AMD EPYC 9355P vs AMD Ryzen Threadripper PRO 9965WX Comparison

The AMD EPYC 9355P and AMD Ryzen Threadripper PRO 9965WX are both Zen 5 powerhouses, yet the benchmark data reveals they are engineered for distinctly different priorities. The EPYC 9355P wins 14 of 17 head-to-head comparisons, leveraging its 32 cores to dominate multi-threaded workloads. However, the Threadripper PRO 9965WX counters with a decisive victory in single-threaded performance, showcasing that core count is not the only metric that matters. This analysis breaks down the numbers to determine which processor's strengths align with specific use cases.

Head-to-Head Benchmarks

The most lopsided result in the entire comparison appears in the PassMark physics test, where the EPYC 9355P scores 13,515 against the Threadripper's 7,529. That is a staggering 79.5% advantage, indicating the EPYC's 32-core configuration provides massive parallel throughput for physics simulations. Similarly, the PassMark find prime numbers test shows the EPYC at 1,044 versus 752, a 38.8% lead. These are not marginal differences; they represent fundamental scaling advantages in integer-heavy, parallelizable workloads.

The data encryption test further underscores the EPYC's dominance, with a score of 80,961 compared to 66,155 — a 22.4% gap. Integer math follows suit, where the EPYC scores 412,067 against 349,195, an 18% advantage. The random string sorting test also favors the EPYC, showing 176,697 versus 149,617 (18.1% ahead). Even floating-point math, often a strength for high-frequency parts, goes to the EPYC at 256,635 versus 229,685, an 11.7% margin. Data compression scores 1,429,976 for the EPYC versus 1,345,230 for the Threadripper, a 6.3% win. The multi-threaded PassMark score gives the EPYC a 4.3% edge (96,603 vs 92,604).

Cinebench results are consistent but closer. Across R15, R20, and R23, the EPYC 9355P wins both multi-core and single-core variants by exactly 2.1% each time. For example, Cinebench R23 multi-core shows 82,666 for the EPYC versus 80,976 for the Threadripper, while single-core shows 11,670 versus 11,431. The consistency of this 2.1% delta across all six Cinebench tests suggests a steady architectural advantage in that specific rendering workload, regardless of thread count.

The Threadripper PRO 9965WX does secure three wins, and they are significant. The PassMark single-thread test is a clear victory: 4,551 versus 3,747, a 17.7% advantage. This is the largest delta in either direction for a mainstream benchmark, showing the Threadripper's higher boost clock of 5.40 GHz delivers tangible results in latency-sensitive tasks. The extended instructions test goes to the Threadripper by a razor-thin margin: 108,753 versus 107,622, a 1% edge. This near-tie suggests the EPYC's extra cores offset the Threadripper's clock advantage in vectorized code.

Architecture Differences

Both processors are built on TSMC's 4 nm process node and use the Zen 5 architecture, but their physical layouts diverge sharply. The EPYC 9355P, codenamed Turin, packs 66,520 million transistors across 8x 70.6 mm² dies. The Threadripper PRO 9965WX, codenamed Shimada Peak, contains 33,260 million transistors on 4x 70.6 mm² dies. This means the EPYC uses twice as many CCDs, which directly enables its 32 cores and 64 threads versus the Threadripper's 24 cores and 48 threads.

Cache configurations follow the core count. The EPYC provides 80 KB of L1 per core, 1 MB of L2 per core, and a massive 256 MB of shared L3 cache. The Threadripper offers 64 KB of L1 per core, the same 1 MB of L2 per core, but only 128 MB of L3. The EPYC's 256 MB L3 is double the Threadripper's, which likely contributes to its wins in data compression and random string sorting, where large working sets benefit from on-die caching.

Memory subsystems also differ substantially. The EPYC 9355P supports twelve-channel DDR5 memory, yielding a theoretical bandwidth of 576.0 GB/s. The Threadripper PRO 9965WX uses eight-channel DDR5, capping at 409.6 GB/s. That is a 40.6% bandwidth advantage for the EPYC, a critical factor for server workloads that stream large datasets. Both support ECC memory and provide 128 PCIe Gen 5 lanes from the CPU, so expansion capability is identical.

Clock speeds tell the opposite story. The Threadripper has a base clock of 4.20 GHz and a boost of 5.40 GHz, while the EPYC runs at 3.55 GHz base and 4.40 GHz boost. The Threadripper's 1 GHz higher boost clock explains its single-thread dominance. The Threadripper also features an unlocked multiplier, whereas the EPYC is locked, meaning the Threadripper can be overclocked for even higher frequencies. Thermal design power differs as well, with the EPYC at 280 W and the Threadripper at 350 W, reflecting the latter's higher clocks.

