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

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 7 265T has 20 cores and 20 threads, while the AMD Ryzen AI Embedded P164 has 8 cores and 16 threads. The Intel part does not use simultaneous multithreading, so its thread count equals its core count.

Q: What is the difference in average benchmark scores between the two?

A: The AMD Ryzen AI Embedded P164 has an average benchmark score of 52901, while the Intel Core Ultra 7 265T scores 47697. The AMD part sits at the 91st percentile of all CPUs, and the Intel part sits at the 90th percentile.

Q: Which processor wins in the PassMark multithread test?

A: The Intel Core Ultra 7 265T wins the PassMark multithread test with a score of 37084 versus 25889 for the AMD Ryzen AI Embedded P164. That is a 30.2% advantage for the Intel processor.

Q: How do the two compare in single-thread performance?

A: The Intel Core Ultra 7 265T scores 4338 in the PassMark single-thread test, while the AMD Ryzen AI Embedded P164 scores 4029. The Intel part leads by 7.1% in this metric.

Q: Do both processors support ECC memory?

A: No. The AMD Ryzen AI Embedded P164 supports ECC memory, while the Intel Core Ultra 7 265T does not. This is a notable difference for embedded and server-oriented workloads.

Q: What are the memory bandwidth figures for each?

A: The AMD Ryzen AI Embedded P164 has a memory bandwidth of 89.6 GB/s, while the Intel Core Ultra 7 265T reaches 102.4 GB/s. The Intel part also supports DDR5 memory, while the AMD part supports both DDR5 and LPDDR5X.

Architecture Differences

The AMD Ryzen AI Embedded P164 uses the Gorgon Point codename and belongs to the Ryzen AI Embedded generation built on Zen 5 / Zen 5c cores. It is manufactured on a 4 nm process at TSMC with a die size of 233 mm². The Intel Core Ultra 7 265T uses the Arrow Lake architecture, specifically Arrow Lake-S, and is built on a 3 nm process, also at TSMC, with a die size of 243 mm². The Intel part has a transistor count of 17,800 million, while the AMD part does not have a recorded transistor count in the database.

The core configuration differs substantially. The AMD chip has 8 cores and 16 threads, indicating simultaneous multithreading is enabled. The Intel chip has 20 cores and 20 threads, meaning it relies on physical cores alone without hyperthreading. Cache hierarchies also diverge: the AMD part has 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3. The Intel part has 192 KB of L1 per core, 3 MB of L2 per core, and 30 MB of shared L3. That gives the Intel processor a much larger total cache footprint, which likely contributes to its advantage in data-heavy workloads.

The memory controllers differ as well. The AMD processor supports both DDR5 and LPDDR5X memory over a dual-channel bus, with a measured bandwidth of 89.6 GB/s. The Intel processor supports only DDR5, also dual-channel, but achieves 102.4 GB/s. The AMD part supports ECC memory, which is absent on the Intel side. PCIe connectivity also differs: the AMD chip provides Gen 4 with 16 lanes (CPU only), while the Intel chip provides Gen 5 with 20 lanes (CPU only). This gives the Intel part both a newer PCIe generation and more lanes.

Integrated graphics are present on both. The AMD Ryzen AI Embedded P164 uses a Radeon 880M, while the Intel Core Ultra 7 265T uses Arc Xe-LPG Graphics with 64 execution units. The market segment also differs: the AMD part is classified as Mobile, while the Intel part is Desktop. The AMD part uses the AMD Socket FP8, while the Intel part uses Intel Socket 1851. The Intel part has a launch MSRP of $384, and the AMD part has no recorded launch MSRP. The AMD part has a base clock of 2.00 GHz and a boost clock of 5.00 GHz, while the Intel part has a base clock of 1.50 GHz and a boost clock of 5.30 GHz. The AMD part has a TDP of 28 watts, and the Intel part has a TDP of 35 watts. The Intel part was released on 2025-01-06, while the AMD part was released on 2026-03-08.

