AMD EPYC 9355P vs AMD Ryzen 9 PRO 9965 Comparison

Head-to-Head Benchmarks

The benchmark data presents a striking asymmetry: the AMD EPYC 9355P wins nine of the eleven head-to-head tests, while the AMD Ryzen 9 PRO 9965 takes only two. The margins are not subtle. In PassMark's Physics test, the EPYC 9355P scores 13,515 against the Ryzen 9 PRO 9965's 3,256, a 315.1% advantage that dwarfs every other gap in the comparison. That single result suggests the EPYC's 32 cores and 64 threads are being leveraged in a way the 16-core, 32-thread Ryzen simply cannot match.

The pattern repeats across compute-heavy workloads. In PassMark's Find Prime Numbers test, the EPYC 9355P scores 1,044 versus 364 for the Ryzen 9 PRO 9965, translating to a 186.8% lead. Integer math shows a 69.4% gap (412,067 versus 243,280), while floating-point math lands at 59.7% (256,635 versus 160,746). Data encryption tells a similar story: 80,961 for the EPYC against 44,920 for the Ryzen, a 80.2% difference. Even random string sorting, which often rewards memory bandwidth, goes decisively to the EPYC 9355P at 176,697 versus 94,915, a 86.2% edge.

The EPYC 9355P also dominates in data compression (1,429,976 versus 908,293, up 57.4%), extended instructions (107,622 versus 71,210, up 51.1%), and the overall multithread score (96,603 versus 66,655, up 44.9%). These are not marginal wins; they are consistent, often doubling or nearly doubling the Ryzen's output. The data points to a processor engineered for throughput, where every additional core and memory channel compounds into large performance deltas.

The Ryzen 9 PRO 9965's two wins come in single-thread tests. In PassMark's Single Thread test, it scores 4,682 against the EPYC's 3,747, a 20% advantage. The identical result appears in the Singlethread test, confirming the pattern. This is a meaningful reversal: when a workload scales poorly across cores, the Ryzen's higher base clock (4.30 GHz versus 3.55 GHz) and boost clock (5.50 GHz versus 4.40 GHz) give it a clear edge. The Ryzen is not merely slower in single-threaded tasks; it is substantially faster, and that could matter for lightly threaded applications.

Looking at the broader benchmark context, the EPYC 9355P's average benchmark score of 160,358 places it in the 98th percentile of all CPUs. Its nearest rivals include the AMD EPYC 9375F (162,497, a -1.3% delta) and the AMD EPYC 7663 (161,973, -1% delta), meaning it sits just below those parts in aggregate performance. The Ryzen 9 PRO 9965, also in the 98th percentile, averages 145,728, with its closest rival being the AMD Ryzen Threadripper PRO 9965WX at 147,009 (-0.9% delta). The EPYC's average score is roughly 10% higher than the Ryzen's, which aligns with the head-to-head results where the EPYC wins the majority of tests.

The Verdict

The data draws a clear line: the AMD EPYC 9355P is the choice for workloads that scale across cores and memory bandwidth. Its wins in multithread, integer math, floating-point math, and data encryption show it excels in server and workstation scenarios where parallel processing is the norm. The Ryzen 9 PRO 9965, despite its lower core count, is the pick for single-threaded performance, where its 20% lead in PassMark Single Thread could translate into faster response times for lightly threaded applications or legacy software.

For a database server, virtualization host, or rendering farm, the EPYC 9355P's 44.9% multithread advantage and 186.8% lead in prime number calculations suggest it will finish batch jobs significantly faster. Its 256 MB of shared L3 cache, versus the Ryzen's 64 MB, also supports the data compression and encryption wins, which often rely on keeping working sets in cache. The Ryzen 9 PRO 9965, with its dual-channel memory bus and 89.6 GB/s bandwidth, is not in the same league for memory-intensive parallel work.

The Ryzen's path to victory is narrow but real. In single-threaded tests, it wins decisively, and its higher clock speeds could make it a better fit for interactive workloads or applications with strict per-core performance requirements. The data shows it is also competitive in the 98th percentile, so it is not a weak part; it is simply aimed at a different use case.

The decision hinges on workload shape. If the question is "which CPU finishes a heavily parallel job first," the answer is the EPYC 9355P without hesitation. If the question is "which CPU handles a single-threaded bottleneck better," the Ryzen 9 PRO 9965 takes it. Neither part is a general-purpose winner; they are specialized tools with complementary strengths.

FAQ

Q: Which CPU has a higher single-thread benchmark score?

A: The AMD Ryzen 9 PRO 9965 scores 4,682 in the PassMark Single Thread test, compared to 3,747 for the AMD EPYC 9355P, giving the Ryzen a 20% advantage.

Q: How much faster is the EPYC 9355P in multithread performance?

A: In the PassMark Multithread test, the EPYC 9355P scores 96,603 versus 66,655 for the Ryzen 9 PRO 9965, a 44.9% lead.

