AMD EPYC 9135 vs Intel Xeon 6520P Comparison

Head-to-Head Benchmarks

The head-to-head comparison between the Intel Xeon 6520P and the AMD EPYC 9135 shows a decisive pattern: the Intel part wins 15 of the 17 recorded benchmarks, while the AMD EPYC takes the remaining 2. The margins, however, are not uniform, and the two AMD victories are worth examining closely.

Starting with the Cinebench suite, the Intel Xeon 6520P holds an 8.9% advantage across all six tests. In Cinebench R15 multicore, the Intel scores 5392 against the AMD's 4952. The single-core R15 test shows 761 versus 699, again an 8.9% delta. This consistency carries through Cinebench R20 and R23: multicore scores are 22467 versus 20637 and 53495 versus 49136 respectively, while single-core scores are 3171 versus 2913 and 7552 versus 6936. The identical 8.9% delta across every Cinebench test indicates a systematic performance difference rather than workload-specific behavior.

The PassMark suite tells a more varied story. The largest Intel win comes in the find prime numbers test, where the Xeon 6520P scores 526 against the EPYC's 292, a massive 80.1% difference. This is the single biggest delta in the entire comparison and suggests the Intel architecture handles this particular integer workload with exceptional efficiency. Floating point math also favors Intel heavily: 162862 versus 126679, a 28.6% margin. The physics test shows a 31.6% advantage for Intel (7209 versus 5477), and extended instructions come in at 15.6% ahead (64557 versus 55822). Data compression shows a 13.8% Intel lead (841518 versus 739277), while data encryption is 12.1% ahead (45188 versus 40295). PassMark multithread gives Intel a 10.1% edge (62936 versus 57170). The narrower Intel wins include integer math at 5.6% (214288 versus 202962) and random string sorting at 6.3% (95736 versus 90064).

The AMD EPYC 9135's two wins come in the PassMark single-thread tests, which appear twice in the database under slightly different test names. In both cases, the AMD scores 3672 against Intel's 3356, an 8.6% advantage for AMD. This is notable because the Cinebench single-core tests all favored Intel by 8.9%. The divergence between these two benchmark families suggests that the EPYC's higher boost clock of 4.30 GHz gives it an edge in certain single-threaded workloads, while the Xeon's architecture responds better to others.

The average benchmark scores reflect the overall picture. The Intel Xeon 6520P has an average benchmark score of 93786, while the AMD EPYC 9135 sits at 82980. Both processors rank in the 96th percentile among all CPUs in the database, indicating that either chip is a high-end option. The nearest rivals for the Intel part include the AMD EPYC 4564P at 95183 (1.5% higher), the AMD EPYC 9175F at 95615 (1.9% higher), and the AMD EPYC 4565P at 95764 (2.1% higher). The AMD EPYC 9135's nearest rivals include the AMD EPYC 4584PX at 83090 (0.1% higher), the Intel Core Ultra 9 285K at 83807 (1.0% higher), and the Intel Core Ultra 9 290K Plus at 84003 (1.2% higher). The EPYC 9135 also sits 1.6% above the AMD EPYC 7443P, which scores 81661.

FAQ

Q: Which processor wins more head-to-head benchmarks?

A: The Intel Xeon 6520P wins 15 of the 17 head-to-head benchmark comparisons. The AMD EPYC 9135 wins only the two PassMark single-thread tests.

Q: How large is the Intel advantage in Cinebench tests?

A: The Intel Xeon 6520P leads by 8.9% in all six Cinebench tests, covering R15, R20, and R23 in both single-core and multicore configurations.

Q: Where does the AMD EPYC 9135 outperform the Intel Xeon 6520P?

A: The AMD EPYC 9135 scores 3672 in the PassMark single-thread tests, compared to 3356 for the Intel part, an 8.6% advantage. This is the only benchmark category where AMD comes out ahead.

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

A: The PassMark find prime numbers test shows the biggest difference, with the Intel Xeon 6520P scoring 526 against the AMD EPYC 9135's 292, an 80.1% margin in favor of Intel.

Q: How do the two processors compare in average benchmark scores?

A: The Intel Xeon 6520P has an average benchmark score of 93786, while the AMD EPYC 9135 averages 82980. Both sit in the 96th percentile of all CPUs in the database.

Q: Are there any benchmark categories where the two processors are closely matched?

A: The closest margins are in PassMark integer math, where Intel leads by 5.6% (214288 versus 202962), and PassMark random string sorting, where Intel leads by 6.3% (95736 versus 90064).

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Xeon 6520P uses the Granite Rapids architecture, part of the Xeon 6 generation built on a 5 nm process at Intel's own foundry. The die size is listed at 598 mm². The AMD EPYC 9135 belongs to the EPYC 9005 series, using the Zen 5 architecture with the codename Turin, manufactured on a 4 nm process at TSMC. The AMD chip uses two chiplets, each measuring 70.6 mm², with a total transistor count of 16,630 million.

Core counts differ substantially. The Intel Xeon 6520P packs 24 cores and 48 threads, while the AMD EPYC 9135 offers 16 cores and 32 threads. This core advantage helps explain the Intel multicore wins, though the AMD chip compensates with higher clock speeds: a base clock of 3.65 GHz and a boost clock of 4.30 GHz, compared to the Intel's 2.40 GHz base and 4.00 GHz boost.

