AMD Ryzen Embedded 9900X vs Intel Core Ultra 5 235H Comparison

The Verdict

The AMD Ryzen Embedded 9900X and the Intel Core Ultra 5 235H serve completely different segments of the market, and the data makes that split clear. The AMD part is a desktop processor on Socket AM5 with 12 cores and 24 threads, a 4.40 GHz base clock, a 5.60 GHz boost clock, and a 120 W TDP. The Intel part is a mobile processor on BGA 2049 with 14 cores and 14 threads, a 2.40 GHz base clock, a 5.00 GHz boost clock, and a 28 W TDP. The AMD chip is built for sustained multi-threaded workloads in a desktop chassis. The Intel chip is built for power-limited mobile systems where efficiency matters more than raw throughput.

The recorded data shows only the Intel Core Ultra 5 235H has benchmark entries. The AMD Ryzen Embedded 9900X has no benchmark scores, no average score, and no nearest rivals in the database. The Intel part sits at the 85th percentile among all CPUs, with an average benchmark score of 37522. Its nearest rivals include the Intel Core i9-13900HK at 37425 (0.3% behind), the Intel Core i5-13600K at 37685 (0.4% ahead), the Intel Core Ultra 5 225F at 37313 (0.6% behind), and the AMD Ryzen AI 5 PRO 435 at 37762 (0.6% ahead). These deltas are all within a single percentage point, which indicates the Core Ultra 5 235H sits in a tightly packed performance cluster. The AMD Ryzen Embedded 9900X has a 50th percentile ranking, but with no benchmark data, that figure carries no interpretive weight.

Who should pick which? For a desktop workstation that needs 24 threads, a 5.60 GHz boost, 64 MB of L3 cache, ECC memory support, and a 120 W power budget, the AMD Ryzen Embedded 9900X is the only logical choice from the data. For a mobile system that needs 14 threads, a 5.00 GHz boost, 18 MB of shared L3 cache, and a 28 W TDP, the Intel Core Ultra 5 235H is the only viable option. There is no overlap in socket, power class, or intended use case.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Core Ultra 5 235H has 14 cores, but only 14 threads, meaning it lacks simultaneous multithreading. The AMD Ryzen Embedded 9900X has 12 cores and 24 threads, giving it 12 more threads than the Intel part.

Q: Which processor has a higher boost clock?

A: The AMD Ryzen Embedded 9900X boosts to 5.60 GHz. The Intel Core Ultra 5 235H boosts to 5.00 GHz. The AMD part has a 0.60 GHz advantage in peak frequency.

Q: Do both processors support ECC memory?

A: No. The AMD Ryzen Embedded 9900X supports ECC memory. The Intel Core Ultra 5 235H does not support ECC memory.

Q: Which processor has more L3 cache?

A: The AMD Ryzen Embedded 9900X has 64 MB of L3 cache. The Intel Core Ultra 5 235H has 18 MB of shared L3 cache. The AMD part has 46 MB more L3 cache.

Q: Which processor has a higher memory bandwidth rating?

A: The Intel Core Ultra 5 235H has a memory bandwidth of 102.4 GB/s. The AMD Ryzen Embedded 9900X has a memory bandwidth of 89.6 GB/s. The Intel part is 12.8 GB/s higher.

Q: Which processor is unlocked for overclocking?

A: The AMD Ryzen Embedded 9900X has an unlocked multiplier. The Intel Core Ultra 5 235H does not have an unlocked multiplier.

Architecture Differences

The AMD Ryzen Embedded 9900X uses the Granite Ridge codename and belongs to the Ryzen Embedded generation based on Zen 5 architecture. It is fabricated on a 4 nm process at TSMC, with 16,630 million transistors across a die size of 2x 70.6 mm². The cache hierarchy uses 80 KB of L1 per core, 1 MB of L2 per core, and a shared 64 MB of L3. It supports DDR5 memory on a dual-channel bus and carries a Radeon Graphics integrated GPU. It provides PCIe Gen 5 with 24 lanes from the CPU only. The socket is AMD Socket AM5.

The Intel Core Ultra 5 235H uses the Arrow Lake codename and belongs to the Core Ultra Series 2 generation with Arrow Lake-H architecture. It is fabricated on a 3 nm process at TSMC. The cache hierarchy uses 192 KB of L1 per core, 3 MB of L2 per core, and 18 MB of shared L3. It supports DDR5 and LPDDR5X memory on a dual-channel bus and carries an Arc Graphics 140T integrated GPU. It provides PCIe Gen 5 with 8 lanes from the CPU only. The socket is Intel BGA 2049.

The architectural split is stark. The AMD part is a desktop-class, multi-chip design with a large shared L3 pool and 24 threads. The Intel part is a mobile-class, monolithic design with a smaller shared L3 pool and no hyper-threading. The AMD part has an unlocked multiplier, while the Intel part is locked. The AMD part supports ECC memory, while the Intel part does not. The Intel part uses a smaller 3 nm process node, while the AMD part uses a 4 nm node. The Intel part also has a larger L1 and L2 allocation per core, but a much smaller L3 pool overall.

