AMD EPYC 7551P vs Intel Core i9-10980XE Comparison

The AMD EPYC 7551P and Intel Core i9-10980XE present a fascinating study in contrasting design philosophies. The data shows a decisive split: the EPYC 7551P dominates in rendering workloads, while the Intel part takes a commanding lead in Geekbench tests. This comparison is not about a single victor but about which processor's strengths align with the specific demands of a given task.

Where Each One Wins

The benchmark results carve out two distinct territories. The AMD EPYC 7551P is the clear winner in every Cinebench test, which are heavily threaded, compute-intensive rendering tasks. It wins the multi-core and single-core variants of Cinebench R15, R20, and R23, securing 6 out of 8 possible benchmark wins. This suggests its 32-core, 64-thread configuration provides a substantial advantage in workloads that scale with core count and raw throughput.

Conversely, the Intel Core i9-10980XE wins both Geekbench tests, which are often more representative of general-purpose computing, memory latency-sensitive tasks, and lightly-threaded applications. Its wins in Geekbench multi-core and single-core, with deltas of -57.3% and -36.8% respectively, are substantial. This indicates that the Intel architecture, with its higher clock speeds, provides superior performance in tasks that depend on single-thread efficiency and lower latency, rather than sheer core count.

The data implies that the EPYC 7551P is better suited for professional content creation, 3D rendering, and scientific computing where render times are paramount. The Core i9-10980XE, on the other hand, appears better suited for desktop workloads, software development, and applications that are not perfectly optimized for massive core counts but benefit from high per-core performance. The Geekbench single-core score of 1453 for Intel versus 919 for AMD highlights a significant gap in tasks that rely on a single processing thread.

Architecture Differences

The fundamental differences between these two processors begin with their core counts and market intent. The AMD EPYC 7551P is a server-class part with 32 cores and 64 threads, built on the Zen architecture (codenamed "Naples"). It uses a 14 nm process from GlobalFoundries and is designed for the AMD Socket SP3 platform. In contrast, the Intel Core i9-10980XE is a desktop extreme processor with 18 cores and 36 threads, based on the Cascade Lake-X architecture. It also uses a 14 nm process, but from Intel's own fabs, and fits the Intel Socket 2066.

Memory architecture is another major differentiator. The EPYC 7551P supports eight-channel DDR4 memory, yielding a massive memory bandwidth of 170.6 GB/s and supporting ECC memory. The Core i9-10980XE supports only quad-channel DDR4, with a memory bandwidth of 93.9 GB/s and no ECC support. This 76.7 GB/s difference in bandwidth is a likely factor in the EPYC's dominance in multi-threaded rendering, where data throughput is critical.

Cache hierarchies also differ significantly. The EPYC 7551P has a large 64 MB shared L3 cache, while the Intel part has a 24.75 MB shared L3 cache. However, the Intel processor has a larger 1 MB L2 cache per core compared to the EPYC's 512 KB per core. The EPYC also has a larger L1 cache per core at 96 KB versus 64 KB. The EPYC 7551P is unlocked for overclocking, and its production status is "Active," while the Intel part is also unlocked but is marked as "End-of-life."

Head-to-Head Benchmarks

The head-to-head data is remarkable for its consistency. In Cinebench R15 multi-core, the AMD EPYC 7551P scores 3276 against Intel's 2734, a 19.8% advantage. This pattern repeats almost exactly in Cinebench R20 multi-core, where the EPYC scores 13650 versus 11394, and in Cinebench R23 multi-core, with a score of 32500 versus 27130. Each of these multi-core wins is accompanied by a nearly identical 19.8% delta, suggesting a consistent scaling advantage for the EPYC in this workload.

The single-core Cinebench results are even more surprising. The EPYC 7551P wins Cinebench R15, R20, and R23 single-core tests with scores of 462, 1926, and 4588, respectively, against Intel's 385, 1608, and 3830. The delta is approximately 20% in each case. This is unexpected, as Intel processors typically have higher clock speeds; the Intel part has a base clock of 3.00 GHz and a boost of 4.80 GHz, versus the EPYC's 2.00 GHz base and 3.00 GHz boost. The data suggests the Zen architecture's efficiency and the larger L3 cache compensate for the lower clock speed in these specific Cinebench workloads.