Cache hierarchies also diverge. The Intel part has 112 KB of L1 cache per core, 2 MB of L2 cache per core, and 144 MB of shared L3 cache. The AMD EPYC 9135 has 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of shared L3 cache. The Intel processor's larger cache, particularly the 144 MB L3, likely contributes to its strong showing in data compression and encryption tests.

Memory architecture is another point of separation. Both support DDR5 and ECC memory, but the Intel Xeon 6520P uses an eight-channel memory bus with 409.6 GB/s of bandwidth, while the AMD EPYC 9135 uses a twelve-channel bus with 576.0 GB/s. The AMD chip offers more memory bandwidth on paper, though the benchmark data does not isolate memory-bound workloads.

PCIe connectivity also differs. The Intel part provides Gen 5 with 88 lanes (CPU only), while the AMD EPYC 9135 offers Gen 5 with 128 lanes (CPU only). Neither has integrated graphics. The Intel processor uses Socket 4710, the AMD uses Socket SP5. Both are currently active production parts, with the AMD released on 2024-10-09 and the Intel on 2025-02-23.

Specification Differences

The core specifications show clear divergence across nearly every field. The Intel Xeon 6520P has 24 cores and 48 threads, while the AMD EPYC 9135 has 16 cores and 32 threads. Base clocks differ significantly: 2.40 GHz for Intel versus 3.65 GHz for AMD. Boost clocks are closer but still distinct: 4.00 GHz for Intel versus 4.30 GHz for AMD. Thermal design power is 210 W for the Intel part and 200 W for the AMD part.

Cache configurations differ at every level. The Intel L1 cache is 112 KB per core versus 80 KB per core for AMD. L2 is 2 MB per core versus 1 MB per core. L3 is 144 MB shared for Intel versus 64 MB shared for AMD.

Memory channels and bandwidth favor AMD: twelve channels and 576.0 GB/s versus eight channels and 409.6 GB/s for Intel. PCIe lanes favor AMD as well: 128 lanes versus 88 lanes, both Gen 5.

Manufacturing details also differ. The Intel part uses a 5 nm process at Intel's foundry with a 598 mm² die. The AMD part uses a 4 nm process at TSMC with two 70.6 mm² chiplets and 16,630 million transistors. The Intel part's transistor count is not recorded in the database.

The launch MSRP for the Intel Xeon 6520P is $1295. The launch MSRP for the AMD EPYC 9135 is $1214. Neither processor has an unlocked multiplier. Part numbers are SRVNQ for Intel and 100-000001150 for AMD.

Where Each One Wins

The Intel Xeon 6520P is the clear winner for multicore and heavily threaded workloads. Its 24 cores and 48 threads give it a structural advantage that shows up across the Cinebench multicore tests, where it leads by 8.9% consistently. The PassMark multithread test confirms this with a 10.1% edge. The largest Intel wins come in specialized workloads: find prime numbers (80.1% ahead), physics (31.6% ahead), and floating point math (28.6% ahead). Data compression, encryption, and extended instructions all favor Intel by double-digit margins. For workloads that rely on large shared caches, the 144 MB L3 cache on the Intel part is a substantial asset compared to the AMD's 64 MB.

The AMD EPYC 9135 wins in one specific area: PassMark single-thread performance. Its 8.6% advantage in that test likely stems from its higher boost clock of 4.30 GHz. The AMD part also offers more memory bandwidth (576.0 GB/s versus 409.6 GB/s) and more PCIe lanes (128 versus 88), which the database shows as specification advantages even though they do not translate into benchmark wins in the recorded tests. The AMD chip is also more power-efficient in terms of TDP, drawing 200 W versus 210 W for Intel, though it delivers fewer cores.

For workloads that fit within the AMD's strengths, such as lightly threaded tasks that benefit from high single-core speed, the EPYC 9135 is the better choice. For everything else in the recorded benchmark suite, the Intel Xeon 6520P holds the advantage.

The Verdict

The data points to a straightforward conclusion for most buyers. The Intel Xeon 6520P wins 15 of 17 benchmarks, including every multicore test and most single-core Cinebench tests. Its 80.1% lead in find prime numbers and 31.6% lead in physics make it the obvious pick for computational workloads. The 28.6% advantage in floating point math and 15.6% in extended instructions further cement its position for scientific and engineering applications.

The AMD EPYC 9135 is the choice for scenarios where single-threaded PassMark performance matters most, as it leads by 8.6% in that specific test. Its higher memory bandwidth and additional PCIe lanes are specification advantages that could matter in memory-heavy or I/O-heavy deployments, even though the recorded benchmarks do not show direct wins in those areas. The AMD part also has a lower TDP of 200 W versus 210 W for Intel, which may factor into dense server environments.

Both processors sit in the 96th percentile of all CPUs in the database, so neither is a weak option. But the benchmark record is unambiguous: the Intel Xeon 6520P delivers superior performance across the vast majority of measured workloads, and its larger core count and cache configuration provide a strong foundation for sustained multicore throughput. The AMD EPYC 9135 is a capable alternative with a single-thread edge and better memory subsystem specifications, but the data shows it trailing the Intel part in nearly every head-to-head comparison.