AMD EPYC 9655P vs Intel Xeon 6780E Comparison

The Verdict

The benchmark database presents a decisive outcome in this comparison. The AMD EPYC 9655P wins all 16 recorded head-to-head tests, with zero wins for the Intel Xeon 6780E. The performance gap is not marginal; it is structural. Across Cinebench releases and Passmark workloads, the AMD part leads by margins ranging from 14% to 138.1%. The average benchmark score for the EPYC 9655P is 397,773, placing it at the 100th percentile of all CPUs, while the Xeon 6780E averages 280,438 and sits at the 99th percentile.

The AMD EPYC 9655P is the choice for workloads where raw throughput and single-thread responsiveness both matter. Its closest rival in the database is the AMD Ryzen Threadripper PRO 9995WX, which scores 3.5% higher on average, and the AMD EPYC 9745, which leads by 6.6%. The Xeon 6780E, by contrast, is closest to the AMD Ryzen Threadripper 9970X, with a 0.2% difference, and trails the AMD EPYC 9565 by 1.8%. The data suggests that the Xeon 6780E competes in a lower performance tier, not against the 9655P.

For buyers needing a server processor with 144 cores and 144 threads, the Xeon 6780E offers a high core count but at significantly lower clocks and cache capacity. However, the benchmark results show that core count alone does not compensate for the architectural and clock disadvantages. The EPYC 9655P delivers 96 cores with 192 threads, and its boost clock of 4.50 GHz versus 3.00 GHz on the Xeon explains much of the single-thread dominance. In multi-threaded tests, the EPYC 9655P still wins by 85% in Cinebench R23 multicore, despite having fewer physical cores.

The verdict from the data is unambiguous: the AMD EPYC 9655P is the superior processor in every measured dimension. The Intel Xeon 6780E is a viable alternative only for scenarios where its specific platform features, such as 144 cores on an eight-channel memory bus, are required, but it will underperform in all benchmark categories recorded here.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD EPYC 9655P averages 397,773 points, while the Intel Xeon 6780E averages 280,438 points. The EPYC 9655P also sits at the 100th percentile of all CPUs, compared to the 99th percentile for the Xeon.

Q: How large is the single-thread performance gap?

A: In Cinebench R23 single-core, the EPYC 9655P scores 19,250 versus 7,373 for the Xeon 6780E, a delta of 85%. In Passmark single-thread, the gap is 100.2%, with scores of 3,849 and 1,923, respectively.

Q: Does the Intel Xeon 6780E win any benchmark tests?

A: No. The database records 16 head-to-head tests, and the AMD EPYC 9655P wins all 16. The Intel Xeon 6780E records zero wins.

Q: What are the core and thread counts for each processor?

A: The AMD EPYC 9655P has 96 cores and 192 threads. The Intel Xeon 6780E has 144 cores and 144 threads. The Xeon has more cores but no simultaneous multithreading.

Q: How do their memory bandwidth figures compare?

A: The AMD EPYC 9655P supports twelve-channel DDR5 with a memory bandwidth of 576.0 GB/s. The Intel Xeon 6780E supports eight-channel DDR5 with a memory bandwidth of 409.6 GB/s.

Q: Which processor has a higher boost clock?

A: The AMD EPYC 9655P boosts to 4.50 GHz, while the Intel Xeon 6780E boosts to 3.00 GHz. The base clocks are 2.60 GHz for the EPYC and 2.20 GHz for the Xeon.

Architecture Differences

The AMD EPYC 9655P is built on the Zen 5 architecture, codenamed Turin, using a 4 nm process at TSMC. The Intel Xeon 6780E uses the Sierra Forest architecture, also its codename, on Intel's 5 nm process. The process node difference is a full nanometer, which contributes to the EPYC's higher clock speeds and efficiency. The EPYC 9655P contains 99,780 million transistors across a die size of 12x 70.6 mm², a chiplet design. The Xeon 6780E has a monolithic die of 578 mm², and its transistor count is not recorded in the database.

Cache organization differs substantially. The EPYC 9655P provides 80 KB of L1 per core, 1 MB of L2 per core, and a shared 384 MB L3 cache. The Xeon 6780E provides 96 KB of L1 per core, 4 MB of L2 per module, and a shared 108 MB L3 cache. The EPYC's L3 cache is over three times larger, which benefits workloads with large working sets. The Xeon's L2 is organized per module rather than per core, a design choice for its efficiency-focused Sierra Forest cores.

