AMD Ryzen Threadripper PRO 9965WX vs Intel Xeon 6745P Comparison

The Intel Xeon 6745P and AMD Ryzen Threadripper PRO 9965WX are both flagship-class workstation processors, but benchmark data shows they are not interchangeable. The AMD part takes a commanding lead in the majority of tests, winning 13 of 16 comparisons, while the Intel Xeon counters with a dominant win in floating-point math and narrow victories in data compression and encryption. The Ryzen Threadripper PRO 9965WX is the faster all-around processor, but the Xeon 6745P holds specific advantages that matter for certain workloads.

Head-to-Head Benchmarks

The most striking result is in single-thread performance, where the AMD Ryzen Threadripper PRO 9965WX leads by a massive margin. In PassMark single-thread tests, AMD scores 4551 against Intel’s 3450, a 24.2% advantage. This is the largest delta in the entire comparison and highlights the architectural efficiency of the Zen 5 core. Cinebench single-core results follow the same pattern, with AMD winning R15 (1152 vs 1018), R20 (4801 vs 4244), and R23 (11431 vs 4244 — the Xeon’s R23 single-core score is not provided separately, but the delta is consistent at -11.6% across all Cinebench single-core tests).

Multi-threaded workloads also favor AMD, though by a narrower margin. In Cinebench R23 multi-core, the Ryzen Threadripper PRO 9965WX scores 80976 versus 71578 for the Xeon, an 11.6% advantage. The same 11.6% delta appears across Cinebench R15 multi-core (8162 vs 7214) and R20 multi-core (34009 vs 30062), indicating a consistent generational performance gap in render workloads. PassMark multithread shows AMD ahead by 9.1% (92604 vs 84210), and the physics test amplifies this further with an 18.4% lead (7529 vs 6144).

The Xeon 6745P’s best result is in floating-point math, where it scores 267438 against AMD’s 229685, a commanding 16.4% victory. This is a decisive win that suggests the Intel architecture handles FP-heavy calculations more efficiently. The Xeon also edges out AMD in data compression (1352801 vs 1345230, +0.6%) and data encryption (66665 vs 66155, +0.8%), though these margins are close to statistical noise. AMD wins the remaining tests: extended instructions (+0.4%), find prime numbers (+9.4%), integer math (+3.5%), and random string sorting (+10.8%).

Architecture Differences

The two processors represent fundamentally different design philosophies. The Intel Xeon 6745P is built on Granite Rapids architecture using a 5 nm process from Intel’s own foundry, with a die size of 2x 598 mm². It packs 32 cores and 64 threads, with a base clock of 3.10 GHz and boost clock of 4.30 GHz. Cache is generous: 112 KB L1 per core, 2 MB L2 per core, and 336 MB shared L3. The AMD Ryzen Threadripper PRO 9965WX uses Zen 5 architecture (codenamed Shimada Peak) on TSMC’s 4 nm process, with a smaller combined die area of 4x 70.6 mm² and 33,260 million transistors. It offers 24 cores and 48 threads at higher clocks: 4.20 GHz base and 5.40 GHz boost. AMD’s cache is smaller at 64 KB L1 per core, 1 MB L2 per core, and 128 MB L3.

Memory support is identical on paper: both use DDR5 with an eight-channel bus and 409.6 GB/s bandwidth, and both support ECC memory. PCIe connectivity differs significantly — Intel provides 88 Gen 5 lanes (CPU only), while AMD offers 128 Gen 5 lanes. Neither has integrated graphics, and both target the server/workstation segment. The Xeon has a TDP of 300 W, while the Threadripper draws 350 W. The Intel part launches at $5250 MSRP, while AMD’s launch MSRP is $2899. The Xeon’s multiplier is locked; AMD’s is unlocked. The AMD chip is part of the 9000 series, while the Xeon belongs to the Xeon 6 (Granite Rapids-SP) generation.

Where Each One Wins

The Intel Xeon 6745P is the clear choice for floating-point-intensive workloads. Its 16.4% lead in floating-point math is substantial, and it also wins in data compression and encryption, making it suitable for scientific computing, financial modeling, and any application that relies heavily on FP operations or secure data handling. The 336 MB L3 cache and 32 cores give it raw throughput for parallel FP tasks, even if its per-core speed is lower.