Specification Differences

The two processors differ across nearly every specification field in the database. The AMD Ryzen Embedded 9900X has 12 cores and 24 threads. The Intel Core Ultra 5 235H has 14 cores and 14 threads. The AMD base clock is 4.40 GHz versus 2.40 GHz for Intel. The AMD boost clock is 5.60 GHz versus 5.00 GHz for Intel. The AMD TDP is 120 W versus 28 W for Intel. The AMD socket is AMD Socket AM5 versus Intel BGA 2049 for Intel. The AMD process node is 4 nm versus 3 nm for Intel. The AMD transistor count is 16,630 million, while the Intel transistor count is not recorded. The AMD die size is 2x 70.6 mm², while the Intel die size is not recorded.

Cache differences are substantial. The AMD part has 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of L3. The Intel part has 192 KB of L1 per core, 3 MB of L2 per core, and 18 MB of shared L3. Memory support differs: AMD lists DDR5 only, while Intel lists DDR5 and LPDDR5X. Both use a dual-channel memory bus. Memory bandwidth differs: AMD at 89.6 GB/s versus Intel at 102.4 GB/s. ECC support is present on AMD and absent on Intel. PCIe lanes differ: AMD provides Gen 5 with 24 lanes from the CPU only, while Intel provides Gen 5 with 8 lanes from the CPU only. Integrated graphics differ: AMD uses Radeon Graphics, Intel uses Arc Graphics 140T.

Market segment differs: AMD is a desktop part, Intel is a mobile part. The AMD release date is 2025-10-06, the Intel release date is 2025-01-12. The AMD multiplier is unlocked, the Intel multiplier is locked. The AMD part number is 100-000000662E, the Intel part number is SRQAP. Neither part has a recorded launch MSRP in the database.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries between the AMD Ryzen Embedded 9900X and the Intel Core Ultra 5 235H. The AMD part has an empty benchmarks array, a zero average benchmark score, and zero wins in head-to-head comparisons. The Intel part has a full set of benchmark scores across Cinebench and Passmark tests, with an average benchmark score of 37522 and zero wins in head-to-head comparisons as well.

The Intel Core Ultra 5 235H benchmark results are as follows. In Cinebench R15, it scores 2580 in multi-core and 364 in single-core. In Cinebench R20, it scores 10751 in multi-core and 1517 in single-core. In Cinebench R23, it scores 25598 in multi-core and 3613 in single-core. In Passmark tests, it scores 301979 in data compression, 23121 in data encryption, 23354 in extended instructions, 239 in find prime numbers, 93509 in floating point math, 74247 in integer math, 30091 in multithread, 1985 in physics, 36208 in random string sorting, and 4359 in single thread.

Without any benchmark data for the AMD Ryzen Embedded 9900X, no direct comparison can be made. The only quantitative context for the Intel part comes from its nearest rivals. The Intel Core Ultra 5 235H is 0.3% ahead of the Intel Core i9-13900HK, 0.4% behind the Intel Core i5-13600K, 0.6% ahead of the Intel Core Ultra 5 225F, and 0.6% behind the AMD Ryzen AI 5 PRO 435. These margins are negligible in practical terms. The Core Ultra 5 235H performs statistically identically to all four rivals in the database.

Where Each One Wins

The AMD Ryzen Embedded 9900X wins in specifications that favor heavy desktop workloads. It has 24 threads versus 14 threads, which gives it a 10-thread advantage for multi-threaded rendering, compilation, or virtualization. It has a 5.60 GHz boost clock versus 5.00 GHz, a 0.60 GHz advantage in peak single-thread frequency. It has 64 MB of L3 cache versus 18 MB, a 46 MB advantage for cache-sensitive workloads. It supports ECC memory, which the Intel part lacks, making it suitable for memory-error-sensitive server or workstation tasks. It has an unlocked multiplier, which allows frequency tuning. It provides 24 PCIe Gen 5 lanes versus 8 lanes, giving it more headroom for expansion cards, storage, and GPUs. It has a 120 W TDP, which indicates a power budget designed for sustained all-core load in a desktop chassis.

The Intel Core Ultra 5 235H wins in specifications that favor mobile efficiency and memory bandwidth. It has 14 cores versus 12 cores, a 2-core advantage for lightly threaded or parallel workloads that do not benefit from SMT. It has a 102.4 GB/s memory bandwidth versus 89.6 GB/s, a 12.8 GB/s advantage for memory-bound tasks. It supports LPDDR5X memory in addition to DDR5, which gives mobile designers more memory options. It has a 28 W TDP versus 120 W, a 92 W lower power budget that suits thin-and-light laptops. It uses a 3 nm process node versus 4 nm, which indicates a denser, potentially more power-efficient transistor layout. It has a larger L1 allocation per core (192 KB versus 80 KB) and a larger L2 allocation per core (3 MB versus 1 MB), which can reduce memory latency for per-core working sets. It sits at the 85th percentile among all CPUs, while the AMD part sits at the 50th percentile, though that comparison is weakened by the AMD part having no benchmark scores.

For practical use-case splits, the data points to the AMD Ryzen Embedded 9900X for desktop workstations that need maximum thread count, high boost clocks, large L3 cache, ECC reliability, and ample PCIe connectivity. The Intel Core Ultra 5 235H is the choice for mobile systems that need a 14-core part with high memory bandwidth, low power draw, and a compact BGA package. The two parts do not compete in the same physical or thermal envelope, and the benchmark data does not put them in direct competition.