The Geekbench results reverse the trend with even greater magnitude. Intel wins Geekbench multi-core with a score of 14742 against AMD's 6296, a 57.3% difference. In Geekbench single-core, Intel scores 1453 against 919, a 36.8% advantage. These are massive swings, indicating that the Core i9-10980XE's higher clock speeds and architectural efficiency for latency-sensitive tasks provide a decisive edge in this benchmark suite. The data shows a complete inversion of performance leadership depending on the benchmark used.

The Verdict

The data presents a clear, use-case-driven verdict. For workloads primarily built around Cinebench-style rendering, the AMD EPYC 7551P is the superior choice. Its consistent 19.8% lead across all Cinebench iterations demonstrates a robust advantage in multi-threaded and even single-threaded rendering tasks. The EPYC's 64 MB L3 cache and eight-channel memory bandwidth appear to be decisive factors for this workload.

For users whose primary workload is better represented by Geekbench, the Intel Core i9-10980XE is the clear winner. Its 57.3% lead in multi-core and 36.8% lead in single-core Geekbench tests are substantial. This performance profile is likely better for general desktop use, gaming (where single-thread performance is critical), and applications that favor lower memory latency over raw bandwidth.

The average benchmark scores are nearly identical, with the EPYC at 7952 and the Intel at 7910, a difference of only 0.5%. Both processors sit at the 64th percentile of all CPUs. This means that on average, they are equivalent performers, but the types of workloads that contribute to that average are entirely different. The choice is not about which is faster, but which is faster for the specific tasks the system will handle.

FAQ

Q: Which processor is faster in Cinebench R23 multi-core?

A: The AMD EPYC 7551P is significantly faster, scoring 32500 compared to the Intel Core i9-10980XE's 27130, a 19.8% difference.

Q: Does the Intel Core i9-10980XE win any benchmarks?

A: Yes, it wins both Geekbench tests. It scores 14742 in multi-core and 1453 in single-core, beating the EPYC 7551P's scores of 6296 and 919, respectively.

Q: What is the difference in memory bandwidth between the two?

A: The AMD EPYC 7551P supports eight-channel memory with a bandwidth of 170.6 GB/s, while the Intel Core i9-10980XE supports quad-channel memory with a bandwidth of 93.9 GB/s.

Q: How do their core counts compare?

A: The AMD EPYC 7551P has 32 cores and 64 threads, while the Intel Core i9-10980XE has 18 cores and 36 threads.

Q: Which processor supports ECC memory?

A: The AMD EPYC 7551P supports ECC memory, but the Intel Core i9-10980XE does not.

Q: What is the average benchmark score for each?

A: The AMD EPYC 7551P has an average score of 7952, while the Intel Core i9-10980XE has an average score of 7910.

Specification Differences

| Specification | AMD EPYC 7551P | Intel Core i9-10980XE |

| :--- | :--- | :--- |

| Cores | 32 | 18 |

| Threads | 64 | 36 |

| Base Clock | 2.00 GHz | 3.00 GHz |

| Boost Clock | 3.00 GHz | 4.80 GHz |

| TDP | 180 W | 165 W |

| Socket | AMD Socket SP3 | Intel Socket 2066 |

| Architecture | Zen | Cascade Lake |

| Codename | Naples | Cascade Lake-X |

| Process Node | 14 nm | 14 nm |

| Foundry | GlobalFoundries | Intel |

| L1 Cache | 96 KB (per core) | 64 KB (per core) |

| L2 Cache | 512 KB (per core) | 1 MB (per core) |

| L3 Cache | 64 MB (shared) | 24.75 MB (shared) |

| Memory Bus | Eight-channel | Quad-channel |

| Memory Bandwidth | 170.6 GB/s | 93.9 GB/s |

| ECC Memory | Yes | No |

| PCIe | Gen 3 | Gen 3, 48 Lanes (CPU only) |

| Production Status | Active | End-of-life |

| Release Date | 2017-06-28 | 2019-10-18 |