AMD Ryzen Threadripper 9970X vs AMD Ryzen Threadripper 9980X Comparison

The AMD Ryzen Threadripper 9980X and AMD Ryzen Threadripper 9970X are both 9000-series desktop processors built on the Zen 5 architecture, sharing the same Shimada Peak codename, 4 nm TSMC process node, and AMD Socket sTR5 platform. The 9980X doubles the core count of the 9970X, offering 64 cores and 128 threads versus 32 cores and 64 threads, while both feature identical 5.40 GHz boost clocks and a 350 W TDP. The data shows a clear performance hierarchy, with the 9980X winning all 11 head-to-head benchmark comparisons, though the 9970X retains a significant advantage in single-threaded efficiency per core and a lower launch MSRP of $2499 versus $4999 for the 9980X.

FAQ

Q: Which processor has more cores and threads?

A: The AMD Ryzen Threadripper 9980X features 64 cores and 128 threads, while the AMD Ryzen Threadripper 9970X has 32 cores and 64 threads. This gives the 9980X exactly double the parallel processing capacity.

Q: How do their base and boost clocks compare?

A: The 9980X has a base clock of 3.20 GHz, while the 9970X starts higher at 4.00 GHz. Both processors share the identical 5.40 GHz boost clock, meaning the 9970X has a higher sustained minimum frequency but the same maximum single-core speed.

Q: What are the cache configurations?

A: The 9980X includes 256 MB of L3 cache, 1 MB of L2 cache per core, and 64 KB of L1 cache per core. The 9970X halves the L3 to 128 MB but retains the same per-core L1 and L2 allocations, resulting in less total cache due to its lower core count.

Q: Which processor offers better multi-threaded performance?

A: The 9980X dominates multi-threaded workloads, winning the PassMark multithread test by 31.9% with a score of 141641 versus 107399 for the 9970X. Its advantage grows in heavily parallel tasks like integer math, where it leads by 87.4%.

Q: Is the 9970X competitive in single-threaded performance?

A: Yes, the 9970X is nearly identical in single-threaded performance, scoring 4530 in the PassMark single-thread test versus 4537 for the 9980X. The delta is just 0.2%, making the two effectively equal for lightly threaded applications.

Q: Do both processors support the same memory and expansion features?

A: Yes, both support DDR5 memory with a quad-channel bus and 204.8 GB/s memory bandwidth, along with ECC memory support. Both also offer PCIe Gen 5 with 80 lanes from the CPU, and neither has integrated graphics.

Where Each One Wins

The AMD Ryzen Threadripper 9980X is the clear choice for heavily parallel, throughput-oriented workloads. Its 64 cores deliver decisive wins in every multi-threaded benchmark category, with the largest margins in integer math (87.4% ahead), floating-point math (80.5% ahead), and data encryption (81.1% ahead). For rendering, compilation, scientific simulations, or any task that scales linearly with core count, the 9980X’s raw compute density makes it the stronger option. The data compression test shows a 69.2% advantage, while extended instructions (typically AVX-style workloads) run 60.9% faster. Even in the less parallel physics simulation test, the 9980X holds a 17.1% lead, suggesting its higher core count compensates for the 9970X’s higher base clock in mixed workloads.

The AMD Ryzen Threadripper 9970X, while losing every head-to-head comparison, wins in scenarios where per-core frequency and cost efficiency matter more than absolute core count. Its 4.00 GHz base clock versus 3.20 GHz on the 9980X means it will feel snappier in lightly threaded tasks, despite the negligible 0.2% single-thread benchmark gap. For users who frequently work in single-threaded applications like legacy software, or who need a 32-core processor that still delivers 99th percentile performance overall, the 9970X provides most of the architectural benefits with half the cores. Its 350 W TDP matches the 9980X, but the lower core count means the 9970X is easier to keep cool under sustained load, making it a better fit for systems where cooling complexity is a concern.

