AMD EPYC 9175F vs AMD Ryzen Threadripper PRO 5965WX Comparison

The AMD Ryzen Threadripper PRO 5965WX and the AMD EPYC 9175F represent two distinct philosophies for high-core-count computing. The data shows a clear split: the Threadripper PRO wins the majority of the head-to-head comparisons, taking 11 of 17 benchmarks, while the EPYC counters with decisive victories in specific, targeted workloads. Benchmark results indicate that the Threadripper PRO’s advantage is broad but often narrow, while the EPYC 9175F’s wins are substantial, particularly in single-threaded and physics-based tests.

Head-to-Head Benchmarks

The most striking pattern in the head-to-head data is the sheer consistency of the Threadripper PRO 5965WX across all Cinebench iterations. In Cinebench R15, R20, and R23, for both single-core and multi-core tests, the Threadripper PRO wins every round with a deltaPct of 0.9% (and 0.8% in R23 single-core). The scores are remarkably close: in Cinebench R23 multi-core, the Threadripper PRO scores 56400 against the EPYC’s 55923, a difference of just 477 points. This suggests that in pure CPU rendering throughput, the two processors are essentially neck-and-neck, with the Threadripper PRO holding a marginal, yet consistent, edge.

The Threadripper PRO’s wins are not limited to rendering. Its largest victories come in data-centric tests. In Passmark data encryption, it posts a score of 63113, which is a massive 49.2% ahead of the EPYC 9175F’s 42297. Similarly, in Passmark integer math, the Threadripper PRO scores 281247, a 28% advantage over the EPYC’s 219800. Other notable wins for the Threadripper PRO include Passmark data compression (16.5% ahead) and Passmark floating point math (8.1% ahead). These are not marginal differences; they indicate a fundamental advantage in workloads that rely on massive parallel throughput across many cores.

The EPYC 9175F, however, does not go down quietly. Its most dramatic victory is in Passmark physics, where it scores 9984 against the Threadripper PRO’s 4251, a staggering 57.4% difference. This is the single largest delta in the entire comparison. The EPYC also wins Passmark single-thread by a significant margin, scoring 4256 versus 3337, a 21.6% advantage. In Passmark find prime numbers, the EPYC scores 741 versus 523, a 29.4% win. Its other wins include Passmark extended instructions (4.6% ahead) and Passmark multithread, where it scores 67634 versus 66353, a 1.9% edge. While the EPYC wins fewer benchmarks, the magnitude of its victories, especially in physics and single-thread performance, is undeniable.

Where Each One Wins

Looking at the distribution of wins, the Threadripper PRO 5965WX is the clear choice for workloads that scale with core count and sustained multi-threaded execution. Its wins in Cinebench multi-core tests, data compression, encryption, integer math, and floating-point math point toward a processor built for heavy, parallel computation. The 49.2% lead in data encryption and the 28% lead in integer math are particularly telling; these are operations that benefit directly from having 24 cores and 48 threads available to chew through large datasets. The Threadripper PRO’s 11 wins out of 17 benchmarks demonstrate that it is the more well-rounded performer for general workstation tasks.

The EPYC 9175F, conversely, wins where clock speed and per-core efficiency matter more than raw core count. Its 57.4% victory in Passmark physics is a clear signal for simulation and physics-based workloads that are often latency-sensitive and less parallelizable. The 21.6% win in single-thread performance further reinforces this, suggesting that lightly-threaded applications or those with a single dominant thread will see a significant boost on the EPYC. The win in find prime numbers (29.4% ahead) also aligns with workloads that are heavily dependent on branch prediction and integer throughput per core rather than aggregate core count. The EPYC’s win in Passmark multithread, albeit narrow at 1.9%, indicates that even in some multi-threaded scenarios, its architecture can hold its own.

Architecture Differences

The fundamental architectural divide between these two CPUs is the difference between Zen 3 and Zen 5. The Threadripper PRO 5965WX is built on the Zen 3 architecture (codename Chagall PRO) using a 7 nm process from TSMC, featuring a transistor count of 16,600 million spread across a die size of 4x 81 mm². In contrast, the EPYC 9175F is a Zen 5 part (codename Turin) manufactured on a 4 nm process, also by TSMC, with a vastly higher transistor count of 133,040 million distributed across 16x 70.6 mm² chiplets. This process shrink and architectural generation leap are the root causes of the EPYC’s higher clock speeds and per-core performance.

Cache configurations also differ significantly. The Threadripper PRO offers 64 KB of L1 cache and 512 KB of L2 cache per core, alongside a 128 MB L3 cache. The EPYC 9175F, on the other hand, provides 80 KB of L1 and 1 MB of L2 per core, with a much larger 512 MB of shared L3 cache. The larger L2 cache per core on the EPYC likely contributes to its superior single-thread performance, while the massive 512 MB L3 cache is a boon for large working sets that can be cached entirely on-chip. The Threadripper PRO’s cache hierarchy, while substantial, is clearly a generation behind.

