AMD EPYC 9554P vs Intel Core Ultra 5 238V Comparison

The Intel Core Ultra 5 238V and AMD EPYC 9554P occupy opposite ends of the computing spectrum: one is a 17W mobile processor built for thin-and-light laptops, the other a 360W server behemoth designed for datacenter workloads. Benchmark results confirm this divide, with the EPYC dominating every shared test, though the Core Ultra’s architecture and feature set make it a compelling choice for its intended mobile market.

Head-to-Head Benchmarks

The head-to-head data is unambiguous: the AMD EPYC 9554P wins all six shared benchmark comparisons, with a consistent performance gap of roughly 82.5% in the EPYC’s favor. In Cinebench R23 multi-core, the EPYC scores 89,490 against the Core Ultra’s 15,645, a delta of -82.5% for the Intel part. The single-core R23 result tells a similar story: EPYC at 12,633 versus Intel at 2,208, again a -82.5% difference. This pattern repeats across Cinebench R15 and R20, with the EPYC scoring 9,020 and 37,585 in R15 and R20 multi-core respectively, while the Core Ultra manages 1,576 and 6,570.

The single-core results are particularly telling. The EPYC’s 3.10 GHz base and 3.75 GHz boost clocks are lower than the Core Ultra’s 2.10 GHz base and 4.70 GHz boost, yet the EPYC still wins single-core tests by the same 82.5% margin. This suggests the Zen 4 architecture’s IPC advantage is substantial, or that the benchmark methodology favors the server chip’s design. The Core Ultra’s higher boost clock does not translate into a win in any test.

With zero wins for the Intel part and six for the AMD part, the data presents a one-sided picture. However, context matters: the EPYC’s 64 cores and 128 threads versus the Core Ultra’s 8 cores and 8 threads mean the multi-core results are almost predetermined. The single-core margin is more surprising and indicates that the EPYC’s Zen 4 cores are simply more efficient per clock than Lunar Lake’s cores in these specific workloads.

FAQ

Q: Which processor has better multi-core performance?

A: The AMD EPYC 9554P wins all multi-core benchmarks by a margin of 82.5%. In Cinebench R23 multi-core, the EPYC scores 89,490 while the Intel Core Ultra 5 238V scores 15,645. The EPYC’s 64 cores and 128 threads provide a massive parallel processing advantage over the Core Ultra’s 8 cores and 8 threads.

Q: Is the Intel Core Ultra 5 238V competitive in single-core tests?

A: No. The EPYC 9554P also wins single-core tests by the same 82.5% margin. In Cinebench R23 single-core, the EPYC scores 12,633 versus the Core Ultra’s 2,208. This holds across R15 and R20 single-core tests as well, with the EPYC scoring 1,273 and 5,305 respectively versus the Core Ultra’s 222 and 927.

Q: What is the power consumption difference?

A: The TDP figures are 17W for the Intel Core Ultra 5 238V and 360W for the AMD EPYC 9554P. This 343W difference reflects their distinct market positions: the Core Ultra is designed for mobile devices where battery life matters, while the EPYC is built for server racks where power is less constrained.

Q: How do their memory architectures compare?

A: The EPYC 9554P supports DDR5 memory across a twelve-channel bus with 460.8 GB/s bandwidth and ECC support. The Core Ultra 5 238V uses dual-channel memory with ECC not supported, and its memory support is listed as dependent on the motherboard. The EPYC’s memory bandwidth is critical for server workloads.

Q: Which processor has more PCIe lanes?

A: The EPYC 9554P provides Gen 5, 128 lanes (CPU only), while the Core Ultra 5 238V offers Gen 5, 4 lanes (CPU only). The EPYC’s 128 lanes enable extensive I/O connectivity for storage and networking, whereas the Core Ultra’s 4 lanes are sufficient for a mobile device’s limited expansion needs.

Q: What are their release dates?

A: The AMD EPYC 9554P was released on 2022-11-09, while the Intel Core Ultra 5 238V was released on 2024-09-23. The EPYC is nearly two years older but still outperforms the newer Intel part in all shared benchmarks.

Architecture Differences

The two processors are built on different architectures from different generations. The Intel Core Ultra 5 238V uses the Lunar Lake architecture, part of the Core Ultra Series 2, fabricated on a 3 nm process by TSMC. The AMD EPYC 9554P uses the Zen 4 architecture, codenamed Genoa, fabricated on a 5 nm process, also by TSMC. Both use TSMC as the foundry, but the process nodes differ: 3 nm for Intel versus 5 nm for AMD.

Cache configurations diverge significantly. The Core Ultra has 192 KB of L1 cache per core, 2.5 MB of L2 cache per core, and 8 MB of shared L3 cache. The EPYC has 64 KB of L1 per core, 1 MB of L2 per core, and a massive 256 MB of shared L3 cache. The EPYC’s 256 MB L3 is 32 times larger than the Core Ultra’s 8 MB, a design choice aimed at server workloads with large working sets. The EPYC also has 52,560 million transistors spread across 8 dies of 72 mm² each, while the Core Ultra’s transistor count and die size are not listed.

