AMD EPYC 72F3 vs Intel Core i9-7980XE Comparison

Where Each One Wins

The benchmark data presents a clear and consistent picture across the entire Cinebench test suite. The Intel Core i9-7980XE wins all six recorded head-to-head comparisons, with no test going to the AMD EPYC 72F3. This is not a narrow victory or a split decision; the Intel part dominates every measured workload category.

The i9-7980XE holds its strongest advantage in the multi-core tests, where its 18 cores and 36 threads can be fully utilized. In Cinebench R15 multi-core, it scores 2527 against the EPYC's 2334, a margin of 8.3%. That same 8.2% delta repeats in Cinebench R20 multi-core, where the Intel part scores 10530 versus 9728, and again in R23 multi-core with 25072 versus 23164. Across every multi-core generation of Cinebench, the Intel processor maintains a nearly identical percentage lead, suggesting a structural advantage rather than a workload-specific quirk.

The single-core results follow the same pattern, though the absolute scores are naturally much lower. The i9-7980XE posts 356 in R15 single-core versus 329 for the EPYC, 1486 versus 1373 in R20, and 3539 versus 3270 in R23. Each of these represents an 8.2% advantage for Intel. This consistency across both single-threaded and multi-threaded tests indicates that the i9-7980XE is simply faster per unit of work in this benchmark family, regardless of how many threads are active.

However, the EPYC 72F3 is not without its own areas of strength, even if those strengths do not appear in the Cinebench results. Its architecture is a Zen 3 Milan design fabricated on a 7 nm process by TSMC, with 33,200 million transistors spread across eight chiplets, each 81 mm². The i9-7980XE, by contrast, is a monolithic Skylake-X die on Intel's 14 nm process, measuring 484 mm². The AMD part also offers a massive 256 MB of shared L3 cache, compared to 24.75 MB for the Intel part, and supports eight-channel DDR4 memory with 204.8 GB/s of bandwidth, more than double the Intel's quad-channel 85.3 GB/s. These differences point toward server-oriented strengths such as memory throughput and cache capacity, which are not captured by the Cinebench suite but are relevant to database, virtualization, and large-scale data workloads.

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Core i9-7980XE is a desktop extreme-edition part built on the Skylake architecture, specifically Skylake-X, and belongs to the Core i9 Extreme X-Series 7th Generation. It is fabricated on a 14 nm process at Intel's own foundry, with a die size of 484 mm². The chip integrates 18 cores and 36 threads, with a base clock of 2.60 GHz and a boost clock of 4.40 GHz. Each core carries 64 KB of L1 cache and 1 MB of L2 cache, while the L3 cache is 24.75 MB shared across all cores. The memory controller supports quad-channel DDR4, delivering 85.3 GB/s of bandwidth, and ECC memory is not supported. PCIe connectivity is Gen 3 with 44 lanes from the CPU. The multiplier is unlocked, allowing overclocking, and the part uses the Intel Socket 2066. It was released in August 2017 and is now end-of-life.

The AMD EPYC 72F3 is a server/workstation processor based on Zen 3, codenamed Milan. It is built on a 7 nm process at TSMC, with a chiplet design consisting of eight dies, each 81 mm², totaling 33,200 million transistors. The CPU has 8 cores and 16 threads, with a base clock of 3.70 GHz and a boost clock of 4.10 GHz. L1 cache is 64 KB per core, L2 is 512 KB per core, and L3 is a massive 256 MB shared pool. Memory support is eight-channel DDR4, providing 204.8 GB/s of bandwidth, and ECC memory is supported, which is expected for a server part. PCIe connectivity is Gen 4 with 128 lanes from the CPU. The multiplier is locked, and the socket is AMD Socket SP3. It launched in March 2021 and remains in active production.

The most striking contrasts are process node, core count, and cache. The EPYC uses a newer, denser 7 nm process with more transistors overall, but the i9-7980XE has more than twice the cores and threads. The EPYC compensates with a much larger L3 cache and far greater memory bandwidth, alongside double the PCIe lanes and a newer PCIe generation. These are architectural choices aimed at different deployment scenarios.

Head-to-Head Benchmarks

The head-to-head data is unusually uniform. Every single test in the comparison shows the Intel Core i9-7980XE ahead by approximately 8.2%. The only exception is Cinebench R15 multi-core, where the margin is 8.3%. This uniformity suggests that the performance gap is driven by a consistent underlying factor, such as clock speed or per-core efficiency, rather than by workload scaling.

In Cinebench R15 multi-core, the i9-7980XE scores 2527 versus 2334 for the EPYC 72F3, a delta of 8.3%. This is the largest percentage margin in the entire comparison. In R15 single-core, the Intel part scores 356 versus 329, a delta of 8.2%. Moving to R20, the multi-core scores are 10530 versus 9728, and single-core is 1486 versus 1373, both with an 8.2% delta. The pattern holds in R23 as well: 25072 versus 23164 in multi-core, and 3539 versus 3270 in single-core, again 8.2%.

Because the delta is nearly identical across all tests, it is reasonable to conclude that the i9-7980XE's advantage is not tied to core count scaling. The EPYC's 8-core design is not being penalized disproportionately in multi-core tests, nor is the Intel part's 18-core design gaining extra leverage. Instead, the Intel part appears to deliver a consistent per-thread performance edge, which then carries through to multi-threaded results.

