AMD EPYC 9335 vs Intel Xeon 6741P Comparison

The Intel Xeon 6741P and AMD EPYC 9335 are both 99th-percentile server processors, but their benchmark profiles could not be more different. The Xeon 6741P wins every single head-to-head benchmark in the data set, while the EPYC 9335 counters with a substantial advantage in memory bandwidth and a lower thermal envelope. The average benchmark scores are nearly identical — 194901 for the Intel part versus 194228 for the AMD part, a margin of just 0.3% — yet the distribution of performance is heavily skewed toward the Intel chip in compute-heavy workloads.

Head-to-Head Benchmarks

The Xeon 6741P dominates the PassMark suite with double-digit wins across the board. The most dramatic gap is in PassMark physics, where the Intel part scores 13890 against the EPYC’s 1905 — a 629.1% advantage. That is not a typo; the Xeon is over seven times faster in that specific physics simulation test. The find prime numbers test shows a similar pattern of Intel superiority, with scores of 1242 versus 340, a 265.3% delta. These two results suggest the Xeon’s architecture handles certain algorithmic workloads with far greater efficiency.

In floating point math, the Xeon 6741P scores 358423 versus 228123 for the EPYC 9335, a 57.1% lead. Integer math follows with 458058 against 346291, a 32.3% advantage. Multithread performance tells the same story: 100660 for the Intel chip versus 65811 for AMD, a 53% gap. Data compression shows 1816408 versus 1203096, a 51% win. Random string sorting is 177322 versus 116608, good for a 52.1% lead. Extended instructions come in at 142682 versus 105706, a 35% margin. Data encryption shows 89746 versus 63159, a 42.1% advantage.

Even single-thread performance, where AMD’s higher clock speeds might suggest an advantage, goes to Intel. The Xeon 6741P scores 3195 in PassMark single-thread versus 2732 for the EPYC 9335, a 16.9% lead. This is notable because the EPYC boasts a boost clock of 4.40 GHz compared to the Xeon’s 3.80 GHz, yet the Intel architecture still wins the single-core race. Across all 11 head-to-head benchmarks, Intel wins 11 and AMD wins 0. The closest contest is extended instructions at 35%, while the average delta across all tests sits around 50% in Intel’s favor.

Where Each One Wins

The Intel Xeon 6741P wins in every benchmark category measured here, so the use-case split is less about workload type and more about workload scale and system constraints. The Xeon’s 48 cores and 96 threads give it a raw parallel processing advantage over the EPYC’s 32 cores and 64 threads. For heavily threaded workloads like physics simulation, data compression, and floating point math, the Xeon is the clear choice. The 629.1% physics delta and 265.3% prime number delta make it obvious that certain compute patterns strongly favor Intel’s Granite Rapids design.

The AMD EPYC 9335 does not win any benchmark in this data, but it holds two structural advantages that matter in real deployments. Its twelve-channel memory bus supports a theoretical bandwidth of 576.0 GB/s, which is 40.6% higher than the Xeon’s 409.6 GB/s from its eight-channel configuration. For memory-bandwidth-bound workloads — large database scans, in-memory analytics, or HPC kernels that stream data — the EPYC could close the gap or even overtake the Xeon in real-world performance, even though the synthetic PassMark suite does not reflect that. Additionally, the EPYC’s TDP is 210 watts versus 300 watts for the Xeon, making it a more manageable part for dense server chassis or power-constrained environments.

The EPYC also has a higher base clock of 3.00 GHz versus 2.50 GHz, which gives it an idle-to-load responsiveness advantage in lightly threaded tasks, despite losing the single-thread benchmark. For workloads that are moderate in thread count but heavy on memory traffic, the EPYC’s twelve-channel memory subsystem is the counterargument to Intel’s core count lead.

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Xeon 6741P uses the Granite Rapids architecture on a 5 nm process, fabricated by Intel. It packs 48 cores and 96 threads, with a die size of 2x 598 mm² — a massive silicon footprint. The AMD EPYC 9335 uses Zen 5 architecture on a 4 nm process from TSMC, with 32 cores and 64 threads spread across 4x 70.6 mm² chiplets, totaling 33,260 million transistors.

Cache hierarchies differ substantially. The Xeon allocates 112 KB of L1 and 2 MB of L2 per core, with a massive 288 MB of shared L3. The EPYC offers 80 KB L1 and 1 MB L2 per core, with 128 MB of shared L3. The Intel part has more than double the L3 cache, which explains its dominance in data compression and random string sorting where large working sets benefit from on-chip storage.

