AMD EPYC 9175F vs Intel Xeon 6736P Comparison

Where Each One Wins

The benchmark split is decisive but not absolute. The AMD EPYC 9175F wins 14 of 17 head-to-head comparisons, while the Intel Xeon 6736P takes 3. The AMD part's dominance is most pronounced in single-threaded workloads, Cinebench rendering, and multithreaded PassMark tests. The Intel chip, meanwhile, claims victories in data encryption, floating-point math, and random string sorting, indicating specific strength in those compute patterns.

For rendering and simulation workloads, the EPYC 9175F is the clear choice. Every Cinebench iteration, from R15 through R23, shows the AMD processor ahead by a uniform 31.4% in both single-core and multi-core runs. The consistency of that margin across all six Cinebench tests suggests an architectural advantage that scales evenly regardless of thread count or workload intensity.

For encryption-heavy tasks, the Xeon 6736P leads. The Intel processor scores 46236 in PassMark data encryption versus 42297 for the AMD chip, an 8.5% advantage. This is the largest single victory for Intel outside of its two other wins, and it points to optimized cryptographic instruction paths in the Granite Rapids design.

Float-physics and sorting workloads reveal a narrower split. The Xeon wins floating-point math by 2.6% (149770 vs 145939) and random string sorting by 7.7% (103723 vs 95783). These are modest margins compared to the AMD wins, but they show that the Intel part retains competence in memory-sorting and FP-heavy operations.

The AMD EPYC 9175F also holds a commanding lead in prime-number finding, scoring 741 versus 392, an 89% advantage. That delta is the second-largest in the entire comparison, behind only single-thread performance. Extended instructions go to AMD by 26.9%, integer math to AMD by 6.3%, and data compression to AMD by 8.9%.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD EPYC 9175F uses Zen 5 architecture on a 4 nm TSMC process, while the Intel Xeon 6736P uses Granite Rapids on Intel's 5 nm node. The AMD chip is built from 16 compute dies, each measuring 70.6 mm², totaling 133,040 million transistors. The Intel die is a single 598 mm² piece.

Core counts diverge sharply. The Xeon 6736P packs 36 cores and 72 threads, more than double the EPYC's 16 cores and 32 threads. Yet the AMD chip compensates with far higher clock speeds: 4.20 GHz base and 5.00 GHz boost versus 2.00 GHz base and 4.10 GHz boost for Intel. The clock advantage, combined with the Zen 5 core design, explains why AMD wins most benchmarks despite being outnumbered 36 to 16.

Cache hierarchies reflect different strategies. The EPYC 9175F provides 80 KB of L1 per core, 1 MB of L2 per core, and a massive 512 MB shared L3 cache. The Xeon 6736P offers 112 KB L1 per core, 2 MB L2 per core, but only 144 MB of shared L3. The AMD chip's L3 cache is more than 3.5 times larger, which likely contributes to its dominance in compression and extended instruction workloads.

Memory channels also differ. The EPYC 9175F runs twelve-channel DDR5 with 576.0 GB/s bandwidth. The Xeon 6736P runs eight-channel DDR5 with 409.6 GB/s. That bandwidth gap of 40.6% favors AMD in memory-hungry tasks, though the Intel chip's sorting win suggests its memory controller handles certain access patterns efficiently despite lower peak bandwidth.

PCIe connectivity goes to AMD as well: 128 Gen 5 lanes versus 88 Gen 5 lanes. Both are server/workstation parts with ECC memory support and no integrated graphics. The AMD part draws 320 W TDP versus 205 W for Intel, a difference that reflects the higher clocks and larger cache. Release dates differ by roughly four months: the EPYC launched in October 2024, the Xeon in February 2025.

Head-to-Head Benchmarks

The single-thread results are dramatic. In PassMark single-thread, the EPYC 9175F scores 4256 against 2024 for the Xeon, a 110.3% advantage. That is the largest delta in the entire dataset and underscores the Zen 5 core's efficiency at high clock speeds. Cinebench R23 single-core confirms the pattern: 7895 for AMD versus 6008 for Intel, a 31.4% gap identical to every other Cinebench test.

Multi-core Cinebench R23 shows 55923 for the EPYC versus 42561 for the Xeon, again 31.4% ahead. This is notable because the Intel chip has 36 cores versus 16, yet still loses by nearly a third. The EPYC's per-core performance is so high that it overcomes a 20-core deficit in threaded workloads.

PassMark multithread tells a similar story but with a larger margin: 67634 for AMD versus 50072 for Intel, a 35.1% lead. Physics tests show the biggest multi-core gap: 9984 versus 6531, a 52.9% advantage for AMD. Prime-number finding, which stresses integer recursion and memory latency, is even more lopsided at 89% (741 vs 392).

