AMD EPYC 7C13 vs AMD EPYC 9335 Comparison

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD EPYC 9335 posts an average benchmark score of 194,228, placing it in the 99th percentile of all CPUs. The EPYC 7C13 averages 167,788, sitting in the 98th percentile. The 9335 leads by roughly 15.8% in average score.

Q: How do the two compare in single-threaded performance?

A: The EPYC 9335 wins the PassMark single-thread test with a score of 2,732 against the 7C13’s 2,618, a 4.4% advantage. The 9335’s boost clock of 4.40 GHz versus 3.68 GHz on the 7C13 helps explain this edge.

Q: Which chip wins more head-to-head benchmark tests?

A: The EPYC 7C13 wins 8 of the 11 head-to-head tests, while the 9335 takes 3. The 7C13 dominates in throughput-oriented workloads like data compression, encryption, and integer math, but the 9335 wins in extended instructions and single-thread tasks.

Q: What is the core and thread count difference?

A: The EPYC 7C13 has 64 cores and 128 threads, double the 32 cores and 64 threads of the EPYC 9335. This core advantage directly drives the 7C13’s wins in heavily parallel workloads.

Q: Do these processors use the same memory technology?

A: No. The EPYC 9335 supports DDR5 memory on a twelve-channel bus with 576.0 GB/s bandwidth. The EPYC 7C13 uses DDR4 on an eight-channel bus with 204.8 GB/s bandwidth. The 9335 has nearly three times the memory bandwidth.

Q: What is the process node difference?

A: The EPYC 9335 is built on a 4 nm process by TSMC, while the EPYC 7C13 uses a 7 nm process, also from TSMC. The 9335’s smaller node contributes to higher clock speeds and greater efficiency per core.

Where Each One Wins

The EPYC 7C13 is the clear winner in raw multi-threaded throughput. Its 64 cores and 128 threads deliver decisive victories in passmark_multithread (76,322 vs 65,811), integer math (492,554 vs 346,291), and floating-point math (266,846 vs 228,123). For workloads that scale linearly with core count—database compression, encryption, physics simulations—the 7C13 is the stronger choice. Its data compression score of 1,562,251 is 23% higher than the 9335’s 1,203,096, and its encryption score of 114,769 beats the 9335 by 45%.

The EPYC 9335 wins where per-core efficiency and modern instruction sets matter. Its 4.4% single-thread advantage (2,732 vs 2,618) and 24.3% lead in extended instructions (105,706 vs 85,034) point to newer architecture benefits. The 9335 also has a massive memory bandwidth advantage—576.0 GB/s over 204.8 GB/s—which can be decisive for memory-bound workloads even when core counts are lower. The 9335’s higher boost clock (4.40 GHz vs 3.68 GHz) reinforces its single-thread edge.

The split is clear: the 7C13 is a brute-force parallel processor, while the 9335 is a more balanced design with superior per-core performance and memory throughput.

Architecture Differences

These are two generations apart in AMD’s EPYC lineage. The EPYC 9335 belongs to the EPYC 9005 series, codenamed Turin, built on Zen 5 architecture. The EPYC 7C13 uses Zen 3 architecture, codenamed Milan. The process node difference is substantial: 4 nm for the 9335 versus 7 nm for the 7C13, both fabricated by TSMC.

The 9335’s die configuration uses 4x 70.6 mm² chiplets, while the 7C13 employs 8x 81 mm² chiplets. Transistor counts are nearly identical—33,260 million for the 9335 versus 33,200 million for the 7C13—but they are packed into very different die layouts.

Cache hierarchies differ significantly. The 9335 has 80 KB of L1 and 1 MB of L2 per core, with 128 MB of shared L3. The 7C13 has 64 KB of L1 and 512 KB of L2 per core, but a larger 256 MB of shared L3. The 9335’s per-core L2 is double the 7C13’s, but the 7C13 has twice the total L3 capacity.

Memory architecture is another major divergence. The 9335 uses DDR5 with a twelve-channel bus and 576.0 GB/s bandwidth. The 7C13 uses DDR4 with eight channels and 204.8 GB/s. Both support ECC memory, but the 9335’s bandwidth advantage is nearly 3x.

PCIe generations also differ: the 9335 offers Gen 5 with 128 lanes (CPU only), while the 7C13 provides Gen 4 with the same 128 lanes. Sockets are incompatible—SP5 for the 9335, SP3 for the 7C13.

Specification Differences

| Specification | AMD EPYC 9335 | AMD EPYC 7C13 |

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

| Cores | 32 | 64 |

| Threads | 64 | 128 |

| Base Clock | 3.00 GHz | 2000.00 MHz (2.00 GHz) |

| Boost Clock | 4.40 GHz | 3.68 GHz |

| TDP | 210 W | 225 W |

| Socket | AMD Socket SP5 | AMD Socket SP3 |

| Architecture | Zen 5 | Zen 3 |

| Codename | Turin | Milan |

| Process Node | 4 nm | 7 nm |

| Die Size | 4x 70.6 mm² | 8x 81 mm² |

| L1 Cache | 80 KB (per core) | 64 KB (per core) |

| L2 Cache | 1 MB (per core) | 512 KB (per core) |

| L3 Cache | 128 MB (shared) | 256 MB (shared) |

| Memory Support | DDR5 | DDR4 |

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

| Memory Bandwidth | 576.0 GB/s | 204.8 GB/s |

| PCIe | Gen 5, 128 Lanes | Gen 4, 128 Lanes |

| Launch MSRP | $3178 | Not available |

Head-to-Head Benchmarks

The EPYC 7C13 dominates the head-to-head matchup, winning 8 of 11 tests. The largest victory comes in passmark_physics, where the 7C13 scores 4,904 against the 9335’s 1,905—a 61.2% advantage. This is the single biggest gap in either direction, reflecting the 7C13’s massive core advantage in physics simulation workloads.

Data encryption shows the second-largest delta: 114,769 for the 7C13 versus 63,159 for the 9335, a 45% lead. Prime number finding follows at 539 versus 340 (36.9% ahead), and integer math at 492,554 versus 346,291 (29.7% ahead). Data compression rounds out the 7C13’s strong showings with 1,562,251 versus 1,203,096 (23% ahead).

The 7C13 also wins in floating-point math (266,846 vs 228,123, up 14.5%), multithread (76,322 vs 65,811, up 13.8%), and random string sorting (131,361 vs 116,608, up 11.2%). These wins are consistent: the 7C13’s double core count powers through parallel workloads with leads ranging from 11% to 61%.

The EPYC 9335’s three wins are narrower but meaningful. Its extended instructions score of 105,706 beats the 7C13’s 85,034 by 24.3%, the largest 9335 victory. Single-thread performance shows a 4.4% edge (2,732 vs 2,618), appearing in both passmark_single_thread and passmark_singlethread. These wins suggest the 9335’s Zen 5 architecture extracts more work per core, particularly with modern instruction sets.

The average benchmark scores tell a complementary story: the 9335’s 194,228 average versus 167,788 for the 7C13. Despite losing most head-to-head tests, the 9335’s higher average indicates that its wins are weighted favorably in aggregate scoring, likely due to the extended instructions and single-thread results.

The Verdict

The data points to a clear split between two very different server processors. The EPYC 7C13 is the choice for workloads that can use every available core. Its 64 cores and 128 threads produce consistent 11% to 61% wins across compression, encryption, physics, integer math, and floating-point tasks. For HPC-style parallel workloads, database compression, or cryptography, the 7C13’s 8-of-11 head-to-head record makes a compelling case.

The EPYC 9335 is the pick for single-thread-sensitive applications and modern instruction set utilization. Its 4.4% single-thread lead and 24.3% extended instructions advantage, combined with 3x memory bandwidth (576.0 GB/s vs 204.8 GB/s) and DDR5 support, position it for memory-bound workloads and newer software optimizations. The 9335 also comes with a 99th percentile ranking versus the 7C13’s 98th, and a higher average benchmark score of 194,228 versus 167,788.

The 9335’s 4 nm process and Zen 5 architecture deliver higher clocks (4.40 GHz boost vs 3.68 GHz) and per-core cache improvements, but the 7C13’s 256 MB L3 cache and double core count remain potent for raw throughput. The 9335’s launch MSRP is $3178; the 7C13 has no listed launch MSRP.

For new deployments with DDR5 infrastructure, the 9335 offers future-proofing with Gen 5 PCIe and a smaller process node. For existing SP3 platforms or maximum parallel throughput, the 7C13’s benchmark dominance in multi-threaded tests is hard to ignore. The verdict depends on whether the workload favors many cores or faster cores—the data supports both interpretations, but the 7C13 wins more tests outright while the 9335 wins the aggregate score.