AMD EPYC 7451 vs Intel Core i9-10920X Comparison

Head-to-Head Benchmarks

The benchmark data presents a remarkably consistent picture: the Intel Core i9-10920X wins every single head-to-head comparison in the database, across all six Cinebench tests. The margin is strikingly uniform, hovering between 6.4% and 6.8% in every workload.

In Cinebench R15 multi-core, the Intel part scores 2203 against the EPYC 7451's 2061, a 6.4% deficit for the AMD chip. The single-core R15 test tells the same story: 311 versus 290, a 6.8% gap, the largest margin recorded in the entire comparison. Moving to Cinebench R20, the Intel processor posts 9181 in multi-core versus 8589, again a 6.4% difference, while single-core shows 1296 against 1212, a 6.5% gap.

The pattern holds in Cinebench R23, the most demanding of the three render workloads. The Intel Core i9-10920X reaches 21861 in multi-core, outpacing the EPYC 7451's 20450 by 6.5%. In single-core R23, the scores are 3086 and 2887 respectively, a 6.4% margin. The uniformity of these deltas is noteworthy: despite the two processors having very different core counts, clock speeds, and architectural lineages, the relative performance advantage stays tightly clustered between 6.4% and 6.8% across every test.

What makes this fascinating is the core count disparity. The AMD EPYC 7451 fields 24 cores and 48 threads, exactly double the Intel's 12 cores and 24 threads. Yet the Intel part still leads in multi-threaded workloads. The data implies that the EPYC's extra cores cannot compensate for the Intel's clock speed and per-core efficiency advantages in these Cinebench workloads. The EPYC 7451 boosts to 3.20 GHz, while the Core i9-10920X reaches 4.80 GHz, a 50% higher ceiling that clearly matters more than raw core quantity in this benchmark suite.

The average benchmark score reinforces the hierarchy. The Intel Core i9-10920X records an average score of 6347 across all tests, placing it in the 62nd percentile of all CPUs in the database. The AMD EPYC 7451 averages 5915, sitting at the 61st percentile. The Intel part also benefits from a broader benchmark footprint, including Geekbench results that the AMD processor lacks entirely. In Geekbench multi-core, the Intel scores 11165, and in single-core it reaches 1673, further extending its lead in the aggregate statistics.

Looking at the nearest rivals in the database provides context for where each processor sits. The EPYC 7451's closest competitor is the Intel Core i9-9920X, which scores 5917 on average, a delta of 0%. The AMD Xeon W-2170B sits 1.2% higher, while the AMD EPYC 7401P trails by 1.7% and the Intel Xeon E-2388G by 1.9%. For the Core i9-10920X, the AMD EPYC 7551 leads it by only 0.7%, while the AMD Ryzen 3 3100U trails by 1.6%, the AMD Ryzen Threadripper 1950X by 1.9%, and the Intel Core i7-12800HE leads by 1.9%. These proximity scores suggest both processors are competitive within their respective performance tiers, even if the Intel part holds the edge in this direct confrontation.

FAQ

Q: Which processor wins in multi-core performance?

A: The Intel Core i9-10920X wins every multi-core benchmark in the comparison. In Cinebench R20 multi-core, it scores 9181 versus the AMD EPYC 7451's 8589, a 6.4% advantage. The R23 multi-core test shows a similar 6.5% lead, with scores of 21861 and 20450 respectively.

Q: How do the single-core results compare?

A: The Intel Core i9-10920X leads in every single-core test as well. The largest margin is in Cinebench R15 single-core, where Intel scores 311 against AMD's 290, a 6.8% gap. The R20 and R23 single-core tests show 6.5% and 6.4% advantages respectively.

Q: Does the EPYC 7451's higher core count make a difference?

A: Despite having 24 cores and 48 threads, double the Intel's 12 cores and 24 threads, the EPYC 7451 still loses all multi-threaded benchmarks. The Intel part's higher boost clock of 4.80 GHz, compared to the EPYC's 3.20 GHz, appears to overcome the core count deficit in these workloads.

Q: What is the overall performance percentile for each processor?

A: The Intel Core i9-10920X sits in the 62nd percentile of all CPUs, while the AMD EPYC 7451 sits in the 61st percentile. The Intel processor also has a higher average benchmark score of 6347, compared to 5915 for the AMD.

Q: Are there any benchmarks where the EPYC 7451 comes out ahead?

A: No. The head-to-head data shows zero wins for the AMD EPYC 7451 and six wins for the Intel Core i9-10920X across all Cinebench tests. The closest the AMD chip comes is in single-core R23, where it trails by 6.4%.

Q: Which processors are the closest rivals to these two?

A: The EPYC 7451's nearest rival is the Intel Core i9-9920X with an average score of 5917, a 0% delta. The Core i9-10920X's nearest rival is the AMD EPYC 7551 at 6391, which leads by 0.7%. The AMD Ryzen 3 3100U trails the Intel part by 1.6%.

The Verdict

The data speaks clearly: the Intel Core i9-10920X is the faster processor in this comparison. It wins all six head-to-head benchmarks, holds a higher average score of 6347 versus 5915, and ranks one percentile higher at 62 versus 61. The consistency of the 6.4% to 6.8% margins across every test suggests this is a systematic performance advantage, not a workload-specific quirk.

For users prioritizing raw Cinebench rendering performance, the Intel Core i9-10920X is the straightforward choice. Its single-core lead is particularly relevant for lightly threaded applications, where the 4.80 GHz boost clock provides a tangible advantage over the EPYC's 3.20 GHz ceiling. The multi-core wins, achieved despite having half the core count, demonstrate that clock speed and architectural efficiency can outweigh sheer core quantity in these workloads.

