AMD EPYC 9755 vs AMD Ryzen Threadripper 9980X Comparison

The AMD EPYC 9755 and AMD Ryzen Threadripper 9980X are both Zen 5 parts, but they target different extremes of the same family tree. The EPYC 9755 is a 128-core server processor built for scale-out throughput, while the Threadripper 9980X is a 64-core desktop part aimed at high-frequency responsiveness. The benchmark data shows a clear split: the EPYC wins 15 of 17 head-to-head tests, while the Threadripper takes only 2, both in single-threaded workloads. The average benchmark scores reinforce this divide, with the EPYC at 505,778 against the Threadripper’s 321,753, a gap of roughly 57%. However, the Threadripper’s higher clock speeds and lower core count make it the better choice for latency-sensitive tasks, as the data will show.

Head-to-Head Benchmarks

The EPYC 9755’s dominance is most pronounced in integer and floating-point math. In PassMark integer math, the EPYC scores 1,549,946 versus the Threadripper’s 872,071, a 77.7% advantage. Floating-point math shows a similar story: 922,900 for the EPYC against 559,003, a 65.1% lead. The EPYC’s 128 cores simply overpower the Threadripper’s 64 in raw parallel compute.

The largest single margin comes in PassMark physics, where the EPYC scores 27,806 compared to the Threadripper’s 8,001, a staggering 247.5% delta. This test likely scales with core count and memory bandwidth, both of which heavily favor the EPYC. Prime number finding also shows a massive gap: 2,047 for the EPYC versus 769, a 166.2% difference, suggesting the EPYC’s larger L3 cache and higher thread count are critical for this workload.

Data compression and encryption follow the same pattern. The EPYC scores 4,517,407 in data compression, 51.9% ahead of the Threadripper’s 2,974,534. Data encryption shows an 81.3% lead for the EPYC (284,927 vs 157,137). Random string sorting is another clear win, with the EPYC at 571,185 versus 292,083, a 95.6% margin. These workloads benefit from the EPYC’s 512 MB of shared L3 cache and twelve-channel memory bus.

Cinebench results are consistent but narrower. Across R15, R20, and R23, the EPYC wins both multi-core and single-core by exactly 8.3% in every test. For multi-core, the EPYC scores 14,250 (R15), 59,378 (R20), and 141,378 (R23) against the Threadripper’s 13,157, 54,822, and 130,529. The single-core Cinebench scores are also 8.3% higher for the EPYC (2,011 vs 1,857 in R15; 8,382 vs 7,739 in R20; 19,959 vs 18,427 in R23). This suggests that despite the Threadripper’s higher boost clock, the EPYC’s architecture delivers better per-thread performance in these rendering tests.

The Threadripper’s only wins come in PassMark single-thread tests. It scores 4,537 against the EPYC’s 3,503, a 22.8% advantage. This is the single largest win for the Threadripper and highlights its strength in low-latency, single-threaded tasks. PassMark multi-thread shows the EPYC ahead by 17.4% (166,328 vs 141,641), which is smaller than the core count difference suggests, indicating that the Threadripper’s higher clocks help it partially close the gap in this particular test.

The EPYC’s average benchmark score of 505,778 places it in the 100th percentile of all CPUs, while the Threadripper’s 321,753 lands in the 99th percentile. The EPYC’s nearest rival is the AMD EPYC 9845, which scores 523,613, putting the 9755 at -3.4% relative to that part. The Threadripper’s nearest rival is the AMD Ryzen Threadripper PRO 9985WX at 320,749, a 0.3% difference, showing they are nearly identical in average performance.

Architecture Differences

Both processors use TSMC’s 4 nm process and the Zen 5 architecture, but their implementations diverge sharply. The EPYC 9755 is built with 133,040 million transistors across 16 chiplets, each 70.6 mm², while the Threadripper uses 66,520 million transistors across 8 chiplets of the same die size. This means the EPYC has exactly double the chiplet count, which explains its 128 cores versus the Threadripper’s 64.

Cache configurations differ significantly. The EPYC provides 80 KB of L1 per core, 1 MB of L2 per core, and a massive 512 MB of shared L3. The Threadripper offers 64 KB of L1 per core, 1 MB of L2 per core, and 256 MB of L3. The EPYC’s L3 is double the Threadripper’s, which is a major factor in its 95.6% lead in random string sorting and its 166.2% advantage in prime number finding.

Memory subsystems are another key divergence. The EPYC uses a twelve-channel DDR5 bus with 576.0 GB/s of bandwidth, while the Threadripper has a quad-channel bus with 204.8 GB/s. Both support ECC memory, but the EPYC’s bandwidth is 2.8 times higher, which is critical for data-heavy server workloads. PCIe lanes also differ: the EPYC provides 128 Gen 5 lanes (CPU only), while the Threadripper offers 80 Gen 5 lanes.

Socket and platform are distinct. The EPYC uses AMD Socket SP5, a server platform, while the Threadripper uses AMD Socket sTR5, aimed at high-end desktops. The EPYC has a locked multiplier, whereas the Threadripper is multiplier-unlocked, allowing overclocking. Power and thermal envelopes differ as well, with the EPYC rated at a 500 W TDP and the Threadripper at 350 W, though exact thermal implications are not in the data.

