AMD Ryzen AI Embedded P164 vs Intel Core Ultra 5 336H Comparison

Head-to-Head Benchmarks

The recorded data shows a clear split between the AMD Ryzen AI Embedded P164 and the Intel Core Ultra 5 336H, with each processor dominating distinct workload types. In the 11 shared PassMark tests, AMD takes 5 wins and Intel takes 6, but the margins tell a more nuanced story than the raw count.

The AMD Ryzen AI Embedded P164 delivers its largest victory in integer math, scoring 87940 against Intel's 62070, a 41.7% advantage. This is the single biggest delta in either direction across the entire benchmark suite. Data compression also goes strongly to AMD, with a score of 327891 versus 280340, a 17% lead. These two results indicate that the AMD part is substantially stronger in workloads that rely on integer throughput and compression algorithms.

The Intel Core Ultra 5 336H counters with equally decisive wins in other areas. Floating point math goes to Intel by a wide margin: 82415 versus 55799, a 32.3% deficit for AMD. The prime number search test shows an even starker gap, with Intel scoring 299 against AMD's 71, a 76.3% difference. Physics simulation also favors Intel heavily, 2682 versus 1210, a 54.9% lead. These three results point to Intel having a significant advantage in floating point heavy and simulation oriented tasks.

The remaining head-to-head results are closer. Intel wins multithread performance with 28545 against 25889, a 9.3% margin, and data encryption with 21291 versus 16055, a 24.6% edge. Extended instructions are nearly identical, with Intel ahead by just 1.4% (24542 versus 24193). Random string sorting goes to AMD by a slim 1.2% (34801 versus 34402). Single-thread performance is effectively a tie, with AMD leading by only 0.4% (4029 versus 4013).

Looking at the broader context from the database, the AMD part sits at the 91st percentile among all CPUs, while Intel sits at the 84th. The average benchmark score confirms this ranking: AMD averages 52901, Intel averages 34485. AMD's nearest rivals include the AMD Ryzen 9 7900X at 53288 (0.7% higher) and the AMD EPYC 7313P at 53206 (0.6% higher), placing it in high-end desktop and server territory. Intel's nearest rivals are notably lower: the Intel Core i7-13700HX at 34554 (0.2% higher) and the Intel Core i5-13450HX at 34333 (0.4% lower), placing it in the mobile performance mainstream.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD Ryzen AI Embedded P164 uses the Gorgon Point codename and belongs to the Ryzen AI Embedded generation built on Zen 5 and Zen 5c cores. It is manufactured on a 4 nm process at TSMC with a die size of 233 mm². The Intel Core Ultra 5 336H uses the Panther Lake architecture, specifically the Panther Lake-H generation, built on a 3 nm process at Intel with no die size recorded.

Core counts differ significantly. AMD provides 8 cores and 16 threads, while Intel provides 16 cores and 16 threads. This means the Intel part relies on more physical cores without hyperthreading, while AMD uses fewer cores with simultaneous multithreading. The cache hierarchy reflects this difference. AMD allocates 80 KB of L1 per core, 1 MB of L2 per core, and 8 MB of L3. Intel allocates 192 KB of L1 per core, 2.5 MB of L2 per core, and 18 MB of shared L3, giving it more than double the last-level cache.

Memory support is similar in type, with both supporting DDR5 and LPDDR5X over dual-channel buses. However, Intel records a higher memory bandwidth of 115.2 GB/s versus AMD's 89.6 GB/s. ECC memory support goes only to AMD, which is a notable differentiator for embedded and reliability-focused deployments. PCIe connectivity also differs: AMD uses Gen 4 with 16 lanes, while Intel uses Gen 5 with 12 lanes. Intel's newer PCIe generation offers higher per-lane bandwidth, though AMD provides more total lanes.

Integrated graphics differ as well. AMD pairs the CPU with the Radeon 880M, while Intel uses Intel Xe3 Graphics. Both are mobile-oriented integrated solutions, but no benchmark data is available in the database to compare them directly.

The production status for both is Active. AMD's release date is recorded as 2026-03-08, while Intel's is 2026-01-04, making Intel the earlier release by roughly two months. Neither processor has a launch MSRP recorded, and both have locked multipliers.

FAQ

Q: Which processor has the higher single-thread performance?

A: The AMD Ryzen AI Embedded P164 leads by a very narrow margin. It scores 4029 in the single-thread test against Intel's 4013, a 0.4% difference. In practice, this is within run-to-run variance and should be treated as a tie.

Q: Why does Intel win the multithread test despite having the same thread count?

A: Intel records 16 physical cores and 16 threads, while AMD records 8 cores and 16 threads. The Intel part's additional physical cores contribute to its 28545 multithread score versus AMD's 25889, a 9.3% advantage, despite the equal thread count.

