AMD EPYC 7C13 vs Intel Xeon 6747P Comparison

The Verdict

The Intel Xeon 6747P and AMD EPYC 7C13 occupy different positions in the server landscape, and the recorded data makes the separation clear. The Intel Xeon 6747P wins 12 of the 14 head-to-head benchmark comparisons, while the AMD EPYC 7C13 takes only 2. The average benchmark score for the Intel part is 238263, which places it in the 99th percentile of all CPUs. The AMD EPYC 7C13 averages 167788, sitting in the 98th percentile. The Intel Xeon 6747P is the outright performance leader in nearly every measured workload, often by substantial margins. The AMD EPYC 7C13 still holds specific wins in data encryption and integer math, and its lower power envelope makes it a legitimate consideration for power-conscious deployments. Ultimately, the Xeon 6747P is the choice for raw compute density and single-thread responsiveness, while the EPYC 7C13 suits workloads that favor its encryption throughput and integer performance with a much lower TDP.

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Xeon 6747P uses the Granite Rapids architecture, built on a 5 nm process at Intel, with a die size of 2x 598 mm². The AMD EPYC 7C13 uses the Zen 3 architecture, codenamed Milan, fabricated by TSMC on a 7 nm process, with a die configuration of 8x 81 mm² and 33,200 million transistors. Core counts differ significantly: the Xeon packs 48 cores and 96 threads, while the EPYC offers 64 cores and 128 threads. The Xeon fights back with higher clock speeds, a base clock of 2.70 GHz and a boost of 3.90 GHz, versus the EPYC's base of 2000.00 MHz and boost of 3.68 GHz.

Cache hierarchies also diverge. The Intel part allocates 112 KB of L1 per core, 2 MB of L2 per core, and a massive 288 MB of shared L3 cache. The AMD part uses 64 KB of L1 per core, 512 KB of L2 per core, and 256 MB of shared L3. Memory support is a generational split: the Xeon uses DDR5 over an eight-channel bus with 409.6 GB/s of bandwidth, while the EPYC uses DDR4 over an eight-channel bus with 204.8 GB/s. PCIe connectivity favors the EPYC in lane count, with Gen 4 and 128 lanes, versus the Xeon's Gen 5 and 88 lanes. Both support ECC memory. The Xeon's launch MSRP is $6497; the EPYC has no recorded launch MSRP.

FAQ

Q: Which processor has more cores?

A: The AMD EPYC 7C13 has 64 cores and 128 threads, compared to the Intel Xeon 6747P's 48 cores and 96 threads.

Q: Which processor is faster in single-threaded workloads?

A: The Intel Xeon 6747P leads in single-thread performance, scoring 3236 in PassMark single-thread tests versus 2618 for the AMD EPYC 7C13, a 23.6% advantage.

Q: How do the two compare in multi-threaded Cinebench tests?

A: The Intel Xeon 6747P wins all three multi-core Cinebench tests by 33.2%, including a Cinebench R23 multi-core score of 86432 versus 64873 for the AMD EPYC 7C13.

Q: Does the AMD EPYC 7C13 win any benchmarks?

A: Yes, it wins two head-to-head tests: PassMark data encryption with 114769 versus 90789 (a 20.9% margin for AMD), and PassMark integer math with 492554 versus 468518 (a 4.9% margin).

Q: What is the memory bandwidth difference?

A: The Intel Xeon 6747P supports DDR5 with 409.6 GB/s of bandwidth, exactly double the 204.8 GB/s of the DDR4-based AMD EPYC 7C13.

Q: Which processor has a higher TDP?

A: The Intel Xeon 6747P has a TDP of 330 watts, while the AMD EPYC 7C13 has a TDP of 225 watts.