Memory and I/O architecture also show a generational leap. The Threadripper PRO supports DDR4 memory over an eight-channel bus, providing 204.8 GB/s of bandwidth. The EPYC 9175F moves to DDR5 over a twelve-channel bus, nearly tripling memory bandwidth to 576.0 GB/s. Both support ECC memory, but the EPYC’s platform is designed for higher bandwidth server workloads. PCIe connectivity also advances from Gen 4 on the Threadripper PRO to Gen 5 on the EPYC, though both offer 128 lanes (CPU only).

Specification Differences

The specification sheets reveal several key divergences beyond the architectural core. The most obvious is core count: the Threadripper PRO 5965WX has 24 cores and 48 threads, while the EPYC 9175F has 16 cores and 32 threads. This 8-core difference explains why the Threadripper PRO often wins multi-threaded benchmarks, despite the EPYC’s higher clock speeds. The EPYC’s base clock is 4.20 GHz and boost clock is 5.00 GHz, compared to the Threadripper PRO’s 3.80 GHz base and 4.50 GHz boost. This 500 MHz advantage in boost clock is directly reflected in the EPYC’s single-thread performance wins.

Thermal design power (TDP) differs as well, with the Threadripper PRO rated at 280 W and the EPYC 9175F at 320 W. The EPYC’s higher TDP reflects its higher clock speeds and more advanced memory controller. The socket and platform are also entirely different: the Threadripper PRO uses AMD Socket WRX8, while the EPYC 9175F uses AMD Socket SP5. This means they are not interchangeable in any system. The release dates are also notable, with the Threadripper PRO launching on 2022-03-07 and the EPYC 9175F on 2024-10-09, reflecting their different generations. The Threadripper PRO has a launch MSRP of $2399, while the EPYC 9175F has a launch MSRP of $4256. Neither processor has an unlocked multiplier, and the EPYC lists integrated graphics as "N/A" while the Threadripper PRO has no integrated graphics listed.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen Threadripper PRO 5965WX has 24 cores and 48 threads, while the AMD EPYC 9175F has 16 cores and 32 threads.

Q: Why does the EPYC 9175F have a much higher Passmark physics score?

A: The EPYC 9175F scores 9984 in Passmark physics, which is 57.4% higher than the Threadripper PRO’s 4251. This is the largest single benchmark delta in the comparison, indicating the EPYC’s architecture is significantly better suited for physics simulation workloads.

Q: Is the Threadripper PRO 5965WX always faster than the EPYC 9175F in multi-core tests?

A: No. While the Threadripper PRO wins all Cinebench multi-core tests by 0.9%, the EPYC 9175F wins the Passmark multithread test, scoring 67634 versus 66353, a 1.9% advantage.

Q: What is the difference in memory bandwidth between the two CPUs?

A: The Threadripper PRO 5965WX supports DDR4 over an eight-channel bus with 204.8 GB/s bandwidth. The EPYC 9175F supports DDR5 over a twelve-channel bus with 576.0 GB/s bandwidth.

Q: Which CPU has a larger L3 cache?

A: The AMD EPYC 9175F has a 512 MB shared L3 cache, which is four times larger than the 128 MB L3 cache on the AMD Ryzen Threadripper PRO 5965WX.

Q: How many benchmark wins does each CPU have in this comparison?

A: The AMD Ryzen Threadripper PRO 5965WX wins 11 of the 17 head-to-head benchmarks, while the AMD EPYC 9175F wins 6.

The Verdict

The data presents a straightforward choice based on workload profile. For users who prioritize raw multi-threaded throughput across a wide variety of tasks, the AMD Ryzen Threadripper PRO 5965WX is the superior option. It wins 11 of 17 benchmarks, including all Cinebench tests, and shows decisive advantages in data encryption (49.2% ahead), integer math (28% ahead), and data compression (16.5% ahead). With 24 cores and 48 threads, it is the better engine for rendering, code compilation, and heavy data processing that can utilize all of its threads effectively. Its average benchmark score of 98504, compared to the EPYC’s 95615, reflects this broader strength.

The AMD EPYC 9175F, however, is the pick for workloads that are sensitive to single-thread performance and clock speed. Its 57.4% lead in Passmark physics and 21.6% lead in single-thread performance are too significant to ignore for specific applications. The 5.00 GHz boost clock and newer Zen 5 architecture deliver a level of per-core performance that the Zen 3-based Threadripper PRO cannot match. For simulation, certain scientific computing tasks, and lightly-threaded applications, the EPYC 9175F is clearly the better choice, despite having 8 fewer cores. The EPYC also offers substantially more memory bandwidth (576.0 GB/s vs 204.8 GB/s) and a larger L3 cache (512 MB vs 128 MB), which could be decisive for data-heavy server workloads. In short, the Threadripper PRO is the multi-threaded generalist, while the EPYC 9175F is the high-frequency specialist.