The AMD Ryzen Threadripper PRO 9965WX wins everywhere else. The 24.2% single-thread advantage makes it superior for responsive, lightly-threaded applications like code compilation, interactive design tools, and database queries that depend on fast single-core execution. Its multi-threaded wins in Cinebench and PassMark multithread indicate better scaling in render farms, video encoding, and physics simulations. The 128 PCIe lanes also make it more suitable for systems with multiple GPUs or high-speed storage arrays. For users who need both strong single-core and multi-core performance, AMD is the better fit.

FAQ

Q: Which processor has more cores?

A: The Intel Xeon 6745P has 32 cores and 64 threads, while the AMD Ryzen Threadripper PRO 9965WX has 24 cores and 48 threads.

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

A: AMD leads by 24.2% in PassMark single-thread tests (4551 vs 3450), and the delta is -11.6% across all Cinebench single-core tests in favor of AMD.

Q: Does the Intel Xeon win any benchmark tests?

A: Yes, the Xeon wins 3 of 16 tests: floating-point math by 16.4%, data compression by 0.6%, and data encryption by 0.8%.

Q: Which chip has a higher boost clock?

A: The AMD Ryzen Threadripper PRO 9965WX boosts to 5.40 GHz, while the Intel Xeon 6745P boosts to 4.30 GHz.

Q: What is the difference in PCIe lane counts?

A: The AMD processor provides 128 Gen 5 lanes, while the Intel processor provides 88 Gen 5 lanes (CPU only).

Q: Do both processors support ECC memory?

A: Yes, both the Intel Xeon 6745P and AMD Ryzen Threadripper PRO 9965WX support ECC memory with an eight-channel DDR5 bus.

The Verdict

The data is unambiguous: the AMD Ryzen Threadripper PRO 9965WX is the superior processor for the majority of workloads. It wins 13 of 16 head-to-head benchmarks, including all Cinebench multi-core and single-core tests, PassMark multithread, physics, integer math, and single-thread tests. The 11.6% lead across all Cinebench tests is consistent and meaningful, indicating that the Zen 5 architecture delivers better performance per clock and better multi-threaded scaling than Granite Rapids.

The Intel Xeon 6745P is not without merit. Its 16.4% victory in floating-point math is the largest single-test win in the entire comparison, and its wins in data compression and encryption make it a niche pick for FP-heavy scientific workloads. However, these three wins are outweighed by AMD’s sweeping dominance elsewhere. The Xeon’s 32 cores and 336 MB L3 cache do not translate into broader benchmark success.

For most users, the AMD Ryzen Threadripper PRO 9965WX is the clear recommendation. It offers better single-core performance, better multi-core performance in most tests, higher clock speeds, more PCIe lanes, a smaller process node, and an unlocked multiplier. The Intel Xeon 6745P should only be chosen if floating-point math, data compression, or encryption are the primary workloads, and even then, the margin in those specific tests is modest compared to AMD’s overall lead.

Specification Differences

| Feature | Intel Xeon 6745P | AMD Ryzen Threadripper PRO 9965WX |

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

| Cores | 32 | 24 |

| Threads | 64 | 48 |

| Base Clock | 3.10 GHz | 4.20 GHz |

| Boost Clock | 4.30 GHz | 5.40 GHz |

| TDP | 300 W | 350 W |

| Socket | Intel Socket 4710 | AMD Socket sTR5 |

| Architecture | Granite Rapids | Zen 5 (Shimada Peak) |

| Process Node | 5 nm (Intel) | 4 nm (TSMC) |

| Die Size | 2x 598 mm² | 4x 70.6 mm² |

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

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

| L3 Cache | 336 MB (shared) | 128 MB |

| PCIe | Gen 5, 88 Lanes (CPU only) | Gen 5, 128 Lanes (CPU only) |

| Multiplier Unlocked | No | Yes |

| Launch MSRP | $5250 | $2899 |