AMD Ryzen AI Embedded P185 vs Intel Core Ultra 5 235HX Comparison

Where Each One Wins

The benchmark split between these two mobile processors is heavily one-sided. Out of 11 recorded head-to-head tests, the Intel Core Ultra 5 235HX claims 10 wins, while the AMD Ryzen AI Embedded P185 secures a single victory. That lone AMD win comes in PassMark integer math, a workload that exercises basic arithmetic operations and often reflects core count and thread scheduling efficiency. The AMD part delivers 117,832 in that test, which is 21.3% ahead of Intel's 97,177.

Every other measured discipline favors the Intel part. The Intel Core Ultra 5 235HX leads in data compression by 12.2%, data encryption by 40%, extended instructions by 23.7%, prime number finding by 64.3%, floating point math by 45.6%, multithread performance by 22.1%, physics by 30.5%, random string sorting by 20.4%, and single-thread performance by 15.1%. These are not marginal edges; several exceed 20% and two surpass 40%.

The use-case split is therefore clear. For integer-heavy code, whether that involves database operations, certain encryption algorithms, or general business logic, the AMD Ryzen AI Embedded P185 holds an advantage. For floating point work, scientific calculations, compression, sorting, and any single-threaded responsiveness, the Intel Core Ultra 5 235HX is the stronger performer by a substantial margin. The multithread score of 40,849 for Intel versus 31,817 for AMD confirms that the Intel part wins in heavily parallel workloads too, despite having fewer threads overall.

Architecture Differences

The two processors take fundamentally different design approaches. The AMD Ryzen AI Embedded P185 is built on Gorgon Point, part of the Ryzen AI Embedded generation using Zen 5 and Zen 5c cores. It has 12 cores and 24 threads, which means it relies on simultaneous multithreading to double its thread count. The base clock is 2.00 GHz and the boost clock reaches 5.10 GHz. It is fabricated on a 4 nm process at TSMC, with a die size of 233 mm². The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of L3.

The Intel Core Ultra 5 235HX belongs to the Core Ultra Series 2, codenamed Arrow Lake-HX. It has 14 cores and 14 threads, meaning there is no hyperthreading; each physical core corresponds to one thread. The base clock is higher at 2.90 GHz, and the boost clock matches at 5.10 GHz. Intel uses a 3 nm process from TSMC, with 17,800 million transistors and a die size of 243 mm². The cache layout is different: 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3.

Memory support also differs. AMD supports both DDR5 and LPDDR5X, with dual-channel configuration and 89.6 GB/s of bandwidth. Intel supports DDR5 only, also dual-channel, but with 102.4 GB/s of bandwidth. The Intel part has a wider memory pipe despite the narrower memory type list. AMD includes ECC memory support, which Intel does not. PCIe connectivity differs as well: AMD offers Gen 4 with 16 lanes from the CPU, while Intel offers Gen 5 with 20 lanes from the CPU.

Integrated graphics are present on both. AMD equips the Radeon 890M, while Intel uses Arc Xe-LPG Graphics with 48 execution units. The Intel part has an unlocked multiplier, whereas the AMD part does not. The power envelopes are distinct: AMD is rated at 28 W TDP, Intel at 55 W TDP. The Intel part draws more power, which partly explains its higher clock base and performance figures. The release dates are also different: Intel launched in January 2025, AMD in February 2026.

Head-to-Head Benchmarks

The data compression test shows the Intel Core Ultra 5 235HX scoring 426,417 against AMD's 374,429, a 12.2% advantage. Compression workloads benefit from large caches and high memory bandwidth, and Intel's 24 MB shared L3 and 102.4 GB/s bandwidth align with that requirement. The encryption test is more decisive: Intel scores 32,697 versus 19,612, a 40% gap. Encryption often relies on dedicated instruction sets and efficient integer pipelines; Intel's extended instruction advantage of 23.7% (34,808 versus 26,544) reinforces that pattern.