Release timing is also different. The EPYC launched on 2024-10-09, while the Threadripper came later on 2025-07-29. Both are listed as Active in production status. The EPYC’s part number is 100-000001443, and the Threadripper’s is 100-000001593.

FAQ

Q: Which processor has more cores?

A: The AMD EPYC 9755 has 128 cores and 256 threads, while the AMD Ryzen Threadripper 9980X has 64 cores and 128 threads. The EPYC has exactly double the core count.

Q: What is the single-threaded performance difference?

A: The Threadripper 9980X wins the PassMark single-thread test, scoring 4,537 against the EPYC’s 3,503, a 22.8% advantage. However, in Cinebench R23 single-core, the EPYC scores 19,959 versus the Threadripper’s 18,427, an 8.3% lead for the EPYC.

Q: How does cache size compare?

A: The EPYC 9755 has 512 MB of shared L3 cache and 80 KB of L1 per core. The Threadripper 9980X has 256 MB of L3 and 64 KB of L1 per core. Both have 1 MB of L2 per core.

Q: What is the memory bandwidth difference?

A: The EPYC 9755 uses a twelve-channel DDR5 bus with 576.0 GB/s bandwidth. The Threadripper 9980X uses a quad-channel bus with 204.8 GB/s. The EPYC offers 2.8 times more bandwidth.

Q: Which processor is better for integer math?

A: The EPYC 9755 scores 1,549,946 in PassMark integer math, which is 77.7% higher than the Threadripper’s 872,071. This indicates the EPYC is far stronger in parallel integer workloads.

Q: Are these processors on the same socket?

A: No. The EPYC 9755 uses AMD Socket SP5, while the Threadripper 9980X uses AMD Socket sTR5. They are not interchangeable.

Specification Differences

| Field | AMD EPYC 9755 | AMD Ryzen Threadripper 9980X |

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

| Cores | 128 | 64 |

| Threads | 256 | 128 |

| Base Clock | 2.70 GHz | 3.20 GHz |

| Boost Clock | 4.10 GHz | 5.40 GHz |

| TDP | 500 W | 350 W |

| Socket | AMD Socket SP5 | AMD Socket sTR5 |

| Codename | Turin | Shimada Peak |

| Transistors | 133,040 million | 66,520 million |

| Die Size | 16x 70.6 mm² | 8x 70.6 mm² |

| L1 Cache | 80 KB (per core) | 64 KB (per core) |

| L3 Cache | 512 MB (shared) | 256 MB |

| Memory Bus | Twelve-channel | Quad-channel |

| Memory Bandwidth | 576.0 GB/s | 204.8 GB/s |

| PCIe | Gen 5, 128 Lanes (CPU only) | Gen 5, 80 Lanes (CPU only) |

| Market Segment | Server/Workstation | Desktop |

| Release Date | 2024-10-09 | 2025-07-29 |

| Launch MSRP | $12984 | $4999 |

| Multiplier Unlocked | false | true |

| Part Number | 100-000001443 | 100-000001593 |

The Verdict

The data is unambiguous: the AMD EPYC 9755 is the superior processor for multi-threaded and server-class workloads. It wins 15 of 17 head-to-head benchmarks, including all Cinebench tests and the vast majority of PassMark tests. Its 128 cores, 512 MB L3 cache, and 576.0 GB/s memory bandwidth give it massive advantages in compute-heavy tasks, ranging from 8.3% in Cinebench to 247.5% in physics simulations. The EPYC also holds a perfect 100th percentile ranking, while the Threadripper sits at 99th. For anyone running parallel workloads that scale with cores and memory bandwidth, the EPYC is the clear choice, despite its higher launch MSRP of $12984.

The Threadripper 9980X has one distinct advantage: single-threaded performance. Its PassMark single-thread score of 4,537 is 22.8% higher than the EPYC’s 3,503. This, combined with its higher boost clock of 5.40 GHz versus 4.10 GHz, makes it better suited for latency-sensitive applications where a single thread is the bottleneck. Its lower launch MSRP of $4999 and unlocked multiplier also make it a more flexible option for desktop users who prioritize overclocking and raw clock speed over core count. However, its quad-channel memory bus and 256 MB L3 cache put it at a structural disadvantage for memory-hungry workloads.

Where Each One Wins

The EPYC 9755 wins in every multi-threaded category. Its 166.2% lead in prime number finding and 247.5% lead in physics suggest it excels in scientific computing and simulation workloads. The 81.3% advantage in data encryption and 77.7% in integer math make it ideal for database processing, financial modeling, and any task involving heavy arithmetic. The 95.6% margin in random string sorting points to strengths in data analysis and sorting algorithms. Its 512 MB L3 cache and twelve-channel memory bus are the likely drivers of these wins, making it the pick for virtualized environments, large-scale data processing, and rendering farms.

The Threadripper 9980X wins where a single thread matters most. Its 22.8% lead in PassMark single-thread performance indicates it is better for older software that relies on one core, or for interactive tasks like gaming, real-time audio processing, or low-latency input handling. Its higher base clock of 3.20 GHz versus 2.70 GHz also helps in bursty workloads that don’t fully utilize all cores. However, the data shows its multi-threaded performance is still strong—only 17.4% behind the EPYC in PassMark multi-thread despite half the cores—which suggests it can handle many tasks well, just not at the extreme scale of the EPYC. For a desktop user who needs occasional high-core-count bursts but prioritizes single-thread speed, the Threadripper is the logical pick.