AMD Ryzen 9 8945HX vs Intel Core Ultra 9 290HX Plus Comparison

Head-to-Head Benchmarks

The benchmark data splits this comparison into two distinct performance profiles. The Intel Core Ultra 9 290HX Plus wins 12 of the 17 recorded head-to-head tests, while the AMD Ryzen 9 8945HX takes 5. The margins, however, tell a more nuanced story than the raw win count suggests.

Starting with multi-threaded Cinebench results, the Intel part dominates the older R15 and R20 versions. In Cinebench R15 multi-core, the Intel scores 5981 against AMD's 4305, a 38.9% advantage. The R20 multi-core test shows Intel ahead at 21198 versus 17939, an 18.2% margin. These are substantial leads in rendering workloads that scale across cores. The R23 multi-core test flips the result, with AMD posting 42713 against Intel's 39684, a 7.1% deficit for Intel. This inconsistency across Cinebench versions suggests the AMD chip responds better to the newer R23 workload's specific instruction mix or thermal profile.

Single-core results present an equally split picture. The R15 single-core test heavily favors AMD: 607 versus 340, a 44% margin for AMD. The R20 single-core test reverses this, with Intel at 2992 versus AMD's 2532, an 18.2% lead. Then R23 single-core swings back decisively to AMD: 6030 versus 2356, a 60.9% advantage. No other workload in the pack shows such dramatic swings between the two processors. The R23 single-core gap is the largest percentage difference recorded in the entire head-to-head set, and it favors AMD by a wide margin.

PassMark workloads further differentiate the two. The Intel chip wins integer-heavy and floating-point tasks with large margins. Floating point math shows Intel at 201773 versus AMD's 117453, a 71.8% lead. Physics tests show Intel at 3387 versus 2168, a 56.2% edge. Prime number finding gives Intel the biggest relative win: 519 versus 262, a 98.1% margin, nearly double AMD's output. Data encryption also favors Intel: 50008 versus 40836, a 22.5% lead. Extended instructions go narrowly to Intel, 51290 versus 49823, a 2.9% margin, while random string sorting is close at 80327 versus 78781, a 2% difference.

AMD's PassMark wins come in integer math and data compression. Integer math shows AMD at 195180 versus Intel's 164839, a 15.5% advantage. Data compression goes to AMD at 677755 versus 658724, a 2.8% edge. The multithread PassMark test favors Intel: 59439 versus 51405, a 15.6% lead. Single-thread PassMark also goes to Intel at 4951 versus 3907, a 26.7% margin.

The aggregate benchmark scores place Intel at 79574 average versus AMD's 76212, a difference of roughly 4.4%. Both chips sit at the 95th percentile among all CPUs in the database. Intel's nearest rivals include the Core i9-14900KF at 79371 (0.3% behind Intel), the Core i9-14900K at 79097 (0.6% behind), the AMD EPYC 7413 at 80041 (0.6% ahead), and the Xeon w5-2565X at 80671 (1.4% ahead). AMD's nearest rivals cluster tighter: the Core Ultra 9 285HX at 76155 (0.1% behind AMD), the EPYC Embedded 8224P at 76492 (0.4% ahead), the Threadripper PRO 9945WX at 76513 (0.4% ahead), and the Ryzen 9 9950X3D at 75779 (0.6% behind). AMD sits within a narrow band of comparable parts, while Intel's closest competitors span a slightly wider range.

Architecture Differences

The two processors come from fundamentally different design philosophies. Intel's Core Ultra 9 290HX Plus uses the Arrow Lake-HX Refresh codename, part of the Core Ultra Series 2 generation. AMD's Ryzen 9 8945HX uses the Dragon Range codename, built on the Zen 4 architecture. Both are mobile parts with active production status, but the underlying silicon differs sharply.

Process node technology separates them clearly. Intel fabricates on a 3 nm process at TSMC, while AMD uses a 5 nm process, also at TSMC. The Intel die measures 243 mm² with 17,800 million transistors. AMD uses a dual-die configuration with 2x 71 mm² dies, totaling 13,140 million transistors. The smaller process node gives Intel a density advantage, while AMD's dual-die approach allows for a different cache topology.

Core counts and threading models are a major architectural divergence. Intel provides 24 cores and 24 threads, meaning no simultaneous multithreading. AMD provides 16 cores and 32 threads, using simultaneous multithreading to double its thread count. This explains why AMD wins some heavily threaded workloads despite fewer physical cores: its 32 threads can fill more execution slots in certain scheduler-friendly tasks. Intel's 24 cores with no SMT rely on raw core count and higher per-core efficiency.

Cache hierarchies differ substantially. Intel uses 192 KB of L1 per core, 3 MB of L2 per core, and 36 MB of shared L3. AMD uses 64 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. The AMD chip holds nearly twice the shared L3 cache, which benefits workloads with large working sets. Intel's larger per-core L1 and L2 caches give each core more local fast storage, which can help latency-sensitive single-threaded tasks.

Clock speeds are close but not identical. Intel lists a 2.70 GHz base clock and 5.50 GHz boost clock. AMD lists a 2.50 GHz base clock and 5.40 GHz boost clock. Intel holds a slight edge in both figures, but the R23 single-core results show AMD pulling far ahead, indicating that raw clock speed alone does not determine single-thread performance in this comparison.

