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

The Verdict

The benchmark data separates these two processors into distinct roles. The AMD Ryzen AI Embedded P185 wins only one of the eleven recorded head-to-head tests, while the Intel Core Ultra 5 235 takes the other ten. The average benchmark score tells a similar story: the AMD part records 62,839 points against 46,062 for the Intel part, which places the AMD processor at the 93rd percentile of all CPUs and the Intel at the 89th. The AMD Ryzen AI Embedded P185 is the pick for integer-heavy workloads and for environments where its 28 W TDP and ECC memory support matter. The Intel Core Ultra 5 235 is the pick for general desktop tasks, single-threaded responsiveness, encryption, and floating-point math. The data does not support a single universal winner; it supports two different deployment profiles.

Architecture Differences

The AMD Ryzen AI Embedded P185 uses the Gorgon Point codename and belongs to the Ryzen AI Embedded generation built on the Zen 5 and Zen 5c hybrid core design. It is manufactured on a 4 nm process at TSMC with a die size of 233 mm². The Intel Core Ultra 5 235 uses the Arrow Lake-S codename, belongs to the Core Ultra Series 2 generation, and is built on a 3 nm process, also at TSMC, with a die size of 243 mm². The Intel part carries 17,800 million transistors on its die.

Core counts differ in a meaningful way. The AMD chip provides 12 cores and 24 threads, relying on simultaneous multithreading to double its thread count. The Intel chip provides 14 cores and 14 threads, with no multithreading, so each core contributes exactly one thread. The AMD part has 80 KB of L1 per core and 1 MB of L2 per core, with 16 MB of shared L3. The Intel part has 192 KB of L1 per core and 3 MB of L2 per core, with 24 MB of shared L3. This gives the Intel design a larger total cache footprint.

The AMD processor supports both DDR5 and LPDDR5X memory across a dual-channel bus, with a recorded memory bandwidth of 89.6 GB/s. The Intel processor supports DDR5 on a dual-channel bus with a higher recorded bandwidth of 102.4 GB/s. ECC memory is supported on the AMD part but not on the Intel part. The AMD chip uses AMD Socket FP8, while the Intel chip uses Intel Socket 1851. The AMD integrated graphics are the Radeon 890M, while the Intel integrated graphics are the Arc Xe-LPG Graphics 24EU. PCIe connectivity also differs: the AMD part runs Gen 4 with 16 CPU lanes, the Intel part runs Gen 5 with 20 CPU lanes. The AMD part enters the market as a mobile segment processor with a 28 W TDP, while the Intel part is a desktop segment processor with a 65 W TDP.

Head-to-Head Benchmarks

The biggest win for the AMD Ryzen AI Embedded P185 comes in PassMark integer math. The AMD part scores 117,832 against 87,948 for the Intel part, a 34 percent advantage. This is the only test where the AMD processor leads, and it is a substantial lead. Integer math workloads, which include general computation, compression logic, and many application-level operations, clearly favor the AMD architecture.

Every other recorded head-to-head test favors the Intel Core Ultra 5 235. The largest margin is in PassMark find prime numbers, where the Intel part scores 371 against 129, a 65.2 percent advantage. This is a heavily single-threaded, branch-heavy workload, and the Intel core design handles it far better. Floating-point math also goes strongly to Intel: 117,951 against 70,587, a 40.2 percent lead. The Intel part leads by 33 percent in data encryption, scoring 29,293 against 19,612. Extended instructions favor Intel by 19 percent, with 32,752 against 26,544. Random string sorting favors Intel by 17.2 percent, with 48,980 against 40,557. Physics simulation favors Intel by 31.1 percent, with 2,570 against 1,772.

In the broader multithreaded test, the Intel part scores 37,816 against 31,817 for the AMD part, a 15.9 percent lead. This is notable because the AMD processor has 24 threads against 14 for the Intel processor, yet the Intel part still wins. The single-thread tests show an 11.9 percent Intel advantage, with 4,516 against 3,977. Data compression is the closest test: Intel scores 390,711 against 374,429, a 4.2 percent lead. The average benchmark score difference is large, with the AMD part at 62,839 and the Intel part at 46,062, but that average is driven by the AMD part's strong integer result and the Intel part's broader consistency across all other workloads.

The nearest rival data for the AMD part shows it sits between the Intel Core Ultra 7 255HX and the AMD Ryzen AI 9 PRO 465. The Intel Core Ultra 7 265HX is 0.5 percent behind the AMD part in average score. The nearest rival data for the Intel part shows it sits between the AMD Ryzen AI 9 HX 375 and the Intel Core i9-13900HX, with the Intel Core i9-13900HX 0.1 percent behind. These placements confirm that both processors compete in a tightly packed performance band.