Looking at the broader database context, the i9-7980XE has an average benchmark score of 7018 and sits at the 63rd percentile of all CPUs. Its nearest rivals include the Intel Xeon Platinum 8260 at 6992, which is only 0.4% behind, and the Intel Core i9-9960X at 6938, which is 1.2% behind. The AMD EPYC 7302P sits at 7100, which is 1.1% ahead of the i9-7980XE. The EPYC 72F3, meanwhile, has an average score of 6700 and sits at the 62nd percentile. Its nearest rival is the Intel Xeon Gold 5317 at 6710, just 0.1% ahead, and the Intel Xeon D-2775TE at 6711, 0.2% ahead. The Intel Core i9-9940X is 0.6% behind at 6657, and the AMD Ryzen Threadripper 2950X is 0.8% behind at 6647.

These percentile positions are worth noting. The i9-7980XE and the EPYC 72F3 are only one percentile point apart overall, despite the Intel part winning every head-to-head Cinebench test. This suggests that across the broader benchmark database, other workloads narrow or even reverse the gap, likely reflecting the EPYC's advantages in memory bandwidth, cache capacity, and server-oriented tasks.

FAQ

Q: Which processor wins in Cinebench multi-core tests?

A: The Intel Core i9-7980XE wins all three multi-core tests. It scores 2527 versus 2334 in R15, 10530 versus 9728 in R20, and 25072 versus 23164 in R23, each with a delta of approximately 8.2%.

Q: Is the AMD EPYC 72F3 competitive in single-core performance?

A: The EPYC 72F3 trails the i9-7980XE in every single-core test. The Intel part scores 356 versus 329 in R15, 1486 versus 1373 in R20, and 3539 versus 3270 in R23, all with an 8.2% margin.

Q: What are the core and thread counts for each processor?

A: The Intel Core i9-7980XE has 18 cores and 36 threads. The AMD EPYC 72F3 has 8 cores and 16 threads.

Q: Which processor has more L3 cache?

A: The AMD EPYC 72F3 has 256 MB of shared L3 cache. The Intel Core i9-7980XE has 24.75 MB of shared L3 cache.

Q: What memory channels does each processor support?

A: The Intel Core i9-7980XE supports quad-channel DDR4 memory with 85.3 GB/s of bandwidth. The AMD EPYC 72F3 supports eight-channel DDR4 memory with 204.8 GB/s of bandwidth.

Q: Does the EPYC 72F3 support ECC memory?

A: Yes, ECC memory is supported on the AMD EPYC 72F3. The Intel Core i9-7980XE does not support ECC memory.

The Verdict

The recorded data points to the Intel Core i9-7980XE as the faster processor in the Cinebench rendering workloads measured here. It wins all six head-to-head tests with a consistent margin of roughly 8.2%, and it holds a higher average benchmark score of 7018 against the EPYC's 6700. For users whose primary workload is rendering, content creation, or any application that behaves like the Cinebench suite, the i9-7980XE is the clear choice based on these numbers.

However, the verdict is not simply a matter of raw Cinebench scores. The EPYC 72F3 is a server-class processor with eight-channel memory support, 204.8 GB/s of memory bandwidth, ECC memory, 128 PCIe Gen 4 lanes, and 256 MB of L3 cache. These features are not reflected in the head-to-head benchmark results, but they are critical for server deployments, large virtualized environments, in-memory databases, and workloads that are sensitive to memory bandwidth and cache capacity. The EPYC also comes from a newer architecture generation on a 7 nm process, and it remains in active production, whereas the Intel part is end-of-life.

The i9-7980XE, despite its age and process disadvantage, demonstrates that a high core count and a high boost clock can still deliver competitive rendering performance. Its unlocked multiplier adds flexibility for enthusiasts, and its desktop Socket 2066 platform is designed for single-socket high-performance systems.

The choice depends on the deployment context. For a workstation focused on rendering and multithreaded desktop applications, the i9-7980XE has the benchmark advantage. For a server environment where ECC memory, massive memory bandwidth, PCIe Gen 4 connectivity, and large cache pools are essential, the EPYC 72F3 brings capabilities that the Intel part simply does not offer. The data does not support calling one universally superior; it supports calling each better suited to different roles.

Specification Differences

| Specification | Intel Core i9-7980XE | AMD EPYC 72F3 |

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

| Cores | 18 | 8 |

| Threads | 36 | 16 |

| Base Clock | 2.60 GHz | 3.70 GHz |

| Boost Clock | 4.40 GHz | 4.10 GHz |

| TDP | 165 W | 180 W |

| Socket | Intel Socket 2066 | AMD Socket SP3 |

| Architecture | Skylake | Zen 3 |

| Codename | Skylake-X | Milan |

| Generation | Core i9 Extreme (X-Series 7th Gen) | EPYC (Zen 3 (Milan)) |

| Process Node | 14 nm | 7 nm |

| Foundry | Intel | TSMC |

| Die Size | 484 mm² | 8x 81 mm² |

| Transistors | Not specified | 33,200 million |

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

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

| L3 Cache | 24.75 MB (shared) | 256 MB (shared) |

| Memory Support | DDR4 | DDR4 |

| Memory Bus | Quad-channel | Eight-channel |

| Memory Bandwidth | 85.3 GB/s | 204.8 GB/s |

| ECC Memory | No | Yes |

| PCIe | Gen 3, 44 Lanes (CPU only) | Gen 4, 128 Lanes (CPU only) |

| Market Segment | Desktop | Server/Workstation |

| Production Status | End-of-life | Active |

| Release Date | 2017-08-31 | 2021-03-14 |

| Launch MSRP | $1999 | $2468 |

| Multiplier Unlocked | Yes | No |

| Part Number | SR3RS | 100-000000327100-100000327WOF |