The Intel wins are concentrated but real. Data encryption sees the Xeon score 46236 versus 42297, an 8.5% margin. Floating-point math is close: 149770 versus 145939, only 2.6% apart. Random string sorting gives Intel its clearest win at 7.7% (103723 vs 95783). These three results suggest that when the workload involves specialized instruction paths or memory-resident data reordering, the Granite Rapids architecture can still outpace Zen 5.

Data compression favors AMD at 867186 versus 796658, an 8.9% margin. Extended instructions go to AMD at 70529 versus 55563, a 26.9% gap. Integer math is closer, with AMD ahead 219800 to 206833, a 6.3% margin. The overall average benchmark score reflects this: the EPYC 9175F averages 95615 across all tests, while the Xeon 6736P averages 87864.

FAQ

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

A: The AMD EPYC 9175F is dramatically faster. PassMark single-thread shows 4256 for AMD versus 2024 for Intel, a 110.3% advantage. Cinebench R23 single-core confirms with 7895 versus 6008, a 31.4% lead.

Q: Does the Intel Xeon 6736P win any benchmarks?

A: Yes, it wins three of seventeen comparisons. These are data encryption (46236 vs 42297, an 8.5% lead), floating-point math (149770 vs 145939, a 2.6% lead), and random string sorting (103723 vs 95783, a 7.7% lead).

Q: How does the core count difference affect results?

A: The Xeon has 36 cores and 72 threads versus 16 cores and 32 threads for the EPYC. Despite this, AMD wins multi-core Cinebench R23 by 31.4% (55923 vs 42561) and PassMark multithread by 35.1% (67634 vs 50072). Higher clocks and superior per-core performance overcome the core-count deficit.

Q: What is the memory bandwidth difference?

A: The EPYC 9175F has twelve-channel DDR5 with 576.0 GB/s bandwidth, while the Xeon 6736P has eight-channel DDR5 with 409.6 GB/s. That is a 40.6% bandwidth advantage for AMD.

Q: Which processor has more cache?

A: The AMD EPYC 9175F has 512 MB of shared L3 cache, while the Intel Xeon 6736P has 144 MB. The AMD chip also has 80 KB L1 and 1 MB L2 per core, versus 112 KB L1 and 2 MB L2 per core for Intel.

Q: How do the average benchmark scores compare?

A: The EPYC 9175F averages 95615 across all recorded benchmarks, while the Xeon 6736P averages 87864. Both sit at the 96th percentile among all CPUs in the database.

Specification Differences

| Specification | AMD EPYC 9175F | Intel Xeon 6736P |

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

| Cores | 16 | 36 |

| Threads | 32 | 72 |

| Base clock | 4.20 GHz | 2.00 GHz |

| Boost clock | 5.00 GHz | 4.10 GHz |

| TDP | 320 W | 205 W |

| Socket | AMD Socket SP5 | Intel Socket 4710 |

| Process node | 4 nm | 5 nm |

| Foundry | TSMC | Intel |

| Die size | 16x 70.6 mm² | 598 mm² |

| Transistors | 133,040 million | Not listed |

| L1 cache | 80 KB (per core) | 112 KB (per core) |

| L2 cache | 1 MB (per core) | 2 MB (per core) |

| L3 cache | 512 MB (shared) | 144 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 | October 2024 | February 2025 |

The data shows a clean architectural split: AMD trades core count for clock speed and cache, while Intel provides more cores at lower frequencies with a smaller L3 pool.

The Verdict

Pick the AMD EPYC 9175F if the workload prioritizes single-thread performance, rendering, or general multithreaded compute. The data shows a 110.3% lead in PassMark single-thread, a 31.4% lead across all Cinebench tests, and a 35.1% lead in PassMark multithread. For simulation, compilation, or database workloads that rely on single-core responsiveness, the EPYC's 5.00 GHz boost clock and 512 MB L3 cache deliver results the Xeon cannot match.

Pick the Intel Xeon 6736P if the workload is dominated by encryption, floating-point math, or random string sorting. The Xeon wins those three specific tests, with an 8.5% edge in data encryption being the most pronounced. Its 36 cores and 72 threads may also be attractive for highly parallel, low-clock-tolerant workloads, though the benchmark data shows it loses multi-core Cinebench and PassMark multithread regardless.

The overall average benchmark score favors AMD, 95615 versus 87864. Both processors sit at the 96th percentile in the database, meaning either is a top-tier server part. But the EPYC 9175F delivers more performance per core, while the Xeon 6736P delivers more cores per watt, with its 205 W TDP being significantly lower than the AMD part's 320 W.

For most users, the AMD EPYC 9175F is the stronger processor. It wins 14 of 17 benchmarks, often by large margins, and its weaknesses are confined to three specialized workloads. The Intel Xeon 6736P is the better choice only for those specific encryption and sorting patterns where its architecture holds a measurable edge.