AMD EPYC 9535 vs Intel Xeon 6780E Comparison
AMD EPYC 9535
Xeon 6780E
PERFORMANCE BENCHMARKS
Analysis: AMD EPYC 9535 vs Intel Xeon 6780E
Where Each One Wins
The AMD EPYC 9535 and Intel Xeon 6780E serve distinctly different workloads despite both targeting the server and workstation segment. The benchmark data shows a clear split: the EPYC 9535 dominates in compute-heavy, latency-sensitive tasks, while the Xeon 6780E excels in memory-bandwidth and vectorized throughput scenarios.
The EPYC 9535 wins 7 of the 11 head-to-head comparisons. Its single-thread advantage is massive, scoring 3720 against the Xeon's 1923, a 93.4% lead. This translates directly into workloads that depend on per-core responsiveness, such as database query processing, code compilation, and container orchestration where individual threads cannot be parallelized further. The extended instructions test tells a similar story, with the EPYC 9535 posting 175784 versus 114008, a 54.2% margin that indicates superior SIMD and cryptography instruction handling.
The Xeon 6780E counters with 4 wins, all in areas where its higher core count and larger aggregate memory subsystem matter. Data compression shows the Xeon ahead at 2557582 versus 2308822, a 9.7% edge. Data encryption is an even bigger win, 193004 against 127372, a 34% gap. The physics benchmark is the most lopsided result in either direction, with the Xeon scoring 10951 versus 3834, a 65% advantage. Random string sorting also favors Intel, 326954 to 247506, a 24.3% lead.
For deployment decisions, the pattern is straightforward. If the workload is dominated by integer math, prime number generation, floating-point calculations, or multithreaded general compute, the EPYC 9535 is the stronger choice. Its multithread score of 114528 beats the Xeon's 86734 by 32%. If the workload involves encryption, compression, physics simulation, or heavy data shuffling, the Xeon 6780E's 144 cores and wider memory channels provide the edge. The data suggests these are complementary designs rather than direct substitutes.
Architecture Differences
The two processors come from fundamentally different design philosophies. The AMD EPYC 9535 uses the Zen 5 architecture, codenamed Turin, built on a 4 nm process at TSMC. The Intel Xeon 6780E uses Sierra Forest, built on Intel's 5 nm node. This process difference helps explain the clock speed gap: the EPYC 9535 runs at 2.40 GHz base and 4.30 GHz boost, while the Xeon 6780E runs at 2.20 GHz base and 3.00 GHz boost.
Core configurations diverge sharply. The EPYC 9535 has 64 cores and 128 threads, meaning each core supports two threads via SMT. The Xeon 6780E has 144 cores but only 144 threads, indicating a single-thread-per-core design. This explains why the EPYC wins single-thread tests by such a wide margin despite having fewer physical cores. The Xeon's approach prioritizes raw core count for parallel throughput, while AMD focuses on per-core efficiency and simultaneous multithreading.
Cache hierarchies are also different. The EPYC 9535 provides 80 KB L1 per core, 1 MB L2 per core, and 256 MB shared L3. The Xeon 6780E provides 96 KB L1 per core, 4 MB L2 per module, and 108 MB shared L3. The EPYC's much larger L3 cache supports its 64-core design with 128 threads, while the Xeon's smaller shared cache reflects a design where each core relies more on its own L2.
Memory architecture differs substantially. The EPYC 9535 supports twelve-channel DDR5 with 576.0 GB/s bandwidth. The Xeon 6780E supports eight-channel DDR5 with 409.6 GB/s. Despite the EPYC's higher bandwidth, the Xeon's higher physical core count can still saturate its memory subsystem in certain workloads, as seen in the compression and encryption results. PCIe lane counts also differ: the EPYC offers 128 Gen 5 lanes, the Xeon offers 88 Gen 5 lanes.
The EPYC 9535 uses 66,520 million transistors across 8 dies of 70.6 mm² each. The Xeon 6780E uses a single 578 mm² die. TDPs are 300 W for AMD and 330 W for Intel. Both support ECC memory and lack integrated graphics. The EPYC 9535 launched on 2024-10-09 with a launch MSRP of $8992. The Xeon 6780E launched on 2024-06-02 with a launch MSRP of $11350.
FAQ
Q: Which processor has better single-thread performance?
A: The AMD EPYC 9535 is far ahead. It scores 3720 in the single-thread PassMark test versus 1923 for the Intel Xeon 6780E, a 93.4% advantage. This is driven by the EPYC's higher boost clock of 4.30 GHz versus 3.00 GHz and its Zen 5 architecture.
Q: Why does the Intel Xeon 6780E win the physics benchmark by such a large margin?
A: The Xeon 6780E scores 10951 in the physics test versus 3834 for the EPYC 9535, a 65% lead. This likely reflects the Xeon's 144 physical cores, which can parallelize physics calculations that do not benefit from SMT or high per-core clock speeds.
Q: Which CPU supports more PCIe lanes?