Q: What is the largest performance gap between the two processors?

A: The largest gap is in the PassMark Physics test, where the EPYC 9355P scores 13,515 against 3,256 for the Ryzen 9 PRO 9965, a 315.1% difference.

Q: Do both CPUs support ECC memory?

A: Yes, both the AMD EPYC 9355P and the AMD Ryzen 9 PRO 9965 list ECC memory support as true.

Q: Which processor has more PCIe lanes?

A: The AMD EPYC 9355P supports Gen 5 with 128 lanes (CPU only), while the AMD Ryzen 9 PRO 9965 supports Gen 5 with 24 lanes (CPU only).

Q: Are both processors in the same performance percentile?

A: Yes, both the EPYC 9355P and the Ryzen 9 PRO 9965 are in the 98th percentile of all CPUs, though their average benchmark scores differ (160,358 versus 145,728).

Specification Differences

The two processors diverge sharply on core and thread counts. The AMD EPYC 9355P has 32 cores and 64 threads, while the AMD Ryzen 9 PRO 9965 has 16 cores and 32 threads. Base clocks differ as well: the EPYC runs at 3.55 GHz, while the Ryzen runs at 4.30 GHz. Boost clocks follow the same pattern, with the EPYC at 4.40 GHz and the Ryzen at 5.50 GHz. Thermal design power is another differentiator: the EPYC is rated at 280 W, while the Ryzen is rated at 170 W.

Memory architecture sets them apart even further. The EPYC 9355P uses a twelve-channel memory bus with 576.0 GB/s of bandwidth, while the Ryzen 9 PRO 9965 uses a dual-channel bus with 89.6 GB/s. L3 cache also differs: the EPYC has 256 MB shared, versus 64 MB for the Ryzen. The sockets are incompatible: the EPYC uses AMD Socket SP5, while the Ryzen uses AMD Socket AM5. The EPYC has no integrated graphics, while the Ryzen includes Radeon Graphics. Transistor counts also differ significantly, with the EPYC at 66,520 million and the Ryzen at 16,630 million.

Architecture Differences

Both processors are built on TSMC's 4 nm process node and use the Zen 5 architecture, but they belong to different product families. The EPYC 9355P is part of the EPYC 9005 series with the codename Turin, while the Ryzen 9 PRO 9965 is in the 9000 series with the codename Granite Ridge. The EPYC uses a chiplet design with 8x 70.6 mm² dies, while the Ryzen uses 2x 70.6 mm² dies. Both have 80 KB of L1 cache per core and 1 MB of L2 cache per core, but the L3 cache differs as noted, with the EPYC offering four times the shared L3 capacity.

The EPYC 9355P's market segment is Server/Workstation, and it has a launch MSRP of $2998. The Ryzen 9 PRO 9965 is also listed as Server/Workstation, but it has no launch MSRP in the data. The EPYC's part number is 100-000001521, while the Ryzen's is 100-000002001. Both are active products, and neither has an unlocked multiplier. The EPYC was released on 2024-10-09, while the Ryzen's release date is listed as 2026-06-29. The EPYC's memory support is DDR5 with twelve channels, while the Ryzen also supports DDR5 but only two channels. Both support ECC memory, which is notable for reliability-focused workloads.

Where Each One Wins

The AMD EPYC 9355P is the clear winner in every parallel workload the data covers. It wins data compression by 57.4%, data encryption by 80.2%, extended instructions by 51.1%, prime number finding by 186.8%, floating-point math by 59.7%, integer math by 69.4%, multithread by 44.9%, physics by 315.1%, and random string sorting by 86.2%. If the task involves processing large datasets, running many virtual machines, or compiling code with multiple threads, the EPYC's 32 cores and twelve-channel memory bandwidth make it the obvious selection. Its 576.0 GB/s memory bandwidth is more than six times the Ryzen's, which likely explains the compression and sorting wins.

The AMD Ryzen 9 PRO 9965 wins only in single-threaded tests, taking PassMark Single Thread and Singlethread by 20% each. Its higher base and boost clocks are the likely explanation, as the data shows a 4.30 GHz base and 5.50 GHz boost versus the EPYC's 3.55 GHz and 4.40 GHz. This makes the Ryzen a better fit for applications that are latency-sensitive or have poor multi-core scaling, such as older database engines, some scientific software, or interactive workloads where a single thread is the bottleneck. Its 170 W TDP also suggests it is a more power-efficient choice for lightly threaded tasks, though the data does not include power consumption figures.

The use-case split is not subtle. The EPYC 9355P is for throughput-centric environments where the goal is maximizing work per unit of time across many cores. The Ryzen 9 PRO 9965 is for scenarios where raw single-core speed matters more than core count, and where the dual-channel memory bus is sufficient. The benchmark results confirm that neither processor is a compromise; each is strongly optimized for its intended role, and the data shows exactly where the boundaries lie.