AMD EPYC 7351P vs Intel Core i9-10900X Comparison

Head-to-Head Benchmarks

The benchmark data presents a stark split between the two processors. The AMD EPYC 7351P wins six of the eight head-to-head tests, all in the Cinebench suite, while the Intel Core i9-10900X takes the two Geekbench tests. The margins are consistent within each suite, which points to a fundamental difference in how these chips handle the two benchmark workloads.

In Cinebench R15 multicore, the EPYC 7351P scores 2231 against the i9-10900X's 1910, a 14.4% advantage. The same delta appears in Cinebench R15 singlecore: 314 versus 269, again a 14.4% gap. Cinebench R20 repeats the pattern exactly—9296 versus 7961 in multicore, and 1312 versus 1123 in singlecore, both with the same 14.4% delta. Cinebench R23 follows suit: 22135 versus 18957 in multicore, and 3125 versus 2676 in singlecore. Every single Cinebench test, regardless of version or core count, shows the EPYC 7351P ahead by precisely 14.4%.

The Geekbench results flip the script dramatically. The i9-10900X scores 10104 in multicore against the EPYC's 4607, a massive 119.3% lead. In singlecore, the i9-10900X posts 1438 versus 733, a 96.2% advantage. These are not small differences—the Intel chip more than doubles the AMD processor's Geekbench multicore score. The contrast between the two benchmark suites is striking: in one, the EPYC wins by a consistent 14.4%, and in the other, the i9-10900X wins by nearly 120%.

The average benchmark scores tell a similar story of overall parity. The i9-10900X averages 5555 per benchmark, while the EPYC 7351P averages 5469. Both chips sit at the 61st percentile among all CPUs. The nearest rivals list shows the i9-10900X at 0.1% below the Intel Celeron G6900's average score of 5561, while the EPYC sits 0.5% above the Intel Xeon W-1290P's 5443. Interestingly, each processor appears in the other's nearest rivals list—the EPYC is 1.5% below the i9-10900X's average, and the i9-10900X is 1.5% above the EPYC's average from the other perspective.

What explains this bifurcation? The Cinebench scores suggest the EPYC's higher core count and larger cache pay off in rendering workloads that scale with cores. The Geekbench results indicate the i9-10900X's higher clock speeds dominate in that benchmark's particular mix of tests. The data does not show one chip as universally better—it shows two chips optimized for different kinds of work.

FAQ

Q: Which processor wins more benchmark tests?

A: The AMD EPYC 7351P wins six of the eight head-to-head tests, all in the Cinebench suite. The Intel Core i9-10900X wins the remaining two, both in Geekbench.

Q: What is the margin in the Cinebench tests?

A: The EPYC 7351P leads by exactly 14.4% in every Cinebench test, from R15 multicore through R23 singlecore. This consistency across all six tests is notable.

Q: How large is the Intel chip's Geekbench advantage?

A: The i9-10900X leads by 119.3% in Geekbench multicore (10104 versus 4607) and by 96.2% in Geekbench singlecore (1438 versus 733). The Intel chip more than doubles the EPYC's multicore Geekbench score.

Q: Do the two processors have similar overall performance?

A: Yes, their average benchmark scores are close—5555 for the i9-10900X and 5469 for the EPYC 7351P. Both rank at the 61st percentile among all CPUs.

Q: Which chip appears in the other's nearest rivals list?

A: Both do. The EPYC 7351P lists the Intel Core i9-10900X as a nearest rival at a 1.5% delta, and the i9-10900X lists the EPYC as a nearest rival at a 0.6% delta from the opposite perspective.

Q: Is the EPYC 7351P a newer processor than the i9-10900X?

A: No. The EPYC 7351P was released on 2017-06-28, while the i9-10900X was released on 2019-10-18. The Intel chip is the newer part by over two years.

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Core i9-10900X uses the Cascade Lake architecture, specifically the Cascade Lake-X variant for the X-Series 10th Gen desktop lineup. The AMD EPYC 7351P uses the original Zen architecture, codenamed Naples, from the EPYC 7001 series. Both are built on a 14 nm process node, but Intel uses its own foundry while AMD uses GlobalFoundries.

The core configurations diverge significantly. The i9-10900X packs 10 cores and 20 threads. The EPYC 7351P offers 16 cores and 32 threads—60% more cores and threads on paper. This explains the EPYC's consistent lead in Cinebench multicore tests, which scale well with additional cores. However, the i9-10900X compensates with much higher clock speeds: a base clock of 3.70 GHz and boost clock of 4.70 GHz, versus the EPYC's 2.40 GHz base and 2.90 GHz boost.

Cache hierarchies also differ. The i9-10900X has 64 KB of L1 cache per core, 1 MB of L2 cache per core, and 19.25 MB of shared L3 cache. The EPYC 7351P has 96 KB of L1 per core, 512 KB of L2 per core, and 64 MB of shared L3 cache—over three times the L3 capacity. The EPYC also reports 4,800 million transistors and a die size of 213 mm², while the i9-10900X's transistor count and die size are not listed in the data.

