AMD Ryzen AI Embedded P164 vs Intel Core Ultra 7 265H Comparison

Head-to-Head Benchmarks

The head-to-head data presents a strikingly one-sided picture. The Intel Core Ultra 7 265H claims victory in 10 of the 11 recorded comparisons, while the AMD Ryzen AI Embedded P164 manages a single win. That solitary AMD success comes in integer math, where it scores 87,940 against Intel's 85,479, a 2.9% margin. It is a narrow result, but it is the only category where the AMD part leads.

The scale of Intel's dominance varies dramatically across workloads. The most extreme gap appears in prime number finding, where Intel scores 335 versus AMD's 71, a 78.8% deficit for the AMD chip. This is not a subtle difference; the Intel part delivers roughly 4.7 times the throughput in this specific test. Floating point math tells a similar story, with Intel at 109,123 against AMD's 55,799, a 48.9% shortfall. Physics results follow the same pattern, Intel's 2,497 nearly doubles AMD's 1,210, a 51.5% lead. These are the categories where the Intel processor's wider execution resources become most visible.

Data encryption shows a 38.3% margin in Intel's favor, with scores of 26,005 and 16,055 respectively. The multithread score, which often serves as a general productivity indicator, gives Intel 34,027 against 25,889, a 23.9% advantage. Random string sorting runs 14.6% faster on Intel, and extended instructions are 9.7% ahead. Even in single-threaded work, Intel leads by 7% with 4,334 versus 4,029. Data compression is the closest contest, Intel's 334,711 barely edges AMD's 327,891, a 2% gap. Across the full set, the pattern is consistent: Intel wins most workloads, often by wide margins, while AMD holds only a slim integer math lead.

The average benchmark scores in the database reinforce the same ordering, though with a different magnitude. AMD's average sits at 52,901, placing it in the 91st percentile of all CPUs. Intel's average is 41,621, which falls into the 88th percentile. The nearest rivals listed for AMD include the AMD Ryzen 5 9500F at 52,873 (0.1% behind) and the Intel Xeon 634 at 52,974 (0.1% ahead). Intel's nearest rivals include the Intel Core 7 251TE at 41,650 (0.1% ahead) and the Intel Core i7-14650HX at 41,576 (0.1% behind). The head-to-head wins do not translate directly into a higher overall average, which suggests the two processors are optimized for different workload distributions.

Architecture Differences

The architectural gap between these two mobile processors is substantial. AMD's Ryzen AI Embedded P164 uses the Gorgon Point codename, built on the Zen 5 / Zen 5c generation, and is fabricated on a 4 nm process at TSMC. Intel's Core Ultra 7 265H belongs to the Core Ultra Series 2, uses the Arrow Lake architecture, specifically Arrow Lake-H, and is built on a 3 nm process, also at TSMC. The process node difference is modest, but the core configurations diverge sharply.

AMD provides 8 cores and 16 threads, while Intel provides 16 cores and 16 threads. This is a critical distinction: Intel has twice the physical core count but no hyperthreading, so thread counts are equal. The AMD part relies on simultaneous multithreading to reach 16 threads, while Intel achieves 16 threads purely through core count. This explains why Intel wins so decisively in floating point and physics workloads, where more physical cores typically provide better scaling. AMD's per-core L1 cache is 80 KB, and its L2 cache is 1 MB per core. Intel's L1 is 192 KB per core, and its L2 is 3 MB per core. The L3 cache also differs: AMD has 8 MB total, while Intel has 24 MB shared. Intel's larger cache hierarchy, combined with more cores, gives it a structural advantage in cache-sensitive workloads.

Clock speeds are close on paper. AMD's base clock is 2.00 GHz with a 5.00 GHz boost, while Intel's base is 2.20 GHz with a 5.30 GHz boost. Both parts are locked, with multipliers not unlocked. The process node difference, 4 nm versus 3 nm, may contribute to Intel's slightly higher clocks, but the recorded data does not isolate that effect. Both processors support DDR5 and LPDDR5X memory over a dual-channel bus. AMD's memory bandwidth is listed at 89.6 GB/s, while Intel reaches 102.4 GB/s. Both support ECC memory, which matters for embedded and reliability-focused deployments.

PCIe connectivity differs in generation and lane count. AMD uses PCIe Gen 4 with 16 lanes from the CPU, while Intel uses PCIe Gen 5 with 8 lanes. The Intel part offers a newer generation but fewer lanes, a trade-off that may affect expansion options. Integrated graphics also diverge: AMD uses the Radeon 880M, while Intel uses Arc Graphics 140T. The die size is recorded only for AMD at 233 mm²; Intel's die size is not listed. The production status for both is active, and neither has a listed launch MSRP.

FAQ

Q: Which processor wins the majority of head-to-head benchmark comparisons?

A: The Intel Core Ultra 7 265H wins 10 of 11 recorded comparisons. The AMD Ryzen AI Embedded P164 wins only one, integer math, by 2.9%.