Specification Differences

| Specification | Intel Xeon 6747P | AMD EPYC 7C13 |

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

| Cores | 48 | 64 |

| Threads | 96 | 128 |

| Base Clock | 2.70 GHz | 2000.00 MHz |

| Boost Clock | 3.90 GHz | 3.68 GHz |

| TDP | 330 W | 225 W |

| Socket | Intel Socket 4710 | AMD Socket SP3 |

| Architecture | Granite Rapids | Zen 3 |

| Codename | Granite Rapids | Milan |

| Process Node | 5 nm | 7 nm |

| Foundry | Intel | TSMC |

| Transistors | Not recorded | 33,200 million |

| Die Size | 2x 598 mm² | 8x 81 mm² |

| L1 Cache | 112 KB per core | 64 KB per core |

| L2 Cache | 2 MB per core | 512 KB per core |

| L3 Cache | 288 MB shared | 256 MB shared |

| Memory Support | DDR5 | DDR4 |

| Memory Bandwidth | 409.6 GB/s | 204.8 GB/s |

| PCIe | Gen 5, 88 lanes | Gen 4, 128 lanes |

| Launch MSRP | $6497 | Not recorded |

Head-to-Head Benchmarks

The Intel Xeon 6747P dominates the benchmark matrix, but the margins vary dramatically by workload. The largest win for Intel comes in PassMark physics, where the Xeon scores 13398 against the EPYC's 4904, a 173.2% advantage. This suggests a massive difference in physics simulation throughput, likely stemming from the Xeon's higher boost clock and newer architecture. PassMark find prime numbers also heavily favors Intel, with the Xeon scoring 1151 versus 539, a 113.5% lead. These two tests alone show that Intel has a decisive edge in compute-heavy single-thread-sensitive workloads.

In extended instruction workloads, the Xeon 6747P scores 142557 against 85034 for the EPYC 7C13, a 67.6% margin. This indicates superior SIMD and vector processing capability, which matters for scientific computing and media encoding. Floating point math also goes to Intel, 365904 versus 266846, a 37.1% lead. Random string sorting follows a similar pattern, with the Xeon at 180382 versus 131361, a 37.3% advantage. These results consistently show Intel's architectural superiority in math and data manipulation tasks.

The Cinebench suite tells a uniform story. Across R15, R20, and R23 multi-core tests, the Intel Xeon 6747P wins by exactly 33.2% each time. The scores are 8712 versus 6539 in R15, 36301 versus 27246 in R20, and 86432 versus 64873 in R23. The consistency of the 33.2% delta across all three Cinebench versions suggests a stable multi-core performance advantage that scales predictably with the newer architecture. PassMark multithread mirrors this, with Intel at 101685 versus 76322, again a 33.2% gap.

Data compression favors the Xeon 6747P, with a score of 1833378 versus 1562251 for the EPYC, a 17.4% lead. This is a smaller margin than the Cinebench results, indicating that the EPYC's extra cores help narrow the gap in memory-bound compression workloads. Single-thread performance goes to Intel by 23.6%, with the Xeon scoring 3236 versus 2618 in both PassMark single-thread and singlethread tests.

The AMD EPYC 7C13's two wins are notable for their specificity. Data encryption shows the EPYC at 114769 versus 90789 for Intel, a 20.9% margin. This is the only test where AMD wins by more than a small amount. Integer math is closer, with the EPYC scoring 492554 versus 468518, a 4.9% lead. These results suggest that AMD's Zen 3 architecture retains strengths in cryptographic operations and integer-heavy code, even as it loses ground in floating point, physics, and instruction-heavy workloads.

The overall benchmark picture places the Intel Xeon 6747P in the 99th percentile of all CPUs with an average score of 238263, while the AMD EPYC 7C13 sits at the 98th percentile with 167788. The Xeon's nearest rivals include the AMD EPYC 9634 at 244274 (2.5% ahead), the Intel Xeon 6980P at 251516 (5.3% ahead), and the AMD EPYC 9455P at 217854 (9.4% behind). The EPYC 7C13's nearest rivals are the AMD Ryzen Threadripper PRO 3995WX at 171748 (2.3% ahead), the AMD EPYC 9375F at 162497 (3.3% behind), and the AMD EPYC 7663 at 161973 (3.6% behind). These comparisons show that the Xeon 6747P competes in a higher performance tier, while the EPYC 7C13 holds its own against older and newer AMD parts in its class.