Integrated graphics differ as well. The Core Ultra includes Arc 130V integrated graphics, while the EPYC has no integrated graphics listed. This aligns with their market segments: mobile processors often need iGPUs for display output, while server processors typically rely on discrete graphics or no graphics at all. The EPYC supports ECC memory, while the Core Ultra does not, reflecting the EPYC’s server-grade reliability requirements.

Specification Differences

The specification sheets reveal stark contrasts in nearly every field. The Core Ultra has 8 cores and 8 threads, while the EPYC has 64 cores and 128 threads — an 8x core count advantage and 16x thread count advantage for AMD. Base clocks are 2.10 GHz for Intel and 3.10 GHz for AMD; boost clocks are 4.70 GHz for Intel and 3.75 GHz for AMD. The Intel part boosts higher, but the AMD part has a higher base clock.

TDP is a major differentiator: 17W for the Core Ultra versus 360W for the EPYC. Socket types are incompatible: Intel BGA 2833 for the Core Ultra versus AMD Socket SP5 for the EPYC. Memory support shows the EPYC with DDR5 and twelve-channel bus, while the Core Ultra’s memory support is listed as unknown and dependent on the motherboard, with a dual-channel bus. The EPYC’s memory bandwidth is 460.8 GB/s, while the Core Ultra’s is not listed. ECC memory is supported on the EPYC but not on the Core Ultra.

PCIe connectivity differs by an order of magnitude: Gen 5 with 4 lanes for the Core Ultra versus Gen 5 with 128 lanes for the EPYC. The EPYC has a launch MSRP of $7104, while the Core Ultra has no launch MSRP listed. The EPYC’s part number is 100-100000804, and the Core Ultra’s is SRPN5SRPN4. Both have locked multipliers, meaning neither supports overclocking. The EPYC’s market segment is Server/Workstation, while the Core Ultra’s is Mobile. Both are Active in production status.

The Verdict

The benchmark data is clear: the AMD EPYC 9554P outperforms the Intel Core Ultra 5 238V in every shared test, with a consistent 82.5% margin. This is not a close contest — the EPYC’s 64-core, 128-thread configuration, 256 MB L3 cache, and twelve-channel DDR5 memory make it overwhelmingly superior for multi-threaded workloads. The single-core results also favor the EPYC, despite its lower boost clock, indicating a fundamental architectural advantage for Zen 4 in these benchmarks.

However, the verdict must account for intended use. The Core Ultra’s 17W TDP, integrated Arc 130V graphics, and mobile socket (Intel BGA 2833) position it for laptops where battery life and portability are paramount. The EPYC’s 360W TDP, 128 PCIe lanes, and server socket (AMD Socket SP5) make it a datacenter workhorse. The Core Ultra’s average benchmark score is 21,981, nearly identical to the EPYC’s 21,899, and both sit at the 75th percentile of all CPUs — but this aggregate metric masks the vast difference in workload suitability.

For anyone building a high-performance server or workstation, the EPYC 9554P is the clear choice from the data. For anyone building a thin-and-light laptop, the Core Ultra 5 238V is the only viable option between these two, given the EPYC’s physical requirements (360W TDP, SP5 socket) make it impossible to install in a mobile chassis. The data does not support a scenario where the Core Ultra wins a benchmark, but it also does not support the EPYC being used in mobile applications.

Where Each One Wins

AMD EPYC 9554P: This processor wins every benchmark category in the head-to-head data. Its strengths lie in multi-threaded workloads: Cinebench R15, R20, and R23 multi-core tests all show an 82.5% advantage over the Core Ultra. The 256 MB L3 cache, 64 cores, 128 threads, and twelve-channel DDR5 memory with 460.8 GB/s bandwidth make it ideal for server virtualization, database processing, scientific computing, and any workload that scales with core count. It also wins single-core tests, suggesting its Zen 4 architecture is more efficient per clock than Lunar Lake in Cinebench workloads. The 128 PCIe Gen 5 lanes support massive I/O configurations for storage arrays and network interfaces. ECC memory support adds reliability for long-running server operations.

Intel Core Ultra 5 238V: While the Core Ultra wins no benchmark comparisons, its architectural features provide advantages in specific use cases not captured by the head-to-head data. The 17W TDP makes it suitable for fanless or low-power mobile designs where the EPYC’s 360W TDP is impossible. Integrated Arc 130V graphics eliminate the need for a discrete GPU in basic display tasks, a feature the EPYC lacks entirely. The 3 nm process node suggests better transistor density and potentially better power efficiency per operation, though no efficiency benchmarks are in the data. Its 4.70 GHz boost clock is higher than the EPYC’s 3.75 GHz, which could benefit bursty single-threaded tasks, though the benchmark data does not reflect this advantage. The mobile form factor (Intel BGA 2833 socket) makes it the only choice for laptops, tablets, or compact embedded systems. For users prioritizing battery life, portability, and integrated graphics over raw compute performance, the Core Ultra 5 238V is the pragmatic selection.