AMD Ryzen Embedded 8845HS vs Intel Core Ultra 5 338H Comparison
AMD Ryzen Embedded 8845HS
Core Ultra 5 338H
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen Embedded 8845HS vs Intel Core Ultra 5 338H
AMD Ryzen Embedded 8845HS vs Intel Core Ultra 5 338H
Head-to-Head Benchmarks
The Intel Core Ultra 5 338H delivers a dominant benchmark performance across all recorded workloads, while the AMD Ryzen Embedded 8845HS has no benchmark scores in the database to compare directly. The database records zero wins for the AMD part in this head-to-head comparison, while the Intel processor records wins in every available category. The Intel Core Ultra 5 338H achieves a multi-core score of 16,331 in Cinebench R23, a result that places it in the 84th percentile of all CPUs in the database. Its single-core result in the same test reaches 2,044 points. In Cinebench R20, the Intel part scores 10,213 multi-core and 1,441 single-core, while in Cinebench R15 it records 2,504 multi-core and 305 single-core.
The PassMark suite shows the Intel processor's strongest areas. The multi-thread score of 28,717 indicates substantial parallel throughput, while the single-thread score of 4,180 confirms strong per-core performance. The data compression workload produces a score of 276,539, far ahead of any other recorded result for this chip. Floating point math reaches 84,067, and integer math scores 64,934. Extended instruction workloads produce 23,906 points, while random string sorting achieves 34,082. Data encryption scores 21,367, physics processing reaches 2,697, and the find prime numbers workload records 304 points.
The average benchmark score for the Intel Core Ultra 5 338H is 33,989. Its nearest rivals in the database are all close competitors: the Intel Core Ultra 7 165H averages 34,083, which is 0.3 percent ahead of the Core Ultra 5 338H. The Intel Core i7-12800HX averages 33,875, 0.3 percent behind. The Intel Xeon 6353P averages 33,844, 0.4 percent behind, and the AMD EPYC 4244P averages 34,220, 0.7 percent ahead. The recorded data indicates the Core Ultra 5 338H sits in a tightly contested performance band among these processors, with all deltas within one percent.
The AMD Ryzen Embedded 8845HS has no benchmark entries and no nearest rivals listed, so no direct score comparisons are possible. The Intel part is therefore the only processor in this comparison with measurable performance data.
Architecture Differences
The two processors come from different manufacturing processes. The AMD Ryzen Embedded 8845HS uses a 4 nm process node fabricated by TSMC, while the Intel Core Ultra 5 338H uses a 3 nm node fabricated by Intel. The AMD chip is built on the Zen 4 architecture with the codename Hawk Point, part of the 8000 series. The Intel part uses the Panther Lake architecture, carries the codename Panther Lake, and belongs to the Core Ultra Series 3 generation, specifically Panther Lake-H. The AMD processor integrates 25,000 million transistors on a 178 mm² die, while the Intel processor has no transistor count or die size recorded in the database.
Core configurations differ significantly. The AMD processor has 8 cores and 16 threads, while the Intel processor has 12 cores and 12 threads. The AMD part therefore offers simultaneous multithreading, doubling its thread count from cores, whereas the Intel part has a one-to-one core-to-thread ratio. The AMD base clock is 3.80 GHz with a boost clock of 5.10 GHz. The Intel base clock is 1.90 GHz with a boost clock of 4.70 GHz. The AMD part runs at a 45 W TDP, while the Intel part runs at 25 W TDP. The AMD processor uses an AMD Socket FP8, and the Intel processor uses an Intel BGA 2540 socket.
Cache hierarchies differ in both capacity and organization. The AMD processor has 64 KB of L1 cache per core, 1 MB of L2 cache per core, and 16 MB of shared L3 cache. The Intel processor has 192 KB of L1 cache per core, 2.5 MB of L2 cache per core, and 18 MB of shared L3 cache. The Intel part has a larger L3 pool and larger per-core L1 and L2 allocations.
Memory support splits along vendor lines. The AMD processor supports DDR5 memory on a dual-channel bus with 89.6 GB/s of memory bandwidth and includes ECC memory support. The Intel processor supports LPDDR5X memory on a dual-channel bus with 136.5 GB/s of memory bandwidth and does not support ECC memory. The Intel memory bandwidth exceeds the AMD figure by 46.9 GB/s. PCIe connectivity also differs: the AMD processor provides Gen 4 with 20 lanes from the CPU, while the Intel processor provides Gen 5 with 4 lanes from the CPU. Integrated graphics differ as well, with the AMD part using the Radeon 780M and the Intel part using the Arc B370. Both processors target the mobile market segment and are currently active in production. The AMD processor was released on April 1, 2024, and the Intel processor on January 4, 2026.
FAQ
Q: Which processor has more cores and threads?
A: The Intel Core Ultra 5 338H has 12 cores and 12 threads. The AMD Ryzen Embedded 8845HS has 8 cores and 16 threads. The AMD part has fewer cores but more threads due to simultaneous multithreading.
Q: What are the clock speed differences?
A: The AMD processor has a base clock of 3.80 GHz and a boost clock of 5.10 GHz. The Intel processor has a base clock of 1.90 GHz and a boost clock of 4.70 GHz. The AMD part runs at higher clocks in both metrics.
Q: Which processor has higher memory bandwidth?
