uses 24 cores and 24 threads. Clock speeds also diverge: AMD has a 2.00 GHz base clock and 5.10 GHz boost clock, while Intel runs at 1.40 GHz base and 5.40 GHz boost. Thermal design power differs by 7 watts, with AMD rated at 28W and Intel at 35W.

Memory support shows a key split. The AMD part supports both DDR5 and LPDDR5X, while Intel supports only DDR5. Both use dual-channel memory buses, but Intel's bandwidth is higher at 102.4 GB/s versus AMD's 89.6 GB/s. ECC memory is supported on Intel but not on AMD.

PCIe generation and lane counts differ. Intel provides Gen 5 with 20 lanes, while AMD offers Gen 4 with 16 lanes. The integrated graphics also differ: AMD uses Radeon 890M, Intel uses Arc Xe-LPG Graphics 64EU.

Market positioning and physical specs vary. AMD is a mobile part on Socket FP8, while Intel is a desktop part on Socket 1851. AMD's die size is 233 mm², Intel's is 243 mm². Intel's transistor count is listed at 17,800 million; AMD's is not recorded. Intel has a launch MSRP of $549; AMD's launch MSRP is not listed. Neither processor has an unlocked multiplier.

Architecture Differences

The underlying architectures come from different design philosophies. AMD uses Zen 5 on a 4 nm TSMC process with the Strix Point codename, part of the Ryzen AI 300 generation that combines Zen 5 and Zen 5c cores. Intel uses Arrow Lake on a 3 nm TSMC process with the Arrow Lake-S codename, part of the Core Ultra Series 2. Both are made by TSMC, but Intel's process node is one step smaller.

Cache hierarchies differ significantly. AMD provides 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of shared L3. Intel provides 192 KB of L1 per core, 3 MB of L2 per core, and 36 MB of shared L3. Intel's larger L3 cache aligns with its higher memory bandwidth and 24-core design.

Feature sets also diverge. Intel supports ECC memory, which AMD does not. Intel's PCIe Gen 5 with 20 lanes offers newer connectivity compared to AMD's Gen 4 with 16 lanes. The release dates differ by roughly six months: AMD launched on 2024-06-30, Intel on 2025-01-06. Both are currently marked as Active in production status. Intel's part number is SRQD3, while AMD's is 100-000000994.