AMD EPYC 9535 vs AMD Ryzen Threadripper PRO 9985WX Comparison

The AMD EPYC 9535 and AMD Ryzen Threadripper PRO 9985WX are both 64-core, 128-thread Zen 5 processors, but benchmark results place them in distinctly different performance tiers. The data shows a decisive overall victory for the Threadripper PRO part, which wins all 11 head-to-head benchmark comparisons, with the largest margins appearing in multi-threaded and physics workloads.

Head-to-Head Benchmarks

The most striking result in the data is the passmark_physics test, where the AMD Ryzen Threadripper PRO 9985WX scores 13,783 against the EPYC 9535’s 3,834. That represents a 72.2% advantage for the Threadripper part, the largest delta across every recorded benchmark. Physics performance often reflects a combination of high clock speeds and efficient thread scheduling, and the Threadripper’s dominance here is unambiguous.

Beyond physics, the Threadripper PRO 9985WX leads by roughly one-fifth to one-quarter in most PassMark workloads. In passmark_multithread, it scores 150,071 versus 114,528, a 23.7% lead. passmark_random_string_sorting shows a 24.8% gap (329,014 vs. 247,506), and passmark_find_prime_numbers favors the Threadripper by 23.2% (1,138 vs. 874). These are not marginal differences; they indicate a consistent throughput advantage across integer-heavy and memory-access-pattern-sensitive tasks.

The passmark_integer_math test shows a 16.3% lead for the Threadripper PRO (872,710 vs. 730,281), while passmark_floating_point_math shows a 17.4% gap (553,348 vs. 457,047). passmark_extended_instructions follows with a 22% delta (225,340 vs. 175,784). Even in passmark_data_encryption, where the gap is smaller at 17.7% (154,824 vs. 127,372), the Threadripper still holds a clear edge. The passmark_data_compression result (2,912,972 vs. 2,308,822) is 20.7% higher for the Threadripper.

Single-thread performance tells the same story. The passmark_single_thread score for the Threadripper PRO 9985WX is 4,482, which is 17% higher than the EPYC 9535’s 3,720. This is notable because the EPYC is designed for dense server deployments, while the Threadripper targets workstation responsiveness; the data suggests the Threadripper delivers both more aggregate throughput and faster per-core execution.

The avgBenchmarkScore metric, which averages across all recorded benchmarks, further contextualizes these results. The EPYC 9535 averages 379,408, while the Threadripper PRO 9985WX averages 320,749. This seems contradictory given the head-to-head results, but it reflects the fact that the EPYC has a higher percentileVsAllCpus (100th) compared to the Threadripper’s 99th, meaning the EPYC outperforms a slightly larger fraction of all CPUs in the database. However, in direct comparison, the Threadripper wins every single workload.

Where Each One Wins

Based strictly on the benchmark data, the AMD Ryzen Threadripper PRO 9985WX wins every measurable performance category. There is no workload in the head-to-head list where the EPYC 9535 posts a higher score. This includes single-threaded tasks, integer math, floating-point math, encryption, compression, sorting, and physics simulation.

The passmark_physics result deserves special attention because the 72.2% delta is so large. This test typically stresses the memory subsystem and cache hierarchy, and the Threadripper’s 3.20 GHz base clock and 5.40 GHz boost clock likely contribute to its superior showing. The EPYC 9535’s base clock of 2.40 GHz and boost clock of 4.30 GHz are lower, and the benchmark data reflects that gap directly.

For the EPYC 9535, the only meaningful advantage visible in the data is its percentileVsAllCpus of 100, which is one point higher than the Threadripper’s 99. This indicates that the EPYC sits at the very top of the global CPU ranking, while the Threadripper is just below it. Additionally, the EPYC’s avgBenchmarkScore of 379,408 is higher than the Threadripper’s 320,749, but this average includes benchmarks not present in the head-to-head set, so it does not contradict the direct comparison results.

In practical terms, the Threadripper PRO 9985WX is the better choice for any workload represented in the head-to-head list. The EPYC 9535 remains a top-tier server processor, but the data does not show a single scenario where it outperforms the Threadripper.

FAQ

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

A: The AMD Ryzen Threadripper PRO 9985WX scores 4,482 in passmark_single_thread, which is 17% higher than the AMD EPYC 9535’s 3,720.

Q: How large is the performance gap in the physics benchmark?

