AMD EPYC 9355P vs Intel Xeon 6732P Comparison

# AMD EPYC 9355P vs Intel Xeon 6732P

The AMD EPYC 9355P is the dominant performer in this head-to-head, winning 16 of 17 benchmark comparisons against the Intel Xeon 6732P. The recorded data shows a consistent and substantial advantage for AMD across nearly every workload category, with the Intel part claiming a single narrow victory in floating-point math. Both processors sit at the 98th percentile among all CPUs in the database, but their average benchmark scores tell a different story: the EPYC 9355P averages 160,358 points while the Xeon 6732P averages 143,444 points. That is an 11.8% gap in overall average score, placing the AMD part alongside rivals like the AMD EPYC 7663 (161,973, 1% higher) and the AMD EPYC 9375F (162,497, 1.3% higher), while the Intel Xeon 6732P aligns more closely with the AMD Ryzen 9 PRO 9965X3D (143,735, 0.2% lower) and the Intel Xeon w9-3575X (144,323, 0.6% higher).

Where Each One Wins

The AMD EPYC 9355P wins in every meaningful performance category except one. Its largest advantages come in integer-heavy and physics-based workloads. The PassMark physics test shows a 66.7% lead, and the find prime numbers test shows a 66.2% lead. These are substantial margins that indicate the Zen 5 architecture handles computational intensity far better than the Granite Rapids design in these specific tasks.

The EPYC 9355P also dominates in single-threaded performance, with a 49.5% lead in the PassMark single-thread test (3747 vs 2506). This advantage carries over to Cinebench single-core results, where the AMD part leads by 29.9% across R15, R20, and R23 versions. For workloads that depend on per-core speed, such as database queries or legacy single-threaded applications, the EPYC 9355P is clearly the stronger choice.

The Intel Xeon 6732P wins exactly one benchmark: PassMark floating-point math, where it scores 261,703 versus 256,635, a 1.9% margin. This suggests the Intel architecture has a slight edge in pure floating-point throughput, but the margin is small and does not compensate for the broad deficits elsewhere.

In multithreaded workloads, the EPYC 9355P leads by 29.1% in PassMark multithread (96,603 vs 74,849) and by 29.9% in Cinebench R23 multicore (82,666 vs 63,621). Data compression favors AMD by 6.8%, data encryption by 26.8%, and random string sorting by 32.4%. The only benchmark where the two are nearly tied is extended instructions, with AMD ahead by just 0.9% (107,622 vs 106,697).

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD EPYC 9355P is built on the Zen 5 architecture with the codename Turin, part of the EPYC 9005 series. It uses a 4 nm process node manufactured by TSMC and packs 66,520 million transistors across eight chiplets, each 70.6 mm² in size. The Intel Xeon 6732P uses the Granite Rapids architecture, part of the Xeon 6 generation, on a 5 nm process node from Intel's own foundry.

Cache configurations differ significantly. The EPYC 9355P has 80 KB of L1 cache per core and 1 MB of L2 cache per core, but its L3 cache is a massive 256 MB shared pool. The Xeon 6732P has larger per-core caches, with 112 KB of L1 and 2 MB of L2 per core, but its shared L3 cache is only 144 MB. This means the AMD part offers 112 MB more shared L3 cache, which can be beneficial for workloads with large working sets that fit in cache.

Memory architecture also diverges. The EPYC 9355P supports twelve-channel DDR5 memory with a theoretical bandwidth of 576.0 GB/s. The Xeon 6732P supports eight-channel DDR5 with 409.6 GB/s of bandwidth. That is a 40.6% bandwidth advantage for AMD, which directly impacts memory-intensive workloads like data compression and random string sorting.

PCIe connectivity differs as well. The EPYC 9355P provides 128 Gen 5 lanes from the CPU, while the Xeon 6732P provides 136 Gen 5 lanes. The Intel part has eight additional lanes, which may matter for systems with many NVMe drives or GPUs, though the practical impact depends on the specific server configuration.

Both processors use the DDR5 memory standard and support ECC memory. Neither includes integrated graphics, and both are locked multipliers aimed at the server and workstation market. The EPYC 9355P uses AMD Socket SP5, while the Xeon 6732P uses Intel Socket 4710.

Head-to-Head Benchmarks

The Cinebench results are remarkably consistent: the EPYC 9355P leads by exactly 29.9% in all six Cinebench tests. In R15 multicore, AMD scores 8,332 versus Intel's 6,412. In R15 single-core, the scores are 1,176 versus 905. The R20 multicore test shows 34,719 against 26,720, and R20 single-core shows 4,901 against 3,772. R23 multicore delivers 82,666 versus 63,621, and R23 single-core delivers 11,670 versus 8,981. This uniform 29.9% margin across all Cinebench versions indicates a consistent architectural advantage in both single-threaded and multi-threaded rendering workloads.

PassMark results show a wider range of outcomes. The largest single advantage is in the physics test, where AMD scores 13,515 versus Intel's 8,109, a 66.7% lead. The find prime numbers test shows a similar 66.2% margin (1,044 vs 628). These two tests highlight the AMD part's strength in computationally intensive, non-floating-point workloads.

Integer math favors AMD by 23.2% (412,067 vs 334,340), and data encryption favors AMD by 26.8% (80,961 vs 63,848). Random string sorting shows a 32.4% advantage (176,697 vs 133,467). Multithread performance is 29.1% higher (96,603 vs 74,849), and single-thread performance is 49.5% higher (3,747 vs 2,506).

