CPU Comparison

The AMD EPYC 9275F and Intel Xeon 6732P represent two distinct philosophies in server processing. The data shows a clear split: the EPYC 9275F dominates in synthetic rendering and single-threaded throughput, while the Xeon 6732P counters with significant wins in specific compute workloads. With 17 head-to-head benchmarks, the EPYC 9275F claims 12 victories, but the Xeon 6732P’s 5 wins are decisive in their respective categories. The following analysis breaks down these results, focusing exclusively on the benchmark scores and architectural facts provided.

Head-to-Head Benchmarks

The most striking pattern in the head-to-head data is the EPYC 9275F’s dominance across the Cinebench suite. In Cinebench R23 multi-core, the AMD part scores 71,927 against Intel’s 63,621, a delta of -11.5% from the Intel perspective. This margin is consistent across R15 and R20 multi-core tests, with deltas of -11.6% and -11.5% respectively. The single-core results are nearly identical in percentage terms: R23 single-core shows 10,154 versus 8,981 (-11.6%), and R15 single-core shows 1,023 versus 905 (-11.5%). This consistency suggests a fundamental per-core performance advantage for the AMD architecture, not a workload-specific anomaly.

The PassMark suite tells a more nuanced story. The EPYC 9275F wins the multi-threaded test with 84,620 against 74,849 (-11.5%), aligning with the Cinebench multi-core trend. However, the Intel Xeon 6732P posts decisive victories in several compute-heavy workloads. The largest win for Intel is in floating-point math: 261,703 versus 201,888, a delta of 29.6%. This is a substantial margin, indicating the Intel part’s AVX-512-style throughput advantage in this specific test. Extended instructions also favor Intel: 106,697 versus 94,889, a 12.4% delta.

The remaining Intel wins are narrower but still meaningful. Data compression shows 1,339,480 versus 1,212,560, a 10.5% delta. Integer math is closer: 334,340 versus 317,777, a 5.2% delta. Data encryption is the tightest contest, with Intel at 63,848 and AMD at 62,664, a 1.9% delta. These results indicate that while the EPYC 9275F has superior raw multi-threaded throughput, the Xeon 6732P excels in specific instruction-heavy and compression workloads.

The AMD part’s wins outside Cinebench are equally pronounced. Physics simulation shows a 12,089 versus 8,109 score, a 32.9% delta in favor of AMD. The largest single delta in the entire dataset is in PassMark single-thread: 3,810 versus 2,506, a 34.2% delta for AMD. Prime number finding also heavily favors AMD: 991 versus 628, a 36.6% delta. Random string sorting is closer, with AMD at 144,037 versus Intel’s 133,467, a 7.3% delta.

The average benchmark scores contextualize these head-to-head results. The Xeon 6732P has an average score of 143,444, placing it in the 98th percentile of all CPUs. The EPYC 9275F’s average is 133,174, in the 97th percentile. Interestingly, the Intel part’s nearest rival is the AMD Ryzen 9 PRO 9965X3D with an average score of 143,735 (-0.2% delta), while the EPYC 9275F’s nearest rival is the Intel Xeon 6710E at 129,930 (2.5% delta). This suggests that the Xeon 6732P’s overall benchmark profile is slightly higher than the EPYC 9275F’s, despite losing the majority of direct comparisons.

Where Each One Wins

The EPYC 9275F is the clear winner in rendering and general-purpose multi-threaded workloads. All four Cinebench tests (R15, R20, R23, both single and multi-core) favor AMD, with consistent deltas around -11.5%. The PassMark multi-threaded test also favors AMD, reinforcing the pattern that the 24-core EPYC 9275F with 48 threads outperforms the 32-core Xeon 6732P with 64 threads in these synthetic loads. This is notable because the Intel part has 8 more cores and 16 more threads, yet cannot match the AMD part’s throughput in these tests.

The EPYC 9275F also dominates in single-threaded performance. The PassMark single-thread score of 3,810 versus 2,506 represents a 34.2% advantage, and the Cinebench R23 single-core score of 10,154 versus 8,981 confirms this trend. This is directly attributable to the higher base and boost clocks: the AMD part runs at 4.10 GHz base and 4.80 GHz boost, while the Intel part runs at 3.80 GHz base and 4.10 GHz boost. For workloads that are latency-sensitive or rely on a few fast cores, the EPYC 9275F is the superior choice.

The Xeon 6732P wins in specific compute-intensive domains. Floating-point math is its strongest category, with a 29.6% delta over the EPYC 9275F. This is a significant margin that would matter for scientific computing, financial modeling, or any workload relying heavily on FP32/FP64 operations. Extended instructions (12.4% delta) and data compression (10.5% delta) are also Intel wins, suggesting strengths in encryption, compression algorithms, and SIMD-heavy code paths. Integer math is a narrower win at 5.2%, but still indicates a slight edge in general integer arithmetic.

Data encryption is the closest contest, with Intel winning by just 1.9%. This near-tie suggests that both parts are well-suited for cryptographic workloads, but the Intel part has a marginal edge. The Xeon 6732P’s larger L3 cache (144 MB shared versus 256 MB shared for AMD) does not translate to a win in every test, but the 144 MB cache is still substantial and likely contributes to the compression and integer wins.