Where Each One Wins

The EPYC 9355P is the clear choice for throughput-oriented environments. Its 79.5% physics benchmark lead and 38.8% prime number advantage show it excels in scientific computing, simulation, and any workload that can scale across 32 cores. The 22.4% encryption win positions it for security-heavy server tasks, while the 18% integer math and 18.1% string sorting results make it ideal for database operations and data processing pipelines. The 576.0 GB/s memory bandwidth further cements its role in high-performance computing clusters where memory bandwidth is the bottleneck.

The Threadripper PRO 9965WX wins where latency and single-thread responsiveness matter. Its 17.7% single-thread PassMark lead is the most pronounced difference in the entire dataset. This makes it better suited for workstation applications like CAD, 3D modeling, and software compilation, where individual thread performance directly impacts UI responsiveness and compile times. The 5.40 GHz boost clock is the key differentiator here, allowing it to outpace the EPYC's 4.40 GHz in lightly threaded tasks. The unlockable multiplier offers additional headroom for users willing to push frequencies further.

The extended instructions test, where the Threadripper wins by just 1%, highlights a nuanced middle ground. In AVX-512 or similar vectorized workloads, the Threadripper's higher clocks nearly compensate for the EPYC's extra cores. However, this is the only benchmark where the Threadripper comes close in a multi-threaded scenario, indicating that the EPYC's core advantage is overwhelming in most parallel tasks.

The Verdict

The data points to a clear split: the AMD EPYC 9355P is the superior processor for server deployments and heavily parallel workloads, winning 14 of 17 benchmarks and offering a 40.6% memory bandwidth advantage. Its 32 cores and 256 MB L3 cache deliver decisive wins in physics, encryption, and integer math, making it the logical choice for data centers running virtualized environments, large-scale databases, or scientific simulations. The 2.1% Cinebench lead across all tests suggests consistent rendering performance, even if not transformative.

The AMD Ryzen Threadripper PRO 9965WX is the better pick for professional workstations where single-thread performance is paramount. Its 17.7% single-thread advantage is the largest margin in any test, and the 5.40 GHz boost clock provides a snappy experience for interactive applications. The 24-core configuration still offers substantial multi-threading capability, and the unlocked multiplier allows enthusiasts to extract even more performance. However, the 128 MB L3 and 409.6 GB/s bandwidth are half the EPYC's figures, so workloads that rely on cache capacity or memory throughput will see tangible penalties.

For buyers deciding between these two, the question is whether the workload scales across cores or depends on clock speed. If the answer is the former, the EPYC 9355P's 38.8% prime number lead and 79.5% physics advantage make it non-negotiable. If the answer is the latter, the Threadripper's 17.7% single-thread win is equally compelling. The EPYC's launch MSRP is $2998, while the Threadripper's is $2899.

FAQ

Q: Which processor has more cores?

A: The AMD EPYC 9355P has 32 cores and 64 threads, while the AMD Ryzen Threadripper PRO 9965WX has 24 cores and 48 threads.

Q: What is the biggest single benchmark margin between the two?

A: The largest margin is in the PassMark physics test, where the EPYC 9355P scores 13,515 versus the Threadripper's 7,529, a 79.5% difference.

Q: Does the Threadripper win any multi-threaded benchmarks?

A: Yes, the Threadripper wins the PassMark extended instructions test with a score of 108,753 versus the EPYC's 107,622, a 1% advantage.

Q: How do their memory bandwidths compare?

A: The EPYC 9355P supports twelve-channel DDR5 memory with 576.0 GB/s bandwidth, while the Threadripper PRO 9965WX uses eight-channel DDR5 with 409.6 GB/s.

Q: Are both processors on the same manufacturing process?

A: Yes, both are built by TSMC on a 4 nm process node, but the EPYC uses 8x 70.6 mm² dies with 66,520 million transistors, while the Threadripper uses 4x 70.6 mm² dies with 33,260 million transistors.

Q: Which chip has a higher boost clock?

A: The Threadripper PRO 9965WX boosts to 5.40 GHz, while the EPYC 9355P peaks at 4.40 GHz, explaining the Threadripper's 17.7% single-thread benchmark win.

Specification Differences

| Specification | AMD EPYC 9355P | AMD Ryzen Threadripper PRO 9965WX |

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

| Cores | 32 | 24 |

| Threads | 64 | 48 |

| Base Clock | 3.55 GHz | 4.20 GHz |

| Boost Clock | 4.40 GHz | 5.40 GHz |

| TDP | 280 W | 350 W |

| Socket | AMD Socket SP5 | AMD Socket sTR5 |

| Codename | Turin | Shimada Peak |

| Transistors | 66,520 million | 33,260 million |

| Die Size | 8x 70.6 mm² | 4x 70.6 mm² |

| L1 Cache | 80 KB (per core) | 64 KB (per core) |

| L3 Cache | 256 MB (shared) | 128 MB |

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

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

| Multiplier Unlocked | No | Yes |

| Launch MSRP | $2998 | $2899 |