AMD Ryzen Threadripper PRO 9975WX vs Intel Xeon 676X Comparison
AMD Ryzen Threadripper PRO 9975WX
Xeon 676X
PERFORMANCE BENCHMARKS
Analysis: AMD Ryzen Threadripper PRO 9975WX vs Intel Xeon 676X
Head-to-Head Benchmarks
The benchmark data paints a remarkably one-sided picture. Across the 16 recorded tests, the AMD Ryzen Threadripper PRO 9975WX takes 14 wins, while the Intel Xeon 676X manages only 2. The margins, however, are not uniform, and the two Intel victories point to a specific workload profile worth noting.
Starting with the Cinebench suite, the AMD part dominates by a near-consistent margin. In Cinebench R15 multi-core, the Threadripper PRO scores 9430 against the Xeon's 7806, a 20.8% lead. The single-core result is almost identical in percentage terms: 1331 versus 1101, a 20.9% advantage. This pattern repeats in R20, where multi-core shows 39292 versus 32527 and single-core shows 5546 versus 4591, both at 20.8% deltas. In Cinebench R23 multi-core, the AMD chip posts 93553 against 77447, again a 20.8% gap. The consistency of these numbers suggests a fundamental clock-speed and IPC advantage for the AMD processor across the entire Cinebench workload family.
The PassMark results tell a more varied story. In data compression, the Threadripper PRO scores 1644573 versus 1355807, a 21.3% win. Data encryption shows a larger gap: 85035 versus 67638, a 25.7% edge for AMD. Extended instructions is the biggest AMD victory, with 137733 versus 105231, a 30.9% margin. Integer math follows closely at 461724 versus 354777, a 30.1% lead. Random string sorting is the largest single delta of all: 188014 versus 137976, a 36.3% blowout in favor of AMD.
The AMD wins continue in floating-point math, where 304833 beats 283570 by 7.5%, and in the PassMark multi-thread test, where 104902 versus 91115 represents a 15.1% advantage. Single-thread performance is closer but still favors AMD: 4408 versus 4015, a 9.8% lead. Notably, the two PassMark single-thread entries report identical scores, confirming data consistency.
The Intel Xeon 676X takes its two wins in specific compute-heavy tests. In the find prime numbers benchmark, the Xeon scores 738 against AMD's 620, a 16% advantage. In physics, the Xeon posts 8281 versus 7288, a 12% lead. These are the only two tests where Intel's architecture shows a clear edge, and both are computationally intensive integer workloads that appear to benefit from the Xeon's larger per-core L2 cache and different memory hierarchy.
Looking at the broader database context, the AMD part's average benchmark score is 182700, placing it in the 98th percentile of all CPUs. The Xeon 676X averages 158540, also in the 98th percentile. The AMD chip's nearest rivals include the AMD EPYC 8534P (1.3% higher average score) and the Intel Xeon 6740E (2.7% higher), while the Xeon 676X sits near the AMD EPYC 9355P (1.1% higher) and the AMD EPYC 9375F (2.4% higher). Neither processor is the outright leader in its respective peer group, but the Threadripper PRO's raw scores are consistently higher across the board.
The Verdict
The data points to a clear hierarchy. The AMD Ryzen Threadripper PRO 9975WX is the faster processor in 14 of 16 benchmarks, with decisive wins in rendering, encryption, compression, and integer operations. The only tests where the Intel Xeon 676X pulls ahead are prime number finding and physics simulation, where its 16% and 12% margins, respectively, show strength in specific integer-heavy workloads.
For users running Cinebench-style rendering workloads, multi-threaded content creation, or general server tasks involving data compression and encryption, the Threadripper PRO 9975WX delivers a consistent 20% or greater advantage. The 30.9% lead in extended instructions and 30.1% lead in integer math suggest the AMD architecture is particularly well-suited to AVX-class and general compute tasks.
The Xeon 676X remains a viable choice only for workloads that mirror the find prime numbers or physics test patterns. If the application is dominated by prime-number sieving or rigid-body physics calculations, the Intel chip's 16% and 12% wins, respