AMD EPYC 8024P vs AMD Ryzen 5 PRO 5655GE Comparison

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD EPYC 8024P has an average benchmark score of 26555, while the AMD Ryzen 5 PRO 5655GE scores 25880. The EPYC is ahead by roughly 2.6%, and both CPUs sit at the 78th percentile among all processors in the database.

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

A: The EPYC 8024P wins every Cinebench single-core test by a consistent 12.7% margin. In Cinebench R23 single-core, the EPYC scores 2466 versus the Ryzen's 2188. This is notable because the Ryzen has a much higher boost clock of 4.40 GHz versus 3.00 GHz on the EPYC.

Q: Which chip wins in PassMark's single-thread test?

A: The Ryzen 5 PRO 5655GE dominates there, scoring 3240 compared to the EPYC's 2371. That is a 26.8% advantage for the Ryzen, which is the largest single-test delta in the entire head-to-head set.

Q: What about multi-threaded workloads?

A: The EPYC 8024P wins PassMark's multithread test with 20556 versus 18057, a 13.8% edge. In Cinebench R23 multicore, the EPYC scores 17472 against the Ryzen's 15504, with the same 12.7% delta seen across all Cinebench tests.

Q: Are there tests where the Ryzen 5 PRO 5655GE wins?

A: Yes, four distinct PassMark workloads favor the Ryzen: extended instructions (15714 vs 14251, a 9.3% win), floating point math (38204 vs 34757, a 9% win), integer math (67582 vs 62128, an 8.1% win), and single-thread (3240 vs 2371, a 26.8% win).

Q: Do these chips have integrated graphics?

A: The Ryzen 5 PRO 5655GE includes Radeon Vega 7 integrated graphics. The EPYC 8024P has no integrated graphics listed in the database, which is typical for a server-focused part.

Architecture Differences

The EPYC 8024P is built on the Zen 4c architecture, codenamed Siena, and belongs to the EPYC 8004 series. It uses a 5 nm process from TSMC, with 8,875 million transistors on a 73 mm² die. The Ryzen 5 PRO 5655GE uses the older Zen 3 architecture, codenamed Cezanne, on a 7 nm process, with 10,700 million transistors on a much larger 180 mm² die. The process node advantage for the EPYC is clear, though the Ryzen packs more transistors overall.

Core counts differ: the EPYC has 8 cores and 16 threads, while the Ryzen has 6 cores and 12 threads. That two-core lead for the EPYC is central to its multi-threaded wins. Cache layouts also diverge. The EPYC provides 1 MB of L2 per core and 32 MB of shared L3, while the Ryzen offers 512 KB of L2 per core and 16 MB of L3. Both share the same 64 KB L1 per core. The EPYC's larger L3 cache aligns with its server role, where data locality matters.

Memory support is a major split. The EPYC uses DDR5 with a six-channel memory bus and 230.4 GB/s of bandwidth. The Ryzen uses DDR4 with a dual-channel bus and 51.2 GB/s. That is a 4.5x bandwidth gap in favor of the EPYC, a decisive factor for memory-intensive server workloads. Both support ECC memory, but the platform context differs: the EPYC sits on AMD Socket SP6, the Ryzen on AMD Socket AM4.

PCIe connectivity also separates them. The EPYC offers Gen 5 with 96 lanes (CPU only), while the Ryzen provides Gen 3 with 16 lanes. The EPYC's lane count and generation are aimed at expansion-heavy environments. The Ryzen's 16 Gen 3 lanes suit a desktop platform. The Ryzen also has integrated Radeon Vega 7 graphics, a feature absent on the EPYC.

Clock speeds tell an interesting story. The Ryzen has a base clock of 3.40 GHz and a boost of 4.40 GHz, both well above the EPYC's 2.40 GHz base and 3.00 GHz boost. Yet the EPYC still wins most Cinebench tests by 12.7%. The Ryzen's higher clocks do not translate into wins there, which points to the EPYC's architectural efficiency per clock and additional cores. The Ryzen's 65 W TDP is lower than the EPYC's 90 W, but the EPYC delivers more throughput per watt in several workloads based on the score deltas.

The Verdict

The data supports a clear split by use case. For server or workstation deployments where memory bandwidth, PCIe lanes, and multi-threaded throughput matter, the EPYC 8024P is the obvious pick. It wins 12 of the 17 head-to-head benchmarks, including every Cinebench test and the PassMark multithread test. Its six-channel DDR5 memory and 96 Gen 5 lanes are platform features the Ryzen cannot match. The 122.4% lead in prime number finding and 194% lead in physics simulation confirm its strength in compute-heavy, parallel tasks.

The Ryzen 5 PRO 5655GE suits desktop workloads where single-thread responsiveness and math throughput take priority. It wins the PassMark single-thread test by 26.8%, and it leads in floating point math by 9%, integer math by 8.1%, and extended instructions by 9.3%. Its higher boost clock of 4.40 GHz and integrated Radeon Vega 7 graphics make it a self-contained desktop processor. For users who do not need the EPYC's server platform, the Ryzen offers competitive performance in math-heavy applications at a lower 65 W TDP.

The EPYC's 12.7% uniform advantage across all Cinebench versions is striking. It suggests the EPYC's Zen 4c architecture delivers consistent per-thread gains over Zen 3 in rendering workloads, despite the Ryzen's clock advantage. The Ryzen's PassMark single-thread win shows the opposite: in that specific workload, raw clock speed matters more. The verdict is not about which CPU is universally better, but which fits the workload. Server and workstation buyers should choose the EPYC 8024P. Desktop users prioritizing single-thread speed and math performance should choose the Ryzen 5 PRO 5655GE.

Specification Differences

| Specification | AMD EPYC 8024P | AMD Ryzen 5 PRO 5655GE |

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

| Cores | 8 | 6 |

| Threads | 16 | 12 |

| Base clock | 2.40 GHz | 3.40 GHz |

| Boost clock | 3.00 GHz | 4.40 GHz |

| TDP | 90 W | 65 W |

| Socket | AMD Socket SP6 | AMD Socket AM4 |

| Architecture | Zen 4c | Zen 3 |

| Codename | Siena | Cezanne |

| Process node | 5 nm | 7 nm |

| Transistors | 8,875 million | 10,700 million |

| Die size | 73 mm² | 180 mm² |

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

| L3 cache | 32 MB (shared) | 16 MB |

| Memory support | DDR5 | DDR4 |

| Memory bus | Six-channel | Dual-channel |

| Memory bandwidth | 230.4 GB/s | 51.2 GB/s |

| PCIe | Gen 5, 96 lanes (CPU only) | Gen 3, 16 lanes (CPU only) |

| Integrated graphics | None | Radeon Vega 7 |

| Market segment | Server/Workstation | Desktop |

| Launch MSRP | $409 | Not listed |

| Release date | 2023-09-17 | 2024-05-06 |

Head-to-Head Benchmarks

The EPYC 8024P wins all six Cinebench tests, and the margin is identical across the board: 12.7%. In Cinebench R15 multicore, the EPYC scores 1761 against the Ryzen's 1562. In R15 single-core, it is 248 versus 220. R20 multicore shows 7338 against 6511, and R20 single-core shows 1035 against 918. R23 multicore has the EPYC at 17472 versus 15504, and R23 single-core has it at 2466 versus 2188. This uniform delta suggests a consistent architectural efficiency gain for Zen 4c in Cinebench workloads.

PassMark results are more varied. The EPYC wins data compression with 232242 versus 228037, a slim 1.8% edge. It wins data encryption with 15809 versus 14509, a 9% margin. Its biggest wins come in prime number finding, 109 versus 49, a 122.4% blowout, and physics, 1905 versus 648, a 194% landslide. Random string sorting also favors the EPYC heavily, 34613 versus 23769, a 45.6% gap. The multithread test goes to the EPYC at 20556 versus 18057, a 13.8% win.

The Ryzen 5 PRO 5655GE takes four PassMark workloads plus the duplicated single-thread entries. Extended instructions show 15714 versus 14251, a 9.3% win for the Ryzen. Floating point math goes 38204 versus 34757, a 9% lead. Integer math goes 67582 versus 62128, an 8.1% edge. The single-thread test is the Ryzen's largest victory: 3240 versus 2371, a 26.8% margin. That result appears twice in the data, once as passmark_single_thread and once as passmark_singlethread, with identical scores.

The overall win count favors the EPYC at 12 wins against 4 for the Ryzen, with one duplicated single-thread entry. The EPYC's wins tend to be larger in magnitude for compute-heavy tasks, while the Ryzen's wins are concentrated in math and single-thread workloads. The EPYC's 194% physics margin and 122.4% prime number margin are the standout deltas, dwarfing the Ryzen's 26.8% single-thread win. Average benchmark scores reflect this: the EPYC sits at 26555, the Ryzen at 25880, a 2.6% gap.

Where Each One Wins

The EPYC 8024P wins in every Cinebench generation tested, from R15 through R23, covering both single-core and multicore. That makes it the stronger choice for rendering, 3D modeling, and any workload that relies on Cinema 4D's engine or similar rendering pipelines. Its PassMark wins in data compression, data encryption, prime number finding, physics, and random string sorting point to strengths in server-side tasks: compression algorithms, cryptographic workloads, scientific simulation, and data sorting. The multithread test win at 13.8% reinforces its suitability for parallel throughput. The 194% physics margin is the largest in the entire comparison, suggesting exceptional performance in physics simulation workloads, likely tied to its Zen 4c architecture and larger L3 cache.

The Ryzen 5 PRO 5655GE wins in PassMark's extended instructions, floating point math, integer math, and single-thread tests. These wins indicate strengths in mathematical computation, including floating point-heavy scientific calculations and integer-heavy logic. The 26.8% single-thread lead is its most significant victory, driven by its 4.40 GHz boost clock. For desktop users running spreadsheets, financial modeling, or single-threaded legacy applications, the Ryzen's high clock speed delivers tangible benefits. Its integrated Radeon Vega 7 graphics also make it a complete desktop package without a discrete GPU, though the database does not include graphics benchmarks.

The split is clean: the EPYC for server and workstation deployments where multi-threaded throughput, memory bandwidth, and expansion capability matter; the Ryzen for desktop use where single-thread speed and math performance take precedence. The EPYC's six-channel DDR5 memory at 230.4 GB/s versus the Ryzen's dual-channel DDR4 at 51.2 GB/s is a decisive platform advantage for memory-bound workloads. The EPYC's 96 Gen 5 PCIe lanes versus the Ryzen's 16 Gen 3 lanes further cements its server positioning. Buyers should weigh these platform differences alongside the benchmark results, as the EPYC's wins are broader and larger in magnitude, while the Ryzen's wins are narrower but significant in specific desktop tasks.