Architecture Differences

Both processors are built on the Zen 5 architecture with the Shimada Peak codename, fabricated on TSMC’s 4 nm process node. The fundamental difference lies in chiplet configuration: the 9980X uses 8 compute dies, each measuring 70.6 mm², while the 9970X uses 4 of the same 70.6 mm² dies. This explains the transistor count disparity—66,520 million for the 9980X versus 33,260 million for the 9970X—exactly double, consistent with the doubling of cores and L3 cache. The per-die design is identical, with each core retaining 64 KB of L1 and 1 MB of L2 cache, but the 9980X aggregates 256 MB of L3 across its 8 dies versus 128 MB across 4 dies for the 9970X.

The memory subsystem is architecturally identical, with both supporting quad-channel DDR5 and 204.8 GB/s of memory bandwidth, but the 9980X’s additional dies mean more total cache to feed the extra cores. Both processors share the same socket (sTR5), PCIe Gen 5 with 80 lanes, and lack integrated graphics. The production status is active for both, and they were released on the same date. The key architectural takeaway is that the 9980X is effectively two 9970X processors fused onto one package, with no changes to the core design, memory controller, or I/O capabilities—only the number of dies and total cache scale.

Specification Differences

The two processors differ primarily in core and thread counts, with the 9980X offering 64 cores and 128 threads versus 32 cores and 64 threads on the 9970X. Base clocks differ significantly, with the 9970X running at 4.00 GHz and the 9980X at 3.20 GHz, though both boost to 5.40 GHz. The L3 cache is 256 MB on the 9980X and 128 MB on the 9970X, while per-core L1 (64 KB) and L2 (1 MB) caches are identical. The 9980X uses 8x 70.6 mm² dies with 66,520 million transistors, while the 9970X uses 4x 70.6 mm² dies with 33,260 million transistors. The launch MSRP is $4999 for the 9980X and $2499 for the 9970X, though both have the same 350 W TDP, DDR5 quad-channel memory support, 204.8 GB/s bandwidth, ECC memory, PCIe Gen 5 with 80 lanes, and no integrated graphics. Both are unlocked for overclocking, ship in the same sTR5 socket, and have distinct part numbers (100-000001593 for 9980X, 100-000001594 for 9970X).

Head-to-Head Benchmarks

The benchmark data shows a comprehensive sweep for the 9980X, winning all 11 comparisons, but the margins vary dramatically by workload. The largest victory comes in integer math, where the 9980X scores 872071 versus 465378 for the 9970X, a 87.4% delta that nearly doubles throughput—expected given the core count advantage. Floating-point math follows closely, with a 80.5% lead (559003 vs 309719), indicating the 9980X scales perfectly for numeric computation. Data encryption shows an 81.1% advantage (157137 vs 86765), while data compression trails slightly at 69.2% (2974534 vs 1757998), suggesting some memory bandwidth limitations but still a massive win.

The 9980X’s lead narrows in less parallel workloads. Extended instructions show a 60.9% advantage (228959 vs 142342), while random string sorting sees a 52.6% delta (292083 vs 191445). The multithread test, which measures overall parallel throughput, shows a 31.9% advantage (141641 vs 107399), a smaller margin than core count alone would suggest—likely due to the 9970X’s higher base clock helping in mixed workloads. Physics simulation shows just a 17.1% lead (8001 vs 6835), and prime number finding narrows to 25% (769 vs 615). The closest result is single-thread performance, where the 9980X edges out the 9970X by only 0.2% (4537 vs 4530), confirming that per-core IPC is identical.

In the context of nearest rivals, the 9980X’s average benchmark score of 321753 places it 0.3% ahead of the AMD Ryzen Threadripper PRO 9985WX, 2% ahead of the Intel Xeon 6781P, 3.2% ahead of the AMD EPYC 9575F, and 3.6% ahead of the AMD EPYC 9734. The 9970X’s average score of 279778 trails the Intel Xeon 6780E by 0.2%, the AMD EPYC 9565 by 2%, the Intel Xeon 696X by 2.2%, and the AMD EPYC 9555P by 2.5%. This places the 9970X in a lower performance tier despite its 99th percentile ranking, while the 9980X sits at the high end of that same percentile. Both processors achieve 99th percentile scores versus all CPUs, but the 9980X’s average score is 15% higher than the 9970X’s, reinforcing the core-count-driven performance gap.