AMD EPYC 7351 vs AMD Ryzen Threadripper 1920 Comparison

Where Each One Wins

The benchmark data presents an unusual picture: the AMD EPYC 7351 wins all six recorded head-to-head comparisons against the AMD Ryzen Threadripper 1920. There is no test in the database where the Threadripper 1920 comes out ahead. That does not mean the two processors serve identical purposes, however. The EPYC 7351 is classified as a Server/Workstation part, while the Threadripper 1920 is a Desktop product. Their different market segments imply different workloads, even when the benchmark results point in one direction.

The EPYC 7351 takes the multi-core crown in every Cinebench generation recorded. In Cinebench R15 multi-core, it scores 1989 against 1890, a 5.2% lead. The margin grows slightly in R20 multi-core, where the EPYC posts 8291 versus 7877, again a 5.3% difference. The R23 multi-core test shows 19742 against 18756, with the same 5.3% gap. These consistent margins suggest the EPYC's additional four cores and eight threads give it a structural advantage in heavily threaded rendering workloads.

Single-core results also favor the EPYC, though the margins are similar. The EPYC wins R15 single-core with 280 versus 266, a 5.3% edge. In R20 single-core it scores 1170 against 1111, again 5.3% higher. The R23 single-core test shows 2787 versus 2647, maintaining the same 5.3% delta. The fact that the single-core lead matches the multi-core lead almost exactly indicates the EPYC's per-thread performance is not compromised by its server-oriented design.

The Threadripper 1920 counters with a lower thermal envelope and a desktop-oriented feature set. Its 140 W TDP versus the EPYC's 170 W means it draws less power under load, though the database does not record wattage figures beyond the TDP specifications. It also uses the SP3r2 socket, which is distinct from the EPYC's SP3, meaning the two are not directly interchangeable in a given motherboard. The Threadripper's quad-channel memory bus, while narrower than the EPYC's eight-channel configuration, aligns with typical high-end desktop builds where memory capacity matters less than low latency.

Architecture Differences

Both processors share the Zen architecture and the 14 nm process node, both manufactured by GlobalFoundries. The silicon itself is similar in origin, but the packaging and configuration diverge significantly. The EPYC 7351 uses a single 213 mm² die with 4,800 million transistors, while the Threadripper 1920 uses two 213 mm² dies, totaling 9,600 million transistors. This dual-die approach in the Threadripper explains its higher transistor count despite fewer active cores.

The core count difference is central. The EPYC 7351 has 16 cores and 32 threads, while the Threadripper 1920 has 12 cores and 24 threads. Both use Zen's simultaneous multithreading, so the thread count is exactly twice the core count for each. The L1 and L2 caches are identical per core: 96 KB L1 and 512 KB L2. The L3 cache differs substantially: the EPYC has 64 MB shared, while the Threadripper has 32 MB. This 32 MB gap in last-level cache could affect workloads that repeatedly access a large working set.

Memory architecture presents another major divergence. The EPYC 7351 supports eight-channel DDR4 with a theoretical bandwidth of 170.6 GB/s. The Threadripper 1920 supports quad-channel DDR4 with 85.3 GB/s. That is exactly half the bandwidth, which matters for memory-bound server workloads such as database transactions or virtual machine consolidation. The EPYC also supports ECC memory, while the Threadripper does not. For error-sensitive server environments, ECC is often a non-negotiable requirement.

Clock speeds favor the Threadripper on paper. It has a 3.20 GHz base clock and a 3.80 GHz boost clock, compared to the EPYC's 2.40 GHz base and 2.90 GHz boost. Despite these lower clocks, the EPYC still wins every single-core benchmark. This suggests the EPYC's architecture, possibly its larger L3 cache or different memory controller behavior, compensates for the clock deficit. The Threadripper's higher clocks do not translate into higher recorded scores.

The EPYC uses AMD Socket SP3, while the Threadripper uses SP3r2. Both are physically similar but electrically distinct, so they are not cross-compatible. The Threadripper's PCIe implementation is listed as Gen 3 with 60 lanes from the CPU, while the EPYC's PCIe details are not enumerated in the database beyond Gen 3 support. The EPYC's market segment is Server/Workstation, and its production status is Active. The Threadripper is marked End-of-life, which affects long-term availability.

Head-to-Head Benchmarks

The head-to-head data shows a clean sweep for the EPYC 7351. Starting with Cinebench R15 multi-core, the EPYC scores 1989 against the Threadripper's 1890, a 5.2% advantage. The single-core R15 test follows a similar pattern: 280 versus 266, a 5.3% lead. These are the earliest recorded tests in the database, and they establish the pattern that repeats across all subsequent Cinebench versions.

Moving to Cinebench R20, the EPYC extends its multi-core lead slightly in percentage terms. It scores 8291 versus 7877, a 5.3% delta. Single-core R20 shows 1170 against 1111, again 5.3%. The absolute point gaps grow with the newer test versions, but the relative margin stays fixed. This consistency across test generations suggests the performance relationship is stable and not dependent on specific Cinebench optimizations.

The newest recorded tests, Cinebench R23, tell the same story. Multi-core scores are 19742 for the EPYC and 18756 for the Threadripper, a 5.3% difference. Single-core scores are 2787 versus 2647, also 5.3% higher. The EPYC's wins total six out of six possible comparisons. The Threadripper records zero wins in the head-to-head dataset.

The average benchmark score reinforces this picture from a different angle. The EPYC 7351 has an average score of 5710 across all recorded benchmarks, placing it in the 61st percentile of all CPUs. The Threadripper 1920 averages 5425, landing in the 60th percentile. The percentile gap is small, just one point, but the raw average score difference is 285 points, about 5% higher for the EPYC.

Looking at nearest rivals, the EPYC 7351 sits close to the AMD EPYC 7F32, which averages 5703, a 0.1% difference. It also compares well to the Intel Core i9-7920X at 5673, where the EPYC is 0.7% ahead, and the Intel Xeon W-2175 at 5770, where it trails by 1%. The Threadripper 1920's nearest rivals include the Intel Xeon W-1290P at 5443, where the Threadripper is 0.3% behind, and the AMD EPYC 7351P at 5469, where it trails by 0.8%. These rival comparisons show both processors are competitive within their respective performance tiers, even though they do not face each other directly in the nearest-rival lists.

The Verdict

The data points to a clear choice for users who need maximum multi-threaded performance: the AMD EPYC 7351 wins every recorded benchmark against the AMD Ryzen Threadripper 1920. The margins are consistent, around 5.2% to 5.3% across all six Cinebench tests. The EPYC's 16 cores and 32 threads outpace the Threadripper's 12 cores and 24 threads in every scenario the database records.

For server and workstation deployments, the EPYC 7351 offers additional advantages beyond raw scores. It supports eight-channel memory with 170.6 GB/s bandwidth, double the Threadripper's 85.3 GB/s. It also supports ECC memory, which is critical for data integrity in server environments. Its Active production status means it remains available for new systems, whereas the Threadripper 1920 is End-of-life.

The Threadripper 1920 is not without merit for desktop users. Its 140 W TDP is lower than the EPYC's 170 W, which can simplify cooling requirements. Its higher base and boost clocks, 3.20 GHz and 3.80 GHz versus 2.40 GHz and 2.90 GHz, suggest it may feel more responsive in lightly threaded desktop tasks, even though the recorded single-core benchmarks favor the EPYC. The quad-channel memory bus is sufficient for most desktop workloads, and the desktop market segment means it fits into consumer-oriented motherboards with the SP3r2 socket.

The production status difference is decisive for long-term planning. The EPYC 7351 is Active, while the Threadripper 1920 is End-of-life. Organizations building new systems should favor the EPYC if they require ongoing availability. Individuals who already own a Threadripper 1920 system can still extract useful performance from it, but the database shows it lags the EPYC in every recorded test.

The percentile rankings are nearly identical, 61st for the EPYC and 60th for the Threadripper, indicating they occupy similar overall performance tiers relative to all CPUs. The EPYC's higher average score, 5710 versus 5425, confirms that its advantage is real but not overwhelming. Users choosing between these two should prioritize the EPYC for server-class workloads, ECC memory needs, and multi-channel bandwidth. Users who value lower TDP and desktop socket compatibility might still consider the Threadripper, but they should accept that it will trail in every Cinebench test recorded.

FAQ

Q: Which processor has more cores and threads?

A: The AMD EPYC 7351 has 16 cores and 32 threads. The AMD Ryzen Threadripper 1920 has 12 cores and 24 threads.

Q: Does the EPYC 7351 support ECC memory?

A: Yes, the EPYC 7351 supports ECC memory. The Ryzen Threadripper 1920 does not support ECC memory.

Q: What is the memory bandwidth difference between the two?

A: The EPYC 7351 has eight-channel memory with 170.6 GB/s bandwidth. The Threadripper 1920 has quad-channel memory with 85.3 GB/s bandwidth.

Q: Which processor has a higher boost clock?

A: The Ryzen Threadripper 1920 has a higher boost clock at 3.80 GHz, compared to the EPYC 7351's 2.90 GHz. The EPYC still wins all single-core benchmarks despite the lower clock.

Q: What are the production statuses of these processors?

A: The EPYC 7351 is marked as Active in production. The Ryzen Threadripper 1920 is marked as End-of-life.

Q: How many benchmark wins does each processor have in the head-to-head data?

A: The EPYC 7351 wins all six recorded head-to-head benchmarks. The Threadripper 1920 has zero wins in the head-to-head dataset.

Specification Differences

| Field | AMD EPYC 7351 | AMD Ryzen Threadripper 1920 |

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

| Cores | 16 | 12 |

| Threads | 32 | 24 |

| Base Clock | 2.40 GHz | 3.20 GHz |

| Boost Clock | 2.90 GHz | 3.80 GHz |

| TDP | 170 W | 140 W |

| Socket | AMD Socket SP3 | AMD Socket SP3r2 |

| Codename | Naples | Zen (Whitehaven) |

| Generation | EPYC (Zen (Naples)) | Ryzen Threadripper (Zen (Whitehaven)) |

| Transistors | 4,800 million | 9,600 million |

| Die Size | 213 mm² | 2x 213 mm² |

| L3 Cache | 64 MB (shared) | 32 MB |

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

| Memory Bandwidth | 170.6 GB/s | 85.3 GB/s |

| ECC Memory | True | False |

| PCIe | Gen 3 | Gen 3, 60 Lanes (CPU only) |

| Market Segment | Server/Workstation | Desktop |

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

| Release Date | 2017-06-28 | Not recorded |

| Average Benchmark Score | 5710 | 5425 |

| Percentile vs All CPUs | 61 | 60 |

The specification table highlights where the two designs diverge. The EPYC 7351 dedicates more silicon to cache and memory throughput, while the Threadripper 1920 uses a dual-die package with higher clock speeds. The EPYC's lower clock but higher benchmark scores suggest that memory bandwidth and cache capacity matter more for the recorded Cinebench workloads than raw clock speed. The Threadripper's higher transistor count comes from its two-die configuration, but those extra transistors do not translate into benchmark wins in this dataset.