Head-to-Head Benchmarks

The head-to-head data shows a clean sweep for the Intel Core Ultra 7 265T across all 11 recorded PassMark benchmarks. The largest margin appears in the find prime numbers test, where Intel scores 322 against AMD's 71, a 78% difference. This is a massive gap and indicates that the Intel architecture handles this particular integer workload far more efficiently, likely due to its larger core count and higher boost clock.

Floating-point math shows another decisive win for Intel. The Intel part scores 129817 versus 55799 for the AMD part, a 57% advantage. This suggests that the Intel processor's 20 physical cores provide a substantial throughput advantage in mathematically intensive tasks. The physics test also favors Intel heavily, with a score of 2391 versus 1210, a 49.4% difference. Physics simulations often scale with core count and memory bandwidth, both of which favor the Intel chip.

Data encryption is another area where Intel dominates, scoring 29687 versus 16055, a 45.9% lead. This workload benefits from the Intel part's higher memory bandwidth and larger L3 cache. Multithread performance shows a 30.2% advantage for Intel, with scores of 37084 versus 25889. This is consistent with the core count difference, though the AMD part's 16 threads partially close the gap versus its 8 physical cores.

Random string sorting shows Intel ahead by 21.6%, with scores of 44400 versus 34801. Integer math gives Intel a 16.2% edge, with scores of 104943 versus 87940. Extended instructions show a 15.4% lead for Intel, with scores of 28609 versus 24193. Data compression shows a more modest 11.4% advantage for Intel, with scores of 370158 versus 327891.

The closest margin is in single-thread performance. Intel scores 4338 versus AMD's 4029, a 7.1% difference. This indicates that while the Intel part has a higher boost clock of 5.30 GHz versus 5.00 GHz, the AMD architecture's Zen 5 cores remain competitive in single-threaded workloads. The single-thread test appears twice in the data (as passmark_single_thread and passmark_singlethread), with identical scores and deltas, confirming consistency in the measurements.

Despite the Intel sweep in the head-to-head PassMark tests, the average benchmark score tells a different story. The AMD part has an average score of 52901, while the Intel part has 47697. This discrepancy arises because the AMD part's average includes its PassMark scores only, while the Intel part's average also incorporates its Cinebench results (R15, R20, R23), which may be weighted differently. The nearest rivals for the AMD part include the AMD Ryzen 5 9500F (delta 0.1%), Intel Xeon 634 (delta -0.1%), AMD EPYC 7313P (delta -0.6%), and AMD Ryzen 9 7900X (delta -0.7%). The Intel part's nearest rivals include the AMD Ryzen 9 PRO 5945 (delta 0.4%), AMD Ryzen 9 7900X3D (delta -0.4%), AMD Ryzen 9 3900 (delta -0.5%), and Intel Core Ultra X9 378H (delta 0.5%).

Specification Differences

The two processors differ in nearly every core specification. The AMD Ryzen AI Embedded P164 has 8 cores and 16 threads, while the Intel Core Ultra 7 265T has 20 cores and 20 threads. Base clocks are 2.00 GHz for AMD and 1.50 GHz for Intel. Boost clocks are 5.00 GHz for AMD and 5.30 GHz for Intel. TDP ratings are 28 watts for AMD and 35 watts for Intel.

Cache configurations are markedly different. The AMD part has 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3. The Intel part has 192 KB of L1 per core, 3 MB of L2 per core, and 30 MB of shared L3. This gives the Intel part a 22 MB advantage in L3 capacity.

Memory support differs: the AMD part supports DDR5 and LPDDR5X, while the Intel part supports only DDR5. Memory bandwidth is 89.6 GB/s for AMD and 102.4 GB/s for Intel. ECC memory is supported on the AMD part but not on the Intel part. PCIe generation and lane counts differ: the AMD part uses Gen 4 with 16 lanes, while the Intel part uses Gen 5 with 20 lanes.