Q: Is the Intel processor faster in every floating point workload?

A: The data shows Intel wins floating point math by 32.3% and physics simulation by 54.9%. The prime number test, which often relies on integer and floating point operations, also goes to Intel by 76.3%. The database does not include additional floating point tests beyond these.

Q: Does AMD have any clear strength in this comparison?

A: Yes. AMD wins integer math by 41.7% and data compression by 17%. These are substantial margins that indicate strong performance in integer-heavy tasks such as compression, encoding, and general data processing.

Q: How do these processors rank against the broader CPU market?

A: AMD ranks at the 91st percentile among all CPUs with an average benchmark score of 52901. Intel ranks at the 84th percentile with an average score of 34485. AMD's nearest rival is the AMD Ryzen 5 9500F at 52873, just 0.1% lower, while Intel's nearest rival is the Intel Core i7-13700HX at 34554, 0.2% higher.

Q: Which processor supports ECC memory?

A: Only the AMD Ryzen AI Embedded P164 supports ECC memory. The Intel Core Ultra 5 336H does not list ECC support in its specifications.

Specification Differences

The two processors differ across nearly every major specification field. AMD uses 8 cores and 16 threads, while Intel uses 16 cores and 16 threads. AMD's base clock is 2.00 GHz with a boost clock of 5.00 GHz. Intel's base clock is 1.90 GHz with a boost clock of 4.60 GHz. AMD holds the clock speed advantage on both ends.

Thermal design power is close but not identical: AMD is rated at 28 W, Intel at 25 W. The socket differs entirely, with AMD using AMD Socket FP8 and Intel using Intel BGA 2540. The process node favors Intel at 3 nm versus AMD's 4 nm, and the foundries differ (Intel versus TSMC). Cache configurations are substantially different, with Intel providing more L1 per core (192 KB versus 80 KB), more L2 per core (2.5 MB versus 1 MB), and more total L3 (18 MB shared versus 8 MB). Memory bandwidth favors Intel at 115.2 GB/s versus 89.6 GB/s. ECC memory support is exclusive to AMD. PCIe generation and lane count differ: AMD offers Gen 4 with 16 lanes, Intel offers Gen 5 with 12 lanes. Integrated graphics are Radeon 880M on AMD and Intel Xe3 Graphics on Intel. The codenames differ (Gorgon Point versus Panther Lake), as do the generations (Ryzen AI Embedded with Zen 5 / Zen 5c versus Ultra 5 with Panther Lake-H). Intel has a recorded part number, SA4RD, while AMD's is unknown. Die size is recorded only for AMD at 233 mm².

The Verdict

The benchmark data indicates two different usage profiles. The AMD Ryzen AI Embedded P164 is the stronger overall processor based on average score and percentile ranking. Its 52901 average benchmark score and 91st percentile placement put it well ahead of Intel's 34485 average and 84th percentile. The closest rivals to AMD are desktop and server class parts like the AMD Ryzen 9 7900X and AMD EPYC 7313P, which speaks to its performance tier. Intel's closest rivals, by contrast, are mobile HX-series parts and the AMD Ryzen 7 3700X, a desktop part from an earlier generation.

However, the head-to-head results complicate the picture. Intel wins more individual tests, including multithread, floating point math, physics, encryption, and prime number search. AMD wins the high-margin integer and compression tests plus the single-thread test by a hair. The choice between these two depends entirely on which workload family matters more. The AMD part also offers ECC memory support, which is a practical consideration for embedded systems where data integrity is critical. The Intel part offers a newer PCIe generation and higher memory bandwidth, which may matter for certain I/O bound tasks.

Where Each One Wins

The AMD Ryzen AI Embedded P164 wins in integer math, data compression, random string sorting, and single-thread performance. The 41.7% integer math lead and 17% compression lead are the strongest arguments for choosing AMD. These workloads appear in database operations, file compression, and general application logic. The single-thread win, while small, confirms that AMD's architecture holds its own in latency-sensitive tasks.

The Intel Core Ultra 5 336H wins in multithread performance, floating point math, physics simulation, prime number search, data encryption, and extended instructions. The multithread lead of 9.3% and the floating point lead of 32.3% make Intel the better choice for rendering, simulation, scientific computing, and encryption workloads. The physics score advantage of 54.9% is particularly notable for any application with physics-based computation.

The overall average benchmark score favors AMD, as does the percentile ranking. Buyers who need a balanced processor with strong integer performance and ECC support should lean toward the AMD part. Buyers who need maximum floating point throughput and physical core count should lean toward Intel. The data does not support a single universal winner; it supports a workload-dependent decision.