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 5 235 has 14 threads from 14 cores. The AMD part uses simultaneous multithreading to double its thread count.

Q: Does the AMD processor support ECC memory?

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

Q: Which processor wins in single-threaded performance?

A: The Intel Core Ultra 5 235 wins the PassMark single-thread test with 4,516 points against 3,977 for the AMD part, an 11.9 percent advantage.

Q: Which processor has the higher boost clock?

A: The AMD Ryzen AI Embedded P185 has a boost clock of 5.10 GHz, which is higher than the Intel Core Ultra 5 235 boost clock of 5.00 GHz.

Q: What is the memory bandwidth of each processor?

A: The AMD part records 89.6 GB/s over dual-channel DDR5 or LPDDR5X, while the Intel part records 102.4 GB/s over dual-channel DDR5.

Q: Which processor wins the most head-to-head benchmarks?

A: The Intel Core Ultra 5 235 wins 10 of the 11 recorded head-to-head tests. The AMD Ryzen AI Embedded P185 wins only the PassMark integer math test.

Where Each One Wins

The AMD Ryzen AI Embedded P185 wins in PassMark integer math by 34 percent. This single result defines its strongest use case: workloads built on integer operations, including many forms of data processing, compression algorithms, and general application logic that rely on integer arithmetic. The AMD part also brings ECC memory support, which matters for data integrity in embedded and server-adjacent deployments. Its 28 W TDP is far lower than the Intel part's 65 W TDP, making it the rational choice for power-constrained mobile or embedded designs. The AMD part's 24 threads give it a theoretical scheduling advantage, but the recorded multithread test shows the Intel part still wins by 15.9 percent, so thread count alone does not translate into a multithreaded win.

The Intel Core Ultra 5 235 wins everywhere else. Its 65.2 percent lead in prime number finding and 40.2 percent lead in floating-point math make it the choice for scientific computation, numerical simulation, and any workload that stresses the FPU. Its 33 percent lead in data encryption makes it the better pick for cryptographic operations and secure communication workloads. Its 31.1 percent lead in physics simulation suits gaming and physics-based modeling. Its 11.9 percent single-thread lead means snappier response in latency-sensitive applications. Its 15.9 percent multithreaded lead, despite fewer threads, shows that the Intel core design extracts more throughput per thread. Its 4.2 percent lead in data compression rounds out a broad sweep of common desktop tasks. The Intel part also offers PCIe Gen 5 with 20 CPU lanes, which the AMD part does not match, and a higher memory bandwidth of 102.4 GB/s against 89.6 GB/s.

Specification Differences

The two processors differ across nearly every recorded specification. The AMD Ryzen AI Embedded P185 uses 12 cores and 24 threads; the Intel Core Ultra 5 235 uses 14 cores and 14 threads. Base clocks differ meaningfully: the AMD part runs at 2.00 GHz, the Intel part at 3.40 GHz. Boost clocks are closer, with the AMD part at 5.10 GHz and the Intel part at 5.00 GHz. TDP differs by more than double: 28 W for AMD against 65 W for Intel.

Cache layouts diverge completely. The AMD part has 80 KB of L1 per core, 1 MB of L2 per core, and 16 MB of L3. The Intel part has 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3. The process node differs by one step, 4 nm for AMD and 3 nm for Intel, both from TSMC. Die size is close, 233 mm² for AMD and 243 mm² for Intel, but the Intel die packs 17,800 million transistors while the AMD transistor count is not recorded.

Memory support differs in type and bandwidth. AMD supports DDR5 and LPDDR5X at 89.6 GB/s; Intel supports DDR5 at 102.4 GB/s. ECC support is present on AMD but absent on Intel. The AMD part uses Gen 4 PCIe with 16 CPU lanes; the Intel part uses Gen 5 PCIe with 20 CPU lanes. Integrated graphics differ as well: the AMD Radeon 890M against the Intel Arc Xe-LPG Graphics 24EU. Sockets are incompatible: AMD Socket FP8 against Intel Socket 1851. The market segments differ, with the AMD part classified as mobile and the Intel part as desktop. The Intel part has a recorded launch MSRP of $257, while the AMD part has no recorded launch MSRP. The AMD part is locked, and the Intel part is also locked; neither offers an unlocked multiplier. The Intel part has a recorded part number of SRQAS, while the AMD part number is recorded as unknown.