A: The Intel Core Ultra 5 338H records 136.5 GB/s of memory bandwidth with LPDDR5X support. The AMD Ryzen Embedded 8845HS records 89.6 GB/s with DDR5 support. The Intel part has the higher bandwidth figure.
Q: Does either processor support ECC memory?
A: The AMD Ryzen Embedded 8845HS supports ECC memory. The Intel Core Ultra 5 338H does not support ECC memory.
Q: Which processor has a larger L3 cache?
A: The Intel Core Ultra 5 338H has 18 MB of shared L3 cache. The AMD Ryzen Embedded 8845HS has 16 MB of shared L3 cache. The Intel part has 2 MB more L3 cache.
Q: What process nodes do the two processors use?
A: The AMD processor uses a 4 nm process node from TSMC. The Intel processor uses a 3 nm process node from Intel. The Intel part uses a smaller process node.
Specification Differences
| Specification | AMD Ryzen Embedded 8845HS | Intel Core Ultra 5 338H |
|---|---|---|
| Series | 8000 series | Core Ultra Series 3 |
| Cores | 8 | 12 |
| Threads | 16 | 12 |
| Base clock | 3.80 GHz | 1.90 GHz |
| Boost clock | 5.10 GHz | 4.70 GHz |
| TDP | 45 W | 25 W |
| Socket | AMD Socket FP8 | Intel BGA 2540 |
| Architecture | Zen 4 | Panther Lake |
| Codename | Hawk Point | Panther Lake |
| Generation | Ryzen Embedded (Zen 4, Hawk Point) | Ultra 5 (Panther Lake-H) |
| Process node | 4 nm | 3 nm |
| Foundry | TSMC | Intel |
| Transistors | 25,000 million | Not recorded |
| Die size | 178 mm² | Not recorded |
| L1 cache | 64 KB per core | 192 KB per core |
| L2 cache | 1 MB per core | 2.5 MB per core |
| L3 cache | 16 MB shared | 18 MB shared |
| Memory support | DDR5 | LPDDR5X |
| Memory bandwidth | 89.6 GB/s | 136.5 GB/s |
| ECC memory | Yes | No |
| PCIe | Gen 4, 20 lanes (CPU only) | Gen 5, 4 lanes (CPU only) |
| Integrated graphics | Radeon 780M | Arc B370 |
| Release date | April 1, 2024 | January 4, 2026 |
| Part number | Unknown | SA4REQ9EW |
The Verdict
The recorded data shows a clear divergence between the two processors. The Intel Core Ultra 5 338H has comprehensive benchmark results and a percentile rank of 84, while the AMD Ryzen Embedded 8845HS has no benchmark scores at all. The Intel processor's average benchmark score of 33,989 places it in the top tier of mobile processors, within 0.7 percent of the AMD EPYC 4244P and 0.3 percent of the Intel Core Ultra 7 165H. The AMD processor cannot be placed in any performance ranking because the database contains no measurements for it.
The Intel part offers 12 cores versus 8, a larger L3 cache of 18 MB versus 16 MB, higher memory bandwidth of 136.5 GB/s versus 89.6 GB/s, and a smaller 3 nm process node versus 4 nm. The AMD part counters with higher clock speeds, a 5.10 GHz boost versus 4.70 GHz, more threads at 16 versus 12, ECC memory support, and more PCIe lanes at 20 versus 4. The AMD part also has a higher TDP of 45 W versus 25 W, which aligns with its higher clocks.
The data supports choosing the Intel Core Ultra 5 338H for any workload where measured performance matters. It delivers verified multi-core and single-core scores across Cinebench and PassMark tests. The AMD Ryzen Embedded 8845HS remains an option for systems requiring ECC memory, higher PCIe lane counts, or the AMD socket ecosystem, but the database provides no performance evidence for it. The Intel processor is the only one of the two with recorded benchmark data, and that data shows strong results across every tested workload.
Where Each One Wins
The Intel Core Ultra 5 338H wins in all recorded benchmark categories. Cinebench R23 multi-core at 16,331 and single-core at 2,044 show strength in both parallel and single-threaded rendering. PassMark multi-thread at 28,717 and single-thread at 4,180 confirm broad workload coverage. The data compression score of 276,539 indicates particular strength in compression tasks, and the floating point math score of 84,067 suggests solid numerical processing capability. The 84th percentile ranking places it above the large majority of all CPUs in the database.
The AMD Ryzen Embedded 8845HS has no benchmark wins in the database. Its advantages exist only in specifications. It has a higher boost clock of 5.10 GHz, which could favor lightly threaded tasks if the chip were measured, but no data confirms this. It has 16 threads versus 12, which could help heavily threaded workloads, but again no scores exist. It supports ECC memory, which matters for data integrity in embedded and server-style deployments. It provides 20 PCIe Gen 4 lanes versus 4 Gen 5 lanes, which supports more expansion devices. Its 45 W TDP suggests higher power draw than the Intel part's 25 W, but the database does not record performance per watt for either processor.
For use cases, the Intel processor suits any application requiring verified throughput: rendering, compression, encryption, and general compute. The AMD processor suits use cases defined by its feature set rather than measured performance: ECC memory reliability, broad PCIe expansion, and the AMD FP8 socket platform. The database records no performance data for the AMD part, so any workload decision must rely on the Intel processor's confirmed results.