Q: How large is the largest performance gap between the two?

A: The largest gap is in prime number finding, where Intel scores 335 versus AMD's 71, a 78.8% deficit for the AMD part. Floating point math is the second largest at 48.9% in Intel's favor.

Q: Do both processors have the same thread count despite different core counts?

A: Yes. AMD has 8 cores and 16 threads, while Intel has 16 cores and 16 threads. Intel reaches 16 threads through physical cores alone, while AMD uses simultaneous multithreading.

Q: What is the memory bandwidth difference?

A: Intel lists 102.4 GB/s, while AMD lists 89.6 GB/s. Both use dual-channel DDR5 or LPDDR5X memory.

Q: Which processor has a higher single-thread score?

A: Intel leads with 4,334 against AMD's 4,029, a 7% margin in the PassMark single-thread test.

Q: How do the overall average benchmark scores compare?

A: AMD has a higher average benchmark score at 52,901, versus Intel's 41,621. This places AMD in the 91st percentile and Intel in the 88th percentile, despite Intel's dominance in individual head-to-head tests.

Specification Differences

The recorded specifications show several clear points of divergence. Core count differs: 8 for AMD, 16 for Intel. Thread count is identical at 16. Base clock differs, AMD at 2.00 GHz and Intel at 2.20 GHz. Boost clock also differs, AMD at 5.00 GHz and Intel at 5.30 GHz. The socket changes completely: AMD uses AMD Socket FP8, Intel uses Intel BGA 2049. The architecture is listed as null for AMD but Arrow Lake for Intel. Codename differs, Gorgon Point for AMD and Arrow Lake-H for Intel. Generation differs, Ryzen AI Embedded (Zen 5 / Zen 5c) versus Ultra 7 (Arrow Lake-H).

Process node differs, 4 nm for AMD and 3 nm for Intel, though both are TSMC. Die size is only listed for AMD at 233 mm². Cache hierarchy differs across all levels: L1 is 80 KB per core for AMD and 192 KB per core for Intel; L2 is 1 MB per core for AMD and 3 MB per core for Intel; L3 is 8 MB for AMD and 24 MB shared for Intel. Memory bandwidth differs, 89.6 GB/s for AMD and 102.4 GB/s for Intel. PCIe differs, Gen 4 with 16 lanes for AMD and Gen 5 with 8 lanes for Intel. Integrated graphics differ, Radeon 880M for AMD and Arc Graphics 140T for Intel. Release dates differ, with AMD listed as 2026-03-08 and Intel as 2025-01-12. Part numbers differ, unknown for AMD and SRQAQ for Intel. Both share the same TDP of 28 watts, dual-channel memory buses, DDR5 and LPDDR5X support, ECC support, mobile market segment, active production status, and locked multipliers.

The Verdict

The data points to a clear split between raw throughput and overall average positioning. Intel wins the vast majority of individual workloads, often by large margins. The physics score, 2,497 versus 1,210, and floating point math, 109,123 versus 55,799, indicate that Intel's 16 physical cores provide a substantial advantage in parallel compute. Prime number finding, where Intel leads by 78.8%, further confirms this. For workloads that scale with core count and cache size, Intel is the stronger choice.

AMD's case rests on a different set of observations. Its average benchmark score is higher, 52,901 versus 41,621, and it holds a 91st percentile ranking against Intel's 88th. The AMD part also wins integer math, which may indicate an advantage in specific integer-heavy tasks. The 8 MB L3 cache and 16 PCIe Gen 4 lanes might suit certain embedded configurations, though the recorded data does not quantify these effects.

The verdict from the measurements is that Intel is the performance leader in most recorded tests, but AMD holds a better overall average and a higher percentile placement. The choice depends on whether the user prioritizes the specific workloads where Intel excels, such as floating point, physics, and multithreaded tasks, or the broader average performance metric where AMD comes out ahead.

Where Each One Wins

Intel's wins concentrate in compute-heavy and parallel workloads. Floating point math, physics, multithread, data encryption, prime number finding, and random string sorting all favor Intel by margins ranging from 14.6% to 78.8%. These are the categories where more physical cores and a larger shared L3 cache produce measurable gains. Data compression also goes to Intel, though by only 2%. Extended instructions and single-thread performance also favor Intel, giving it a broad base of wins across both multithreaded and lightly threaded tasks.

AMD's single win is integer math, with a 2.9% margin. That is a narrow lead, but it marks the one recorded area where AMD's architecture, with its 8 cores and 16 threads, outperforms Intel's 16-core design. The higher average benchmark score and 91st percentile placement also count as qualitative wins for AMD, even if they do not appear in the head-to-head table. The recorded data does not show a second category where AMD leads, so its case rests on the integer result and the overall average. For workloads that resemble the PassMark integer test, the AMD part is the better fit. For everything else in the recorded set, Intel delivers a faster result, often by a wide margin.