The process node is 4 nm for AMD and 3 nm for Intel, both at TSMC. The die size is 233 mm² for AMD and 243 mm² for Intel. The Intel part has a recorded transistor count of 17,800 million; the AMD part has none. The AMD part uses the AMD Socket FP8, while the Intel part uses Intel Socket 1851. The integrated graphics are Radeon 880M for AMD and Arc Xe-LPG Graphics 64EU for Intel. The market segment is Mobile for AMD and Desktop for Intel. The AMD part has no launch MSRP, while the Intel part has a launch MSRP of $384.

The Verdict

The recorded data shows that the Intel Core Ultra 7 265T wins every head-to-head PassMark benchmark, with margins ranging from 7.1% in single-thread to 78% in prime number finding. This is a comprehensive victory in the tested workloads. The Intel part's 20 physical cores, larger L3 cache, higher boost clock, and greater memory bandwidth all contribute to these results. The 30.2% multithread advantage and 57% floating-point advantage are particularly significant for compute-heavy tasks.

However, the AMD Ryzen AI Embedded P164 has a higher average benchmark score of 52901 versus 47697 for Intel, and it sits at the 91st percentile versus the 90th percentile for Intel. The AMD part also offers ECC memory support, a lower TDP of 28 watts versus 35 watts, and support for LPDDR5X memory. These features make it more suitable for embedded and mobile scenarios where reliability, power efficiency, and memory flexibility are priorities.

The Intel part is classified as a Desktop processor, while the AMD part is classified as Mobile. The Intel part has a higher boost clock (5.30 GHz versus 5.00 GHz) and a larger L3 cache (30 MB versus 8 MB). The AMD part has a higher base clock (2.00 GHz versus 1.50 GHz) and a smaller die size (233 mm² versus 243 mm²). The Intel part uses PCIe Gen 5 with 20 lanes, while the AMD part uses PCIe Gen 4 with 16 lanes.

For users prioritizing raw performance in the tested PassMark workloads, the Intel Core Ultra 7 265T is the clear choice. For users prioritizing ECC memory, lower power consumption, and a mobile form factor, the AMD Ryzen AI Embedded P164 offers distinct advantages. The Intel part has a launch MSRP of $384, while the AMD part has no recorded launch MSRP.

Where Each One Wins

The Intel Core Ultra 7 265T wins in every recorded head-to-head benchmark. The largest margins are in prime number finding (78% ahead), floating-point math (57% ahead), and physics (49.4% ahead). These are workloads that scale strongly with core count and memory bandwidth. The Intel part also leads in data encryption by 45.9%, which benefits from its larger cache and higher bandwidth. Multithread performance is 30.2% higher, making it the better choice for parallel processing tasks. Even in single-thread performance, the Intel part leads by 7.1%, so it holds the advantage across both single-threaded and multi-threaded scenarios.

The AMD Ryzen AI Embedded P164, despite losing all head-to-head benchmarks, has a higher average benchmark score (52901 versus 47697) and a higher percentile ranking (91st versus 90th). This suggests that in a broader benchmark suite, the AMD part may perform differently, particularly in workloads not captured by the PassMark tests. The AMD part also wins in power efficiency, with a TDP of 28 watts versus 35 watts, and offers ECC memory support, which the Intel part lacks. The AMD part supports LPDDR5X memory, which can be advantageous in low-power mobile designs.

For embedded applications requiring ECC memory, the AMD Ryzen AI Embedded P164 is the only option of the two. For desktop applications where maximum performance in integer, floating-point, and encryption workloads is needed, the Intel Core Ultra 7 265T is the superior choice. The Intel part's PCIe Gen 5 support with 20 lanes also makes it more suitable for high-bandwidth peripheral connectivity. The AMD part's lower TDP and mobile classification make it a better fit for thermally constrained or battery-powered systems. The data does not show any benchmark where the AMD part outperforms the Intel part, so the AMD part's advantages are purely in features, power, and memory support rather than in the tested performance metrics.