AMD EPYC 9335 vs AMD Ryzen Threadripper PRO 9975WX Comparison

Both the AMD EPYC 9335 and the AMD Ryzen Threadripper PRO 9975WX are 32-core Zen 5 processors built on the same 4nm TSMC process, but benchmark results show they are not equals. Across the eleven head-to-head PassMark tests, the Threadripper PRO 9975WX wins decisively, taking all 11 contests. The data indicates that the EPYC 9335, while a strong server part, is consistently outpaced by the Threadripper part in every measurable workload, with the most significant gaps appearing in single-threaded and physics-based tasks.

Head-to-Head Benchmarks

The benchmark data paints a clear picture of dominance for the AMD Ryzen Threadripper PRO 9975WX. In the passmark_single_thread test, the Threadripper scores 4408 against the EPYC's 2732, a delta of -38% in favor of the Threadripper. This indicates a substantial advantage in applications that rely on per-core speed, likely stemming from its higher boost clock. The same pattern holds in passmark_physics, where the Threadripper's score of 7288 dwarfs the EPYC's 1905, representing a massive -73.9% delta. This suggests the Threadripper is far better suited for simulation and physics-based workloads that are sensitive to raw core performance.

The Threadripper's lead persists even in heavily multithreaded tests. In passmark_multithread, it scores 104902 versus the EPYC's 65811, a -37.3% delta. This is a significant margin, meaning the Threadripper delivers more than a third more throughput in parallel workloads. Similarly, in passmark_integer_math, the Threadripper scores 461724 compared to the EPYC's 346291, a -25% delta, and in passmark_floating_point_math, it scores 304833 versus 228123, a -25.2% delta. These results show that despite having the same core and thread count, the Threadripper's architecture allows it to execute more instructions per cycle.

The data compression and encryption tests reinforce this trend. In passmark_data_compression, the Threadripper scores 1644573 against the EPYC's 1203096, a -26.8% delta. In passmark_data_encryption, it scores 85035 versus 63159, a -25.7% delta. Even in passmark_extended_instructions, which often highlights architectural efficiency, the Threadripper's 137733 score beats the EPYC's 105706 by a -23.3% delta. The only test where the gap is relatively smaller is passmark_find_prime_numbers, where the Threadripper scores 620 against the EPYC's 340, still a substantial -45.2% delta. Across the board, the Threadripper PRO 9975WX is the faster processor.

Where Each One Wins

Based on the benchmark data, the AMD Ryzen Threadripper PRO 9975WX is the clear winner in every single category tested. There are no benchmark wins for the EPYC 9335; the winsA count is 0, while the winsB count is 11. This means that for any workload represented by these PassMark tests, the Threadripper part offers superior performance. The most pronounced advantage for the Threadripper is in physics simulations, where its score is nearly four times higher, indicating a massive lead in tasks that require complex floating-point calculations and single-core speed.

The EPYC 9335's role is not defined by winning benchmark tests but by its positioning within the server market. Its nearest rival, the Intel Xeon 6741P, posts an average score of 194901, which is only -0.3% different from the EPYC's 194228 average. This places the EPYC in a highly competitive field for server workloads. The Threadripper, in contrast, has a lower average benchmark score of 182700, but its architecture is clearly tuned for maximum performance per thread, as evidenced by its benchmark victories. Therefore, the data suggests the Threadripper is the choice for raw performance in a workstation, while the EPYC is a more balanced server part that trades top-end performance for a broader competitive standing.

Architecture Differences

Both processors are built on the Zen 5 architecture and use the same 4nm process node from TSMC, with a transistor count of 33,260 million and a die size of 4x 70.6 mm². However, their codenames differ: the EPYC 9335 is from the "Turin" generation, while the Threadripper PRO 9975WX is from "Shimada Peak". A key architectural difference lies in the cache hierarchy. The EPYC 9335 features 80 KB of L1 cache per core, while the Threadripper PRO 9975WX has a smaller 64 KB of L1 cache per core. Both have 1 MB of L2 cache per core and 128 MB of shared L3 cache.

The memory interface is another significant architectural divergence. The EPYC 9335 supports a twelve-channel memory bus, delivering a memory bandwidth of 576.0 GB/s. In contrast, the Threadripper PRO 9975WX uses an eight-channel memory bus with a bandwidth of 409.6 GB/s. This indicates that the EPYC is designed for memory-intensive server workloads that benefit from higher aggregate bandwidth, while the Threadripper's lower bandwidth is offset by its higher clock speeds and superior per-core performance. Both support DDR5 memory and ECC memory, and both lack integrated graphics.