Prime number finding is the largest relative win for Intel. The score of 361 versus 129 represents a 64.3% lead. This workload is extremely sensitive to clock speed and integer execution efficiency, and Intel's higher base clock of 2.90 GHz compared to 2.00 GHz is a likely contributor. Floating point math shows Intel at 129,819 against AMD's 70,587, a 45.6% edge. This is a strong indicator for scientific computing and any workload using trigonometric or transcendental functions. The physics test, which simulates rigid body dynamics and collision detection, gives Intel 2,550 versus 1,772, a 30.5% lead. Random string sorting, a memory-latency-sensitive benchmark, shows Intel at 50,929 versus 40,557, a 20.4% advantage.

The single-thread tests are notable. Intel scores 4,683 in both PassMark single thread and singlethread tests, while AMD scores 3,977 in both. The 15.1% gap confirms that Intel has superior per-core performance, which matters for legacy applications and lightly threaded software. The multithread test gives Intel 40,849 versus 31,817, a 22.1% lead. This is surprising given that AMD has 24 threads versus Intel's 14, but Intel's higher clocks and wider memory bandwidth compensate for the thread deficit.

The only AMD win, integer math, shows 117,832 versus 97,177, a 21.3% margin. This test measures basic integer operations, and AMD's higher thread count of 24 likely helps here, as integer math can scale well across many threads without requiring the same memory bandwidth as other workloads. The overall benchmark averages reflect this split: AMD's average benchmark score is 62,839, while Intel's is 52,073. However, the average includes all tests, and the head-to-head data shows Intel winning most of them. AMD's percentile rank among all CPUs is 93, while Intel sits at 91, indicating that AMD's overall position in the broader database is slightly higher despite losing the direct comparison in the majority of tests.

The Verdict

The data directs different users to different processors. The Intel Core Ultra 5 235HX is the choice for any workload that stresses floating point math, encryption, compression, physics simulation, or single-threaded responsiveness. Its 45.6% lead in floating point and 64.3% lead in prime number finding are decisive. The multithread advantage of 22.1% also makes it the better option for parallel rendering or video encoding tasks that use all available threads.

The AMD Ryzen AI Embedded P185 is the pick for integer-heavy code that can exploit its 24 threads. The 21.3% win in integer math is the only benchmark victory, but it is a meaningful one for server-side processing, data transformation, and certain financial calculations. The lower 28 W TDP also indicates a more power-efficient design, which matters for embedded or battery-constrained systems. The ECC memory support is another differentiator; for reliability-critical embedded applications, that feature alone could swing the decision toward AMD.

There is no universal winner. The Intel part wins 10 of 11 tests and by larger margins in several cases, so for most general-purpose mobile computing, the Intel Core Ultra 5 235HX is the stronger performer. The AMD part serves a narrower but legitimate niche: integer-heavy multithreaded workloads in power-limited, ECC-equipped embedded environments. The data does not support calling one objectively superior; it supports matching the processor to the workload profile.

FAQ

Q: Which processor has more threads?

A: The AMD Ryzen AI Embedded P185 has 24 threads from 12 cores, while the Intel Core Ultra 5 235HX has 14 threads from 14 cores. AMD uses simultaneous multithreading, Intel does not.

Q: What is the largest single benchmark margin between the two?

A: The largest margin is in PassMark find prime numbers, where the Intel Core Ultra 5 235HX scores 361 versus 129 for AMD, a 64.3% advantage.

Q: Does the AMD processor win any benchmark?

A: Yes, the AMD Ryzen AI Embedded P185 wins PassMark integer math with a score of 117,832, which is 21.3% ahead of Intel's 97,177.

Q: How do their memory bandwidth figures compare?

A: The Intel Core Ultra 5 235HX has 102.4 GB/s of memory bandwidth, while the AMD Ryzen AI Embedded P185 has 89.6 GB/s. Intel's bandwidth is higher, and it supports DDR5 only, whereas AMD supports both DDR5 and LPDDR5X.

Q: Which processor has ECC memory support?

A: The AMD Ryzen AI Embedded P185 supports ECC memory. The Intel Core Ultra 5 235HX does not support ECC memory.

Q: What are the TDP ratings for each processor?

A: The AMD Ryzen AI Embedded P185 is rated at 28 W TDP, while the Intel Core Ultra 5 235HX is rated at 55 W TDP. The Intel part consumes more power and delivers higher performance in most tests.