AMD Ryzen Threadripper 9970X vs Intel Xeon 6960P Comparison

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon 6960P has 72 cores and 144 threads, while the AMD Ryzen Threadripper 9970X has 32 cores and 64 threads. The Xeon offers more than double the core and thread count of the Threadripper.

Q: What are the clock speed differences between the two processors?

A: The AMD Ryzen Threadripper 9970X has a base clock of 4.00 GHz and a boost clock of 5.40 GHz. The Intel Xeon 6960P has a lower base clock of 2.70 GHz and a boost clock of 3.90 GHz. The Threadripper holds a significant clock speed advantage.

Q: How do they compare in single-threaded performance?

A: The AMD Ryzen Threadripper 9970X wins the single-thread benchmark with a score of 4530, compared to the Intel Xeon 6960P's score of 3287. This represents a 27.4% advantage for the Threadripper in that test.

Q: Which processor has higher memory bandwidth?

A: The Intel Xeon 6960P features a twelve-channel memory bus with 614.4 GB/s of bandwidth. The AMD Ryzen Threadripper 9970X uses a quad-channel memory bus with 204.8 GB/s. The Xeon provides three times the memory bandwidth of the Threadripper.

Q: What is the thermal design power (TDP) of each processor?

A: The Intel Xeon 6960P has a TDP of 500 watts, while the AMD Ryzen Threadripper 9970X has a TDP of 350 watts. The Xeon draws substantially more power.

Q: Which chip has a higher overall benchmark average?

A: The Intel Xeon 6960P has an average benchmark score of 365194, placing it in the 100th percentile of all CPUs. The AMD Ryzen Threadripper 9970X has an average benchmark score of 279778, placing it in the 99th percentile.

Architecture Differences

The Intel Xeon 6960P and AMD Ryzen Threadripper 9970X represent fundamentally different design philosophies. The Xeon 6960P is built on Intel's Granite Rapids architecture, specifically the Granite Rapids-AP generation, using a 5 nm process fabricated by Intel. The Threadripper 9970X uses AMD's Zen 5 architecture, codenamed Shimada Peak, fabricated by TSMC on a 4 nm process. The Xeon uses a massive die configuration of 3x 598 mm², while the Threadripper uses 4x 70.6 mm² chiplets, and AMD lists 33,260 million transistors for its design.

The core configurations diverge sharply. The Xeon 6960P packs 72 cores with 144 threads, each core carrying 112 KB of L1 cache and 2 MB of L2 cache, backed by a shared 432 MB L3 cache. The Threadripper 9970X offers 32 cores and 64 threads, with 64 KB of L1 cache per core, 1 MB of L2 per core, and 128 MB of L3. The Xeon's cache hierarchy is clearly engineered for massive data residency, with over three times the L3 capacity.

Memory subsystems differ as much as the core counts. The Xeon 6960P supports DDR5 across a twelve-channel memory bus, delivering 614.4 GB/s of bandwidth. The Threadripper 9970X also supports DDR5 but on a quad-channel bus, yielding 204.8 GB/s. Both support ECC memory. PCIe connectivity also differs: the Xeon provides Gen 5 with 96 lanes from the CPU, while the Threadripper provides Gen 5 with 80 lanes.

The market positioning is explicit in the data. The Xeon 6960P targets the Server/Workstation segment, uses Intel Socket 7529, has a locked multiplier, and launched on 2024-09-23 with a launch MSRP of $9625. The Threadripper 9970X targets the Desktop segment, uses AMD Socket sTR5, has an unlocked multiplier, and launched on 2025-07-29 with a launch MSRP of $2499. Neither chip includes integrated graphics.

Head-to-Head Benchmarks

The head-to-head benchmark results show a dominant performance profile for the Intel Xeon 6960P, which wins 9 of the 11 recorded tests. The most lopsided result is in the passmark_physics test, where the Xeon scores 24937 against the Threadripper's 6835, a 264.8% advantage. This test rewards the Xeon's massive core and thread count, and the margin is enormous.

The Xeon also excels in passmark_find_prime_numbers, scoring 1484 versus 615 for the Threadripper, a 141.3% lead. Data encryption shows an 86.7% advantage for the Xeon (162013 versus 86765), and random string sorting shows a 94.2% lead (371795 versus 191445). Floating point math goes to the Xeon by 70.3% (527473 versus 309719), while integer math favors the Xeon by 56.4% (727750 versus 465378). Data compression shows a 59.1% Xeon advantage (2797724 versus 1757998), and extended instructions land at 35.9% in favor of the Xeon (193404 versus 142342).