Specification Differences

The specification sheets reveal distinct target markets for these two processors. The most obvious difference is in clock speeds: the EPYC 9335 has a base clock of 3.00 GHz and a boost clock of 4.40 GHz, while the Threadripper PRO 9975WX runs at a base of 4.00 GHz and boosts to 5.40 GHz. This 1.00 GHz advantage in boost clock is the primary driver of the Threadripper's benchmark dominance. The thermal design power (TDP) also reflects this, with the EPYC rated at 210 W and the Threadripper at 350 W. The EPYC uses AMD Socket SP5, while the Threadripper uses AMD Socket sTR5, making them incompatible with each other's motherboards.

Other key differences include the multiplier unlock status; the Threadripper PRO 9975WX has an unlocked multiplier, allowing for overclocking, while the EPYC 9335 does not. The release dates also differ, with the EPYC launching on 2024-10-09 and the Threadripper on 2025-07-22. The launch MSRP is $3178 for the EPYC and $4099 for the Threadripper. The part numbers are also different: 100-000001149 for the EPYC and 100-000000723 for the Threadripper. While both have 128 PCIe Gen 5 lanes, the EPYC's twelve-channel memory bus versus the Threadripper's eight-channel bus is a major specification split.

FAQ

Q: Which processor has a higher single-thread performance?

A: The AMD Ryzen Threadripper PRO 9975WX is significantly faster in single-threaded tests, scoring 4408 in the passmark_single_thread test compared to the AMD EPYC 9335's 2732, a -38% delta.

Q: How do the two compare in multithreaded workloads?

A: The Threadripper PRO 9975WX also leads in multithreaded tests, scoring 104902 in passmark_multithread against the EPYC 9335's 65811, a -37.3% delta, despite both having 32 cores and 64 threads.

Q: What is the difference in memory bandwidth?

A: The AMD EPYC 9335 has a memory bandwidth of 576.0 GB/s via its twelve-channel memory bus, whereas the AMD Ryzen Threadripper PRO 9975WX has a bandwidth of 409.6 GB/s over an eight-channel bus.

Q: Are both processors based on the same architecture?

A: Yes, both are built on the Zen 5 architecture using a 4nm process, but the EPYC 9335 has the codename "Turin" while the Threadripper PRO 9975WX is codenamed "Shimada Peak".

Q: Which processor has a higher boost clock speed?

A: The AMD Ryzen Threadripper PRO 9975WX has a higher boost clock of 5.40 GHz, while the AMD EPYC 9335 has a boost clock of 4.40 GHz.

Q: What are the TDP ratings for each processor?

A: The AMD EPYC 9335 has a TDP of 210 W, while the AMD Ryzen Threadripper PRO 9975WX has a TDP of 350 W.

The Verdict

The data is unambiguous: the AMD Ryzen Threadripper PRO 9975WX is the superior processor for performance-focused users. It wins all 11 head-to-head benchmark tests, with particularly large margins in physics (-73.9%) and single-threaded (-38%) workloads. Its higher base and boost clocks, 4.00 GHz and 5.40 GHz respectively, compared to the EPYC's 3.00 GHz and 4.40 GHz, give it a clear edge in almost any application. The Threadripper's unlocked multiplier also offers additional flexibility for users willing to push beyond stock settings, though its 350 W TDP is substantially higher than the EPYC's 210 W.

The AMD EPYC 9335, however, is not without its merits. It is designed for server environments where memory bandwidth is critical, offering a twelve-channel memory bus with 576.0 GB/s throughput compared to the Threadripper's eight-channel, 409.6 GB/s setup. Its average benchmark score of 194228 is also higher than the Threadripper's 182700, and it sits competitively against its nearest rivals like the Intel Xeon 6741P, which is only -0.3% different in average score. This makes the EPYC a strong choice for server racks where balanced performance and high memory bandwidth are prioritized over raw per-core speed.

For a workstation user, the choice is clear: the Threadripper PRO 9975WX offers superior speed in every measured category. For a server administrator needing high memory bandwidth and a more competitive average score, the EPYC 9335 is the appropriate pick. The data supports the Threadripper as the performance champion, while the EPYC serves as a specialized server part with a different set of strengths. There is no benchmark-based argument for choosing the EPYC over the Threadripper for raw compute performance.