The Verdict

The data presents a clear choice based on workload priorities. For users running Cinebench-style rendering, physics simulations, or any single-threaded heavy application, the AMD EPYC 9275F is the definitive pick. Its 11.5% advantage in multi-core Cinebench, 32.9% advantage in physics, and 34.2% advantage in single-threaded PassMark make it the stronger all-around processor for these tasks. The higher clock speeds (4.10 GHz base, 4.80 GHz boost) directly explain this advantage.

For compute-heavy scientific and financial workloads, the Intel Xeon 6732P is the better option. Its 29.6% win in floating-point math is the largest margin in the entire dataset, and the 12.4% win in extended instructions indicates a strong SIMD implementation. Data compression (10.5% delta) and integer math (5.2% delta) are also Intel advantages, making this part suitable for database, compression, and general server compute tasks that leverage these specific instruction paths.

The average benchmark scores complicate a simple recommendation. The Xeon 6732P’s average of 143,444 is 7.7% higher than the EPYC 9275F’s 133,174, and the Intel part sits in the 98th percentile versus AMD’s 97th. This suggests that on a balanced mix of workloads, the Intel part edges ahead. However, the EPYC 9275F wins 12 of 17 direct comparisons, indicating that its wins are more numerous, while Intel’s wins are more concentrated in specific high-margin categories.

The nearest rival data further clarifies the positioning. The Xeon 6732P is most closely matched by the AMD Ryzen 9 PRO 9965X3D (-0.2% delta) and the Intel Xeon w9-3575X (-0.6% delta), while the EPYC 9275F is most closely matched by the Intel Xeon 6710E (2.5% delta) and the AMD EPYC 9354 (5% delta). This indicates that the EPYC 9275F is slightly ahead of its closest competitors, while the Xeon 6732P is essentially tied with its nearest rivals, with one rival (Xeon 674X) slightly ahead at 0.2% delta.

FAQ

Q: Which processor has the higher overall benchmark average?

A: The Intel Xeon 6732P has an average benchmark score of 143,444, which is higher than the AMD EPYC 9275F’s average of 133,174.

Q: How significant is the difference in single-threaded performance?

A: The AMD EPYC 9275F wins the PassMark single-thread test with 3,810 against Intel’s 2,506, a 34.2% delta. The Cinebench R23 single-core score is 10,154 versus 8,981, an 11.6% delta.

Q: Does the Intel part win any benchmark by a large margin?

A: Yes. The Intel Xeon 6732P wins floating-point math by 29.6% (261,703 versus 201,888) and extended instructions by 12.4% (106,697 versus 94,889).

Q: What is the core and thread configuration of each processor?

A: The Intel Xeon 6732P has 32 cores and 64 threads, while the AMD EPYC 9275F has 24 cores and 48 threads.

Q: How does the cache configuration differ between the two?

A: The Intel part has 144 MB of shared L3 cache, while the AMD part has 256 MB of shared L3 cache. Both have per-core L1 and L2 caches, with Intel at 112 KB and 2 MB per core, and AMD at 80 KB and 1 MB per core.

Q: What are the memory bandwidth specifications?

A: The AMD EPYC 9275F supports twelve-channel DDR5 with a maximum bandwidth of 576.0 GB/s, while the Intel Xeon 6732P supports eight-channel DDR5 with a maximum bandwidth of 409.6 GB/s.

Architecture Differences

The architectural divide between these two processors is substantial. The Intel Xeon 6732P is built on Granite Rapids architecture, using a 5 nm process node from Intel’s own foundry. It features 32 cores and 64 threads, with a base clock of 3.80 GHz and a boost clock of 4.10 GHz. The thermal design power is 350 W, and it uses the Intel Socket 4710. The cache hierarchy includes 112 KB of L1 per core, 2 MB of L2 per core, and a shared 144 MB L3 cache.

The AMD EPYC 9275F is based on Zen 5 architecture (codenamed Turin) and is part of the EPYC 9005 series. It is manufactured on a 4 nm process node by TSMC, with 66,520 million transistors spread across 8 chiplets, each measuring 70.6 mm². This is a chiplet design, unlike the Intel monolithic approach. The AMD part has 24 cores and 48 threads, with a base clock of 4.10 GHz and a boost clock of 4.80 GHz. The TDP is 320 W, and it uses the AMD Socket SP5. The cache configuration is 80 KB of L1 per core, 1 MB of L2 per core, and a shared 256 MB L3 cache.

Memory support differs significantly. The AMD EPYC 9275F supports twelve-channel DDR5 memory with a peak bandwidth of 576.0 GB/s, while the Intel Xeon 6732P supports eight-channel DDR5 with 409.6 GB/s. This gives AMD a 40.6% bandwidth advantage, which aligns with its multi-threaded wins. Both support ECC memory. PCIe connectivity is similar, with Intel offering Gen 5 with 136 lanes and AMD offering Gen 5 with 128 lanes (CPU-only).

The release dates differ by about seven months, with the AMD part launching on 2024-10-09 and the Intel part on 2025-05-21. The launch MSRP for the Intel Xeon 6732P is $5295, while the AMD EPYC 9275F has a launch MSRP of $3439. Neither processor has an unlocked multiplier, and both target the server/workstation market segment. The Intel part’s part number is SRVP2, while AMD’s is 100-000001144. Both are currently in active production.