A: The Threadripper PRO 9985WX scores 13,783 in passmark_physics, while the EPYC 9535 scores 3,834. This represents a 72.2% advantage for the Threadripper, the largest delta in the head-to-head results.

Q: Does the EPYC 9535 win any benchmark in the head-to-head comparison?

A: No. The data shows the Threadripper PRO 9985WX winning all 11 head-to-head benchmarks, with zero wins for the EPYC 9535.

Q: What is the average benchmark score for each processor?

A: The EPYC 9535 has an average benchmark score of 379,408, while the Threadripper PRO 9985WX has an average of 320,749. Note that this average includes benchmarks outside the head-to-head set.

Q: How do the two processors compare in data compression?

A: The Threadripper PRO 9985WX scores 2,912,972 in passmark_data_compression, which is 20.7% higher than the EPYC 9535’s 2,308,822.

Q: What is the percentile ranking for each processor?

A: The EPYC 9535 ranks in the 100th percentile versus all CPUs, while the Threadripper PRO 9985WX ranks in the 99th percentile.

Specification Differences

The two processors differ in several core specifications, despite sharing the same core and thread counts. The EPYC 9535 has a base clock of 2.40 GHz and a boost clock of 4.30 GHz, while the Threadripper PRO 9985WX runs at a base clock of 3.20 GHz and a boost clock of 5.40 GHz. The Threadripper also has a higher thermal design power (TDP) at 350 compared to the EPYC’s 300.

The memory bus differs significantly: the EPYC 9535 uses a twelve-channel memory bus with a memory bandwidth of 576.0 GB/s, while the Threadripper PRO 9985WX uses an eight-channel bus with a memory bandwidth of 409.6 GB/s. Both support DDR5 and ECC memory, but the EPYC’s wider memory interface gives it a theoretical bandwidth advantage that does not translate into benchmark wins in the data.

The sockets are different: the EPYC 9535 uses AMD Socket SP5, while the Threadripper PRO 9985WX uses AMD Socket sTR5. The multiplier is locked on the EPYC 9535, whereas the Threadripper PRO 9985WX has an unlocked multiplier. The release dates also differ, with the EPYC 9535 launching on 2024-10-09 and the Threadripper PRO 9985WX launching on 2025-07-22.

The launch MSRP for the EPYC 9535 is $8992, while the Threadripper PRO 9985WX has a launch MSRP of $7999.

Architecture Differences

Both processors are built on the Zen 5 architecture and manufactured on a 4 nm process at TSMC, with identical transistor counts of 66,520 million and a die size of 8x 70.6 mm². The codenames differ: the EPYC 9535 is codenamed "Turin" and belongs to the EPYC 9005 series, while the Threadripper PRO 9985WX is codenamed "Shimada Peak" and belongs to the 9000 series.

The cache hierarchy shows a notable difference in L1 cache. The EPYC 9535 has 80 KB of L1 cache per core, while the Threadripper PRO 9985WX has 64 KB per core. Both share 1 MB of L2 cache per core and 256 MB of L3 cache, with the Threadripper listing its L3 as 256 MB and the EPYC listing 256 MB (shared). No 3D V-Cache is present on either processor.

Both processors support PCIe Gen 5 with 128 lanes (CPU only) and have no integrated graphics. The EPYC 9535 is a server/workstation part, and the Threadripper PRO 9985WX is also classified as server/workstation. Both are currently in active production.

The Verdict

The benchmark data is unequivocal: the AMD Ryzen Threadripper PRO 9985WX outperforms the AMD EPYC 9535 in every head-to-head workload recorded. With a 72.2% lead in physics, a 23.7% lead in multithread performance, and a 17% lead in single-thread performance, the Threadripper is the stronger processor for any task represented in this data set.

The EPYC 9535’s advantages are structural rather than performance-based. It has a higher percentile ranking (100th vs. 99th), a wider twelve-channel memory bus with higher theoretical bandwidth (576.0 GB/s vs. 409.6 GB/s), and a lower TDP (300 vs. 350). These features make it suitable for dense server deployments where memory bandwidth per socket and power efficiency are priorities, even if the benchmark scores do not reflect a performance win.

For users who prioritize raw compute performance, the AMD Ryzen Threadripper PRO 9985WX is the clear choice. Its higher base and boost clocks, combined with its benchmark dominance, make it the better option for workstation workloads. The EPYC 9535 remains a top-tier server processor, but the data shows that the Threadripper PRO 9985WX delivers superior results in direct comparison.