Memory channels are a key differentiator. The Xeon uses eight-channel DDR5 with 409.6 GB/s bandwidth. The EPYC uses twelve-channel DDR5 with 576.0 GB/s bandwidth. This is the single area where AMD has a clear architectural edge. PCIe lanes are close: Intel provides Gen 5 with 136 lanes, while AMD provides Gen 5 with 128 lanes. Both support ECC memory and have no integrated graphics.

The process node difference (Intel 5 nm vs TSMC 4 nm) explains the transistor density advantage for AMD, but the Xeon’s larger die and higher core count deliver more raw compute throughput. The EPYC’s smaller chiplets and lower TDP (210W vs 300W) make it a more efficient part per watt, though the benchmark data does not include power efficiency scores to quantify that directly.

FAQ

Q: Which processor has a higher single-thread score?

A: The Intel Xeon 6741P scores 3195 in PassMark single-thread, which is 16.9% higher than the AMD EPYC 9335’s 2732, despite the EPYC having a higher boost clock of 4.40 GHz versus 3.80 GHz.

Q: How much larger is the Intel Xeon’s L3 cache?

A: The Xeon 6741P has 288 MB of shared L3, which is 160 MB more than the EPYC 9335’s 128 MB. That is a 125% increase in on-chip cache capacity.

Q: What is the memory bandwidth difference between the two?

A: The AMD EPYC 9335 has a twelve-channel memory bus with 576.0 GB/s theoretical bandwidth, while the Intel Xeon 6741P has an eight-channel bus with 409.6 GB/s. The EPYC offers 166.4 GB/s more bandwidth.

Q: Which chip has more cores and threads?

A: The Intel Xeon 6741P has 48 cores and 96 threads, compared to the AMD EPYC 9335’s 32 cores and 64 threads. The Xeon provides 16 additional cores and 32 additional threads.

Q: How does the power draw compare?

A: The Intel Xeon 6741P has a TDP of 300 watts, while the AMD EPYC 9335 has a TDP of 210 watts. The EPYC draws 90 watts less under maximum rated load.

Q: What is the release date and launch MSRP for each?

A: The Intel Xeon 6741P launched on 2025-02-23 with a launch MSRP of $4421. The AMD EPYC 9335 launched earlier on 2024-10-09 with a launch MSRP of $3178.

The Verdict

The data is unambiguous: for raw compute performance across every benchmark in the PassMark suite, the Intel Xeon 6741P is the better processor. It wins all 11 head-to-head tests, with an average score of 194901 versus 194228 for the EPYC, and it does so while offering 48 cores versus 32. The 629.1% physics advantage and 265.3% prime number advantage show that certain algorithmic workloads will run dramatically faster on the Intel part.

However, the AMD EPYC 9335 is the better choice for memory-bandwidth-sensitive applications. Its 576.0 GB/s twelve-channel memory bandwidth is 40.6% higher than the Xeon’s 409.6 GB/s, making it the superior platform for workloads that stream large datasets. The EPYC also consumes 90 fewer watts at TDP, which simplifies cooling and power delivery in dense server deployments. Its earlier release date and lower launch MSRP of $3178 versus $4421 make it the more economical platform choice, though the Xeon’s performance lead justifies its premium for compute-bound users.

Choose the Intel Xeon 6741P if your workloads are dominated by floating point math, integer math, physics simulation, or data compression. Choose the AMD EPYC 9335 if your bottleneck is memory bandwidth or if power density and thermal management are critical constraints.

Specification Differences

| Specification | Intel Xeon 6741P | AMD EPYC 9335 |

|---|---|---|

| Cores | 48 | 32 |

| Threads | 96 | 64 |

| Base Clock | 2.50 GHz | 3.00 GHz |

| Boost Clock | 3.80 GHz | 4.40 GHz |

| TDP | 300 W | 210 W |

| Socket | Intel Socket 4710 | AMD Socket SP5 |

| Architecture | Granite Rapids | Zen 5 |

| Codename | Granite Rapids | Turin |

| Process Node | 5 nm | 4 nm |

| Foundry | Intel | TSMC |

| Die Size | 2x 598 mm² | 4x 70.6 mm² |

| Transistors | N/A | 33,260 million |

| L1 Cache | 112 KB (per core) | 80 KB (per core) |

| L2 Cache | 2 MB (per core) | 1 MB (per core) |

| L3 Cache | 288 MB (shared) | 128 MB (shared) |

| Memory Bus | Eight-channel | Twelve-channel |

| Memory Bandwidth | 409.6 GB/s | 576.0 GB/s |

| PCIe | Gen 5, 136 Lanes | Gen 5, 128 Lanes |

| Release Date | 2025-02-23 | 2024-10-09 |

| Launch MSRP | $4421 | $3178 |

| Part Number | SRVEY | 100-000001149 |