Data compression is closer, with AMD ahead by 6.8% (1,429,976 vs 1,339,480). Extended instructions are nearly identical, with AMD leading by just 0.9% (107,622 vs 106,697). The only Intel win is floating-point math, where the Xeon 6732P leads by 1.9% (261,703 vs 256,635).

Specification Differences

The core and thread counts are identical: both processors have 32 cores and 64 threads. The differences begin with clock speeds. The Intel Xeon 6732P has a higher base clock of 3.80 GHz versus 3.55 GHz for the EPYC 9355P. However, the AMD part has a higher boost clock of 4.40 GHz versus 4.10 GHz for Intel. This combination explains the single-threaded performance gap: AMD's higher boost clock, combined with the Zen 5 architecture, delivers a 49.5% lead in PassMark single-thread testing.

Thermal design power differs substantially. The EPYC 9355P has a TDP of 280 watts, while the Xeon 6732P has a TDP of 350 watts. The Intel part draws 25% more power at the TDP level, which has implications for cooling and power delivery in dense server deployments.

L3 cache is a major differentiator: 256 MB for AMD versus 144 MB for Intel. The L1 and L2 caches are larger on Intel (112 KB and 2 MB per core versus 80 KB and 1 MB per core), but the shared L3 advantage for AMD is significant.

Memory bandwidth favors AMD heavily: 576.0 GB/s versus 409.6 GB/s, a 40.6% difference. The memory bus is twelve-channel on AMD versus eight-channel on Intel. PCIe lanes favor Intel slightly: 136 versus 128, both Gen 5.

The EPYC 9355P launched on October 9, 2024, with a launch MSRP of $2998. The Xeon 6732P launched on May 21, 2025, with a launch MSRP of $5295. The AMD part became available roughly seven months earlier. Both processors have part numbers listed in the database: 100-000001521 for AMD and SRVP2 for Intel.

FAQ

Q: Which processor has higher single-threaded performance?

A: The AMD EPYC 9355P is significantly ahead. In PassMark single-thread testing, it scores 3,747 versus 2,506 for the Intel Xeon 6732P, a 49.5% advantage. Cinebench R23 single-core confirms this with a 29.9% lead (11,670 vs 8,981).

Q: How do the two processors compare in memory bandwidth?

A: The AMD EPYC 9355P supports twelve-channel DDR5 memory with 576.0 GB/s of bandwidth. The Intel Xeon 6732P supports eight-channel DDR5 with 409.6 GB/s. This gives AMD a 40.6% bandwidth advantage.

Q: Is there any workload where the Intel Xeon 6732P wins?

A: Yes, in the PassMark floating-point math test, the Xeon 6732P scores 261,703 versus 256,635 for the EPYC 9355P, a 1.9% margin. This is the only benchmark win for Intel in the head-to-head comparison.

Q: What are the core and thread counts for each processor?

A: Both processors have 32 cores and 64 threads. The difference lies in clock speeds: Intel has a higher base clock (3.80 GHz vs 3.55 GHz), while AMD has a higher boost clock (4.40 GHz vs 4.10 GHz).

Q: How do the cache configurations differ?

A: The AMD EPYC 9355P has 80 KB of L1 and 1 MB of L2 per core, with 256 MB of shared L3 cache. The Intel Xeon 6732P has 112 KB of L1 and 2 MB of L2 per core, with 144 MB of shared L3 cache. AMD's L3 cache is 112 MB larger.

Q: Which processor has higher power consumption?

A: The Intel Xeon 6732P has a TDP of 350 watts, while the AMD EPYC 9355P has a TDP of 280 watts. The Intel part has a 25% higher TDP rating.

The Verdict

The data is unambiguous: the AMD EPYC 9355P is the superior processor for nearly every workload measured. It wins 16 of 17 head-to-head benchmarks and holds a 11.8% advantage in average benchmark score (160,358 vs 143,444). The only Intel victory is a narrow 1.9% margin in floating-point math, which is unlikely to influence most purchasing decisions.

For single-threaded performance, the EPYC 9355P is the clear choice. Its 49.5% lead in PassMark single-thread and 29.9% lead across all Cinebench single-core tests make it suitable for workloads that depend on per-core speed. The higher boost clock of 4.40 GHz, combined with the Zen 5 architecture, delivers results that the Xeon 6732P cannot match.

For multithreaded workloads, the EPYC 9355P maintains a consistent 29.9% lead in Cinebench multicore tests and a 29.1% lead in PassMark multithread. The 256 MB L3 cache and 576.0 GB/s memory bandwidth provide additional advantages for data-heavy applications. The 66.7% lead in physics and 66.2% lead in prime number finding indicate particular strength in computational intensity.

The Intel Xeon 6732P is only competitive in floating-point math and offers more PCIe lanes (136 vs 128). It also has a higher base clock (3.80 GHz vs 3.55 GHz), but the lower boost clock (4.10 GHz vs 4.40 GHz) and smaller L3 cache (144 MB vs 256 MB) limit its practical performance. Its 350 watt TDP is also 70 watts higher than the AMD part.

The launch dates and MSRP figures are notable: the EPYC 9355P launched on October 9, 2024, with a launch MSRP of $2998, while the Xeon 6732P launched on May 21, 2025, with a launch MSRP of $5295. The AMD processor became available earlier and carries a lower launch MSRP.

For server and workstation buyers, the EPYC 9355P is the recommended choice across the board. Its wins in single-thread, multithread, encryption, compression, and physics workloads cover the majority of enterprise computing scenarios. The Xeon 6732P should only be considered for deployments where floating-point throughput is the dominant requirement, or where the additional 8 PCIe lanes are essential. In all other cases, the recorded data points to the AMD EPYC 9355P as the higher-performing processor.