Memory architecture marks another major split. Both support DDR4 memory, but the i9-10900X uses a quad-channel memory bus while the EPYC 7351P uses an eight-channel bus. The EPYC's memory bandwidth is listed at 170.6 GB/s, while the i9-10900X's memory bandwidth is not specified. The EPYC supports ECC memory; the i9-10900X does not. This places the EPYC firmly in server territory, while the i9-10900X targets desktop enthusiasts.

The sockets are incompatible: Intel Socket 2066 for the i9-10900X, AMD Socket SP3 for the EPYC. Both support PCIe Gen 3, and neither has integrated graphics. Both have unlocked multipliers. The EPYC carries the part number PS735PBEVGPAF and is listed as a Server/Workstation part, while the i9-10900X is classified as a Desktop processor.

Specification Differences

The table below highlights where the two processors differ in their specifications.

| Specification | Intel Core i9-10900X | AMD EPYC 7351P |

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

| Cores | 10 | 16 |

| Threads | 20 | 32 |

| Base clock | 3.70 GHz | 2.40 GHz |

| Boost clock | 4.70 GHz | 2.90 GHz |

| TDP | 165 W | 170 W |

| Socket | Intel Socket 2066 | AMD Socket SP3 |

| Architecture | Cascade Lake | Zen |

| Codename | Cascade Lake-X | Naples |

| Generation | Core i9 (X-Series 10th Gen) | EPYC (Zen (Naples)) |

| Foundry | Intel | GlobalFoundries |

| Transistors | Not listed | 4,800 million |

| Die size | Not listed | 213 mm² |

| L1 cache | 64 KB (per core) | 96 KB (per core) |

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

| L3 cache | 19.25 MB (shared) | 64 MB (shared) |

| Memory bus | Quad-channel | Eight-channel |

| Memory bandwidth | Not listed | 170.6 GB/s |

| ECC memory | No | Yes |

| Market segment | Desktop | Server/Workstation |

| Release date | 2019-10-18 | 2017-06-28 |

The Verdict

The data does not crown a single winner. The AMD EPYC 7351P dominates Cinebench across every version, winning all six tests by a consistent 14.4% margin. The Intel Core i9-10900X dominates Geekbench, winning multicore by 119.3% and singlecore by 96.2%. Their average benchmark scores are nearly identical—5555 versus 5469—and both sit at the same 61st percentile among all CPUs.

The EPYC 7351P is the choice for workloads that resemble Cinebench: heavily multithreaded rendering, scientific computing, or server tasks that scale across its 16 cores and 32 threads. Its 64 MB of shared L3 cache and eight-channel memory with ECC support reinforce its server pedigree. The i9-10900X is the choice for workloads that resemble Geekbench: applications that favor high clock speeds, lower core counts, and the responsiveness of a desktop platform. Its 4.70 GHz boost clock and quad-channel memory bus suit desktop use cases where single-threaded performance matters.

The release dates add context. The EPYC 7351P launched on 2017-06-28, over two years before the i9-10900X's 2019-10-18 release. Despite being the older chip, the EPYC still wins the Cinebench suite decisively, which speaks to its core-count advantage. The i9-10900X's newer architecture and higher clocks give it the Geekbench edge, but the data shows the EPYC's 16-core design remains competitive in threaded workloads even against a newer desktop part.

Neither processor is a clear overall winner. The 61st percentile ranking for both, and the fact that each appears in the other's nearest rivals list, indicates they occupy a similar tier of overall performance. The choice comes down to workload, not superiority.

Where Each One Wins

AMD EPYC 7351P wins in:

• Multi-threaded rendering workloads, as shown by its 14.4% lead in Cinebench R15, R20, and R23 multicore tests
• Single-threaded Cinebench performance, where it also leads by 14.4% across all three Cinebench versions
• Server and workstation deployments, given its 16 cores, 32 threads, ECC memory support, and eight-channel memory bus
• Memory bandwidth-intensive tasks, with its 170.6 GB/s of bandwidth versus the i9-10900X's unspecified but quad-channel configuration
• Cache-heavy workloads, thanks to 64 MB of shared L3 cache compared to the Intel chip's 19.25 MB

Intel Core i9-10900X wins in:

• Geekbench multicore workloads, with a 119.3% advantage over the EPYC (10104 versus 4607)
• Geekbench singlecore workloads, with a 96.2% lead (1438 versus 733)
• Desktop use cases, given its classification as a Desktop processor with an Intel Socket 2066
• Clock-sensitive applications, with a 4.70 GHz boost clock versus the EPYC's 2.90 GHz
• Scenarios where power draw matters at the margins, with its 165 W TDP versus the EPYC's 170 W

The practical takeaway from the data is straightforward. If your work resembles Cinebench—rendering, simulation, or other heavily threaded compute—the EPYC 7351P's 14.4% consistent lead makes it the stronger option. If your work resembles Geekbench—general desktop productivity, software development, or mixed single-threaded tasks—the i9-10900X's near-doubling of Geekbench scores gives it a decisive advantage. Neither chip is a bad choice; they simply aim at different targets.