A: The AMD EPYC 9535 provides 128 Gen 5 lanes (CPU only), while the Intel Xeon 6780E provides 88 Gen 5 lanes. This gives the EPYC more headroom for high-bandwidth peripherals such as GPUs, NVMe storage, and network adapters.
Q: How do the memory bandwidth figures compare?
A: The EPYC 9535 has a twelve-channel DDR5 memory bus delivering 576.0 GB/s, versus the Xeon 6780E's eight-channel bus at 409.6 GB/s. The EPYC offers roughly 40% more theoretical bandwidth, which supports its higher single-thread and multithread scores in memory-sensitive tests.
Q: Is the Intel Xeon 6780E always faster in multithreaded workloads?
A: No. Despite having 144 cores versus 64, the Xeon 6780E loses the multithread PassMark test, scoring 86734 versus 114528 for the EPYC 9535, a 32% deficit. The EPYC's SMT support (128 threads) and higher clocks compensate for the core count difference in general multithreaded code.
Q: What is the production status of both processors?
A: Both are listed as Active in the database. The AMD EPYC 9535 was released on 2024-10-09, and the Intel Xeon 6780E was released on 2024-06-02.
Specification Differences
The recorded data shows the following differences between the two processors.
| Specification | AMD EPYC 9535 | Intel Xeon 6780E |
|---|---|---|
| Cores | 64 | 144 |
| Threads | 128 | 144 |
| Base clock | 2.40 GHz | 2.20 GHz |
| Boost clock | 4.30 GHz | 3.00 GHz |
| TDP | 300 W | 330 W |
| Socket | AMD Socket SP5 | Intel Socket 4710 |
| Architecture | Zen 5 | Sierra Forest |
| Process node | 4 nm (TSMC) | 5 nm (Intel) |
| Transistors | 66,520 million | Not listed |
| Die size | 8x 70.6 mm² | 578 mm² |
| L1 cache | 80 KB per core | 96 KB per core |
| L2 cache | 1 MB per core | 4 MB per module |
| L3 cache | 256 MB shared | 108 MB shared |
| Memory bus | Twelve-channel | Eight-channel |
| Memory bandwidth | 576.0 GB/s | 409.6 GB/s |
| PCIe lanes | Gen 5, 128 lanes | Gen 5, 88 lanes |
| Launch MSRP | $8992 | $11350 |
| Release date | 2024-10-09 | 2024-06-02 |
| Part number | 100-000001147 | SRPG3 |
Both support DDR5 memory and ECC. Neither has integrated graphics. Both have locked multipliers. The EPYC 9535's percentile is 100 versus 99 for the Xeon 6780E.
Head-to-Head Benchmarks
The head-to-head results reveal where each architecture's strengths manifest. The largest win for the AMD EPYC 9535 is in single-thread performance, where it scores 3720 versus 1923, a 93.4% delta. This is the most decisive result in the comparison. The EPYC also wins extended instructions by 54.2% (175784 to 114008), multithread by 32% (114528 to 86734), and find prime numbers by 23.4% (874 to 708). Integer math goes to AMD by 13.8% (730281 to 641817), and floating point math by 5.3% (457047 to 433862).
The Intel Xeon 6780E's biggest win is in physics, where it nearly triples the EPYC's score: 10951 versus 3834, a 65% delta. Data encryption shows a 34% Intel lead (193004 to 127372). Random string sorting favors Intel by 24.3% (326954 to 247506). Data compression goes to Intel by 9.7% (2557582 to 2308822).
The multithread result deserves attention. The Xeon has 144 cores versus 64, yet loses by 32%. The EPYC's 128 threads, combined with its 4.30 GHz boost clock, produce a higher aggregate score. This suggests that SMT and clock speed matter more than raw core count for the PassMark multithread test. However, the Xeon's wins in physics and encryption indicate that certain workloads scale with physical cores and can overcome the per-core performance deficit.
The average benchmark scores from the database reinforce this split. The EPYC 9535 has an average score of 379408, placing it above the Intel Xeon 6960P (365194, 3.9% lower) and the AMD EPYC 9754 (364371, 4.1% lower). The Xeon 6780E averages 280438, nearly identical to the AMD Ryzen Threadripper 9970X (279778, 0.2% higher) and below the AMD EPYC 9565 (285471, 1.8% higher). The EPYC 9535 sits in the 100th percentile of all CPUs, while the Xeon 6780E sits in the 99th. These rankings show that the EPYC 9535 competes at the very top of the database, while the Xeon 6780E, despite its 144 cores, lands in a slightly lower tier for overall benchmark performance.
For encryption-heavy environments, the Xeon's 34% lead in data encryption is significant and likely stems from its core count advantage in AES operations. For general-purpose server workloads, the EPYC 9535's higher single-thread and multithread scores make it the better all-rounder. The database records 7 wins for the EPYC 9535 and 4 for the Xeon 6780E, a ratio that matches the narrative of AMD's per-core strength versus Intel's core-count strategy.