Memory architecture also diverges. The EPYC 9655P uses a twelve-channel memory bus, while the Xeon 6780E uses an eight-channel bus. This results in a memory bandwidth advantage of 576.0 GB/s versus 409.6 GB/s for the Xeon. Both support DDR5 and ECC memory. PCIe connectivity also differs: the EPYC 9655P provides Gen 5 with 128 lanes, while the Xeon 6780E provides Gen 5 with 88 lanes. Neither has integrated graphics.

The socket and platform are entirely different. The EPYC 9655P uses AMD Socket SP5, while the Xeon 6780E uses Intel Socket 4710. The Xeon 6780E was released earlier, on June 2, 2024, while the EPYC 9655P followed on October 9, 2024. The production status for both is active. Neither processor has an unlocked multiplier.

Specification Differences

| Field | AMD EPYC 9655P | Intel Xeon 6780E |

|---|---|---|

| Cores | 96 | 144 |

| Threads | 192 | 144 |

| Base Clock | 2.60 GHz | 2.20 GHz |

| Boost Clock | 4.50 GHz | 3.00 GHz |

| TDP | 400 W | 330 W |

| Socket | AMD Socket SP5 | Intel Socket 4710 |

| Process Node | 4 nm | 5 nm |

| Foundry | TSMC | Intel |

| Transistors | 99,780 million | Not recorded |

| Die Size | 12x 70.6 mm² | 578 mm² |

| L1 Cache | 80 KB (per core) | 96 KB (per core) |

| L2 Cache | 1 MB (per core) | 4 MB (per module) |

| L3 Cache | 384 MB (shared) | 108 MB (shared) |

| Memory Bus | Twelve-channel | Eight-channel |

| Memory Bandwidth | 576.0 GB/s | 409.6 GB/s |

| PCIe | Gen 5, 128 Lanes | Gen 5, 88 Lanes |

| Release Date | 2024-10-09 | 2024-06-02 |

| Launch MSRP | $10811 | $11350 |

| Part Number | 100-000001522 | SRPG3 |

The table highlights that the Xeon 6780E has more cores (144 vs. 96) but fewer threads (144 vs. 192), indicating the lack of SMT. The EPYC 9655P has higher clocks, a smaller process node, a much larger L3 cache, and greater memory bandwidth. The Xeon has a higher TDP at 400 W, but the EPYC 9655P consumes 330 W. Both support DDR5 and ECC, and both are server/workstation segments.

Head-to-Head Benchmarks

The largest single advantage for the AMD EPYC 9655P appears in Passmark prime number finding, where it scores 1,686 versus 708, a delta of 138.1%. This workload is highly dependent on integer arithmetic and clock speed, both of which favor the EPYC. Similarly, in Passmark physics, the EPYC leads by 136%, with scores of 25,847 and 10,951. These results indicate that the Zen 5 architecture excels at latency-sensitive, high-frequency operations.

Cinebench results are consistently one-sided. In Cinebench R15 multicore, the EPYC 9655P scores 13,744 versus 7,431, a 85% lead. The single-core test shows 1,940 versus 1,049, a 84.9% lead. Cinebench R20 repeats this pattern: multicore 57,268 versus 30,963, single-core 8,085 versus 4,371, both at 85% deltas. Cinebench R23 multicore shows 136,354 versus 73,723, an 85% lead, and single-core shows 19,250 versus 7,373, also 85%. The consistency across Cinebench versions suggests a stable architectural advantage.

Passmark integer math shows a 90.9% lead, with scores of 1,225,251 and 641,817. Floating point math shows a 65% lead, with 715,866 versus 433,862. Extended instructions show a 102.3% lead, with 230,609 versus 114,008. These are substantial margins that point to the EPYC's superior execution resources and higher clock speeds.

Data compression results show a 36.3% lead, with 3,486,158 versus 2,557,582. Random string sorting shows a 38.2% lead, with 451,824 versus 326,954. Data encryption shows the smallest margin at 14%, with 220,074 versus 193,004. Even the narrowest win is still a clear victory, indicating that the Xeon 6780E cannot match the EPYC in any recorded workload.

Passmark multithread scores are 160,490 for the EPYC and 86,734 for the Xeon, an 85% lead. This result is telling because the Xeon has 144 cores versus 96 cores, yet the EPYC still outperforms by a wide margin. The data shows that the EPYC's 192 threads and higher clocks overcome the Xeon's core count advantage. The nearest rivals for the EPYC 9655P, such as the AMD EPYC 9745, score 6.6% higher on average, while the Xeon 6780E's closest competitor, the AMD Ryzen Threadripper 9970X, is only 0.2% different. This places the Xeon in a different competitive bracket entirely.