The AMD EPYC 7451, however, should not be dismissed. Its 24 cores and 48 threads target a different market segment entirely, that of servers and workstations, where the eight-channel memory bus and ECC support matter more than Cinebench scores. The benchmark data does not capture memory bandwidth scalability, multi-socket environments, or sustained server workloads. The EPYC's 170.6 GB/s memory bandwidth, a figure not available for the Intel part, hints at capabilities these benchmarks do not measure. For its intended server and workstation role, the EPYC 7451 remains a credible option despite losing this head-to-head.

Specification Differences

The two processors diverge substantially in nearly every specification category. The AMD EPYC 7451 packs 24 cores and 48 threads, while the Intel Core i9-10920X offers 12 cores and 24 threads, exactly half. Base clocks show the Intel part at 3.50 GHz versus the EPYC's 2.30 GHz, and boost clocks widen the gap further: 4.80 GHz for Intel against 3.20 GHz for AMD. Thermal design power differs modestly, with the EPYC rated at 180 watts and the Intel at 165 watts.

Memory architecture highlights the EPYC's server pedigree. The AMD chip supports eight-channel DDR4 memory with 170.6 GB/s of bandwidth, while the Intel part uses quad-channel DDR4 with no bandwidth figure recorded. ECC memory is supported on the EPYC but not on the Intel. The EPYC 7451 uses AMD Socket SP3, whereas the Intel Core i9-10920X uses Intel Socket 2066. Both processors support PCIe Gen 3 and feature unlocked multipliers, but their market segments differ: the EPYC targets Server/Workstation, while the Intel part is aimed at Desktop.

The EPYC 7451 carries a 64 MB shared L3 cache, dwarfing the Intel's 19.25 MB. Per-core L2 cache tells the opposite story, with Intel providing 1 MB per core versus AMD's 512 KB per core. L1 caches are 96 KB per core on the EPYC and 64 KB per core on the Intel. The AMD processor includes 4,800 million transistors on a 213 mm² die, while no transistor count or die size is recorded for the Intel part. Both use 14 nm process nodes, though from different foundries: GlobalFoundries for AMD and Intel for itself. The EPYC 7451 was released on June 28, 2017, over two years before the Intel's October 18, 2019 launch date.

Architecture Differences

The architectural divide here is generational and philosophical. The AMD EPYC 7451 belongs to the EPYC 7001 series, built on the Zen architecture with the Naples codename. The Intel Core i9-10920X comes from the Core 10th Gen family, using Cascade Lake architecture with the Cascade Lake-X codename. Both are manufactured on 14 nm process nodes, but the design philosophies diverge sharply.

Zen, as implemented in Naples, prioritizes core density and memory throughput. The 24-core, 48-thread configuration, combined with the eight-channel memory controller and 64 MB shared L3 cache, targets data center workloads where parallel throughput and memory bandwidth are paramount. The 4,800 million transistors on a 213 mm² die reflect a design focused on packing as many compute units as possible.

Cascade Lake-X, by contrast, emphasizes per-core performance and clock headroom. The 12-core, 24-thread configuration with a 4.80 GHz boost clock and 19.25 MB shared L3 cache is engineered for desktop enthusiasts who need strong single-threaded performance alongside respectable multi-threaded capability. The Intel architecture's higher clock ceilings compensate for its lower core count and smaller shared cache.

The memory controllers reflect these different priorities. The EPYC's eight-channel DDR4 support with 170.6 GB/s bandwidth is a server-class feature, enabling high-throughput access to large memory pools. The Intel's quad-channel support without a recorded bandwidth figure serves desktop workloads adequately but does not match the EPYC's memory scalability. ECC support on the AMD chip further cements its enterprise positioning, while its absence on the Intel part aligns with consumer desktop expectations.

Where Each One Wins

The Intel Core i9-10920X wins across every benchmark in this comparison, making its case straightforward. In all six Cinebench tests, spanning R15, R20, and R23 in both single-core and multi-core configurations, the Intel part holds a 6.4% to 6.8% advantage. This consistency suggests it is the better choice for rendering workloads, content creation, and any application that scales with Cinebench-style performance. Its 4.80 GHz boost clock gives it a decisive edge in single-threaded tasks, where it leads by margins between 6.4% and 6.8% across the three R-series versions.

The AMD EPYC 7451's strengths lie outside the measured benchmarks. Its 24 cores and 48 threads provide double the parallel compute resources of the Intel part, a characteristic that could benefit workloads beyond Cinebench's scope. The eight-channel memory bus with 170.6 GB/s bandwidth and ECC support position it for server environments where data integrity and memory throughput are critical. Its larger 64 MB L3 cache, over three times the Intel's 19.25 MB, may aid workloads with large working sets that benefit from extensive on-chip caching.

The market segments tell the story: the EPYC 7451 is a Server/Workstation part, while the Core i9-10920X is a Desktop chip. For users building a workstation where Cinebench-style rendering is the primary task, the Intel processor's benchmark dominance makes it the logical pick. For data center deployments requiring ECC memory, massive memory bandwidth, and high core counts for virtualization or database workloads, the EPYC 7451's specification sheet offers capabilities the benchmarks do not capture. The database records zero wins for the EPYC and six for the Intel, but the EPYC's server-oriented feature set suggests the right choice depends heavily on the intended deployment environment.