The multithread benchmark is closer, with the Xeon at 130659 and the Threadripper at 107399, a 21.7% margin. This narrower gap is notable because it suggests the Threadripper's higher clocks help it close some distance in a mixed workload, though the Xeon still takes the win.

The single-thread tests are the only category where the AMD Ryzen Threadripper 9970X comes out ahead. The Threadripper scores 4530 in passmark_single_thread against the Xeon's 3287, a 27.4% advantage. This result aligns with the Threadripper's 5.40 GHz boost clock versus the Xeon's 3.90 GHz, and it is the clearest signal of where the AMD chip's architectural strengths lie.

Specification Differences

The specification table between the two processors shows differences across nearly every major category. Core count: the Xeon 6960P has 72 cores versus 32 for the Threadripper 9970X. Thread count: 144 versus 64. Base clock: 2.70 GHz versus 4.00 GHz. Boost clock: 3.90 GHz versus 5.40 GHz. TDP: 500 watts versus 350 watts.

Cache hierarchies are distinct across all levels. L1 cache is 112 KB per core on the Xeon versus 64 KB per core on the Threadripper. L2 cache is 2 MB per core versus 1 MB per core. L3 cache is 432 MB shared versus 128 MB. The Xeon's total cache footprint is substantially larger at every level.

Memory channels and bandwidth differ significantly. The Xeon uses twelve channels with 614.4 GB/s, while the Threadripper uses four channels with 204.8 GB/s. PCIe lanes also differ: 96 versus 80, both Gen 5. Socket types are entirely different, with Intel Socket 7529 on the Xeon and AMD Socket sTR5 on the Threadripper.

Process technology and foundry differ as well. The Xeon is built on a 5 nm process at Intel, while the Threadripper uses a 4 nm process at TSMC. The die sizes are radically different: the Xeon uses 3x 598 mm² dies, while the Threadripper uses 4x 70.6 mm² chiplets. The Threadripper explicitly lists 33,260 million transistors, while the Xeon does not provide a transistor count in the database.

Other differences include the multiplier: the Xeon is locked, the Threadripper is unlocked. The release dates are roughly ten months apart, with the Xeon launching on 2024-09-23 and the Threadripper on 2025-07-29. The launch MSRP is $9625 for the Xeon and $2499 for the Threadripper. The Xeon's production status is Active, as is the Threadripper's.

Where Each One Wins

The Intel Xeon 6960P is the clear winner for heavily parallel, throughput-oriented workloads. Its 9 benchmark wins out of 11 head-to-head tests include dominant margins in physics simulation, prime number finding, encryption, random string sorting, floating point math, integer math, and data compression. The 264.8% lead in physics and the 141.3% lead in prime numbers indicate that any workload which scales with core count will strongly favor the Xeon. Its 432 MB of L3 cache and 614.4 GB/s of memory bandwidth make it suited for large in-memory datasets and multi-tenant server environments. The nearest rivals in the database for the Xeon are AMD EPYC parts, with the EPYC 9754 within 0.2% of its average score, which shows the Xeon competes at the top tier of server silicon.

The AMD Ryzen Threadripper 9970X wins where clock speed and single-thread efficiency matter. Its 27.4% advantage in single-thread performance over the Xeon is the standout result, a direct consequence of its 5.40 GHz boost clock versus 3.90 GHz. The Threadripper's nearest rivals in the database are Intel Xeon 6780E and AMD EPYC 9565, with scores within 2% of the Threadripper's average, indicating it sits in a different performance class than the Xeon 6960P. For desktop workstation users who run lightly threaded applications, interactive workloads, or software that responds to high per-core performance, the Threadripper is the better fit. Its unlocked multiplier also allows overclocking, which the Xeon does not permit.

The trade-off between the two is sharp and structural. The Xeon 6960P offers massive parallelism, huge cache, and three times the memory bandwidth, but at 500 watts TDP and a $9625 launch MSRP. The Threadripper 9970X offers higher clocks, better single-thread performance, lower power draw at 350 watts, and a $2499 launch MSRP, but with a fraction of the core count and memory bandwidth. The average benchmark score of 365194 for the Xeon versus 279778 for the Threadripper reflects the Xeon's overall throughput dominance, yet the Threadripper's 99th percentile ranking among all CPUs shows it remains a high-end part in its own right. Workloads that can use 144 threads will find the Xeon nearly unstoppable. Workloads that need fast single-core response will prefer the Threadripper. The data does not show a single winner across all scenarios; it shows two processors optimized for different ends of the computing spectrum.