Memory support matches on type: both use DDR5 with dual-channel memory buses. Bandwidth differs, with Intel at 102.4 GB/s and AMD at 83.2 GB/s, a 23% advantage for Intel. ECC memory support also differs: Intel supports ECC, AMD does not. PCIe lane allocation shows AMD with more CPU-attached lanes: 28 Gen 5 lanes versus Intel's 20 Gen 5 lanes.

Integrated graphics differ as well. Intel pairs an Arc Xe-LPG Graphics 64EU solution, while AMD includes Radeon 610M. Both are integrated options for mobile platforms, but the database records no graphics benchmarks for either.

Release dates place Intel later: Intel's release date is 2026-03-16, while AMD's is 2025-04-22. Both are listed as active production parts. Neither has a recorded launch MSRP in the database.

FAQ

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

A: The Intel Core Ultra 9 290HX Plus wins 12 of the 17 recorded tests. The AMD Ryzen 9 8945HX wins 5.

Q: Where does the AMD processor show its largest advantage?

A: AMD's biggest margin is in Cinebench R23 single-core, where it scores 6030 versus Intel's 2356, a 60.9% lead. It also wins R15 single-core by 44% and integer math by 15.5%.

Q: What is the Intel processor's strongest test result?

A: Intel's largest relative win is in PassMark find prime numbers, scoring 519 versus AMD's 262, a 98.1% margin. It also leads floating point math by 71.8% and physics by 56.2%.

Q: How do the core and thread counts compare?

A: Intel has 24 cores and 24 threads, with no simultaneous multithreading. AMD has 16 cores and 32 threads, using simultaneous multithreading to reach 32 threads.

Q: Which processor has more L3 cache?

A: AMD has 64 MB of shared L3 cache. Intel has 36 MB of shared L3 cache. Intel compensates with larger per-core L1 (192 KB versus 64 KB) and L2 (3 MB versus 1 MB) caches.

Q: Do both processors support ECC memory?

A: No. Intel supports ECC memory, while AMD does not. Both support DDR5 with dual-channel memory buses.

Specification Differences

The two processors differ across every major specification field in the database.

| Specification | Intel Core Ultra 9 290HX Plus | AMD Ryzen 9 8945HX |

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

| Series | Core Ultra Series 2 | 8000 series |

| Cores | 24 | 16 |

| Threads | 24 | 32 |

| Base Clock | 2.70 GHz | 2.50 GHz |

| Boost Clock | 5.50 GHz | 5.40 GHz |

| Socket | Intel BGA 2114 | AMD Socket FL1 |

| Architecture | (not listed) | Zen 4 |

| Codename | Arrow Lake-HX Refresh | Dragon Range |

| Process Node | 3 nm | 5 nm |

| Transistors | 17,800 million | 13,140 million |

| Die Size | 243 mm² | 2x 71 mm² |

| L1 Cache | 192 KB (per core) | 64 KB (per core) |

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

| L3 Cache | 36 MB (shared) | 64 MB |

| Memory Bandwidth | 102.4 GB/s | 83.2 GB/s |

| ECC Memory | Yes | No |

| PCIe Lanes | Gen 5, 20 Lanes (CPU only) | Gen 5, 28 Lanes (CPU only) |

| Integrated Graphics | Arc Xe-LPG Graphics 64EU | Radeon 610M |

| Release Date | 2026-03-16 | 2025-04-22 |

Both processors share a 55 W TDP, dual-channel DDR5 memory support, mobile market segment, active production status, and unlocked multiplier. Neither has a recorded launch MSRP.

The Verdict

The data points to distinct use cases for each processor. The Intel Core Ultra 9 290HX Plus is the stronger all-around performer by aggregate score, with an average benchmark score of 79574 versus AMD's 76212. It wins the majority of head-to-head tests and holds decisive leads in floating point math, physics, encryption, and prime number workloads. Its higher memory bandwidth (102.4 GB/s versus 83.2 GB/s), ECC support, and 3 nm process node give it structural advantages for compute-heavy mobile tasks.

The AMD Ryzen 9 8945HX excels in specific workloads, particularly Cinebench R23 multi-core and single-core, where it beats Intel by 7.1% and 60.9% respectively. Its 32 threads from 16 cores with SMT, combined with 64 MB of L3 cache, make it effective for threaded rendering tasks and integer math operations. AMD also wins data compression and integer math tests, suggesting strength in productivity and content-creation workloads that rely on those patterns.

For users who prioritize Cinebench-style rendering and single-threaded responsiveness, the recorded data favors AMD. For users who need broader compute coverage across physics, floating point, encryption, and prime number workloads, Intel has the edge. The 95th percentile ranking for both chips means either sits among the top mobile processors in the database.

The verdict from the numbers: Intel offers the higher aggregate performance and wider benchmark coverage, while AMD offers superior results in specific rendering and integer workloads. The choice depends on whether the target software aligns with AMD's strengths or requires Intel's broader compute lead. Neither chip dominates across the full test suite, and the 4.4% aggregate gap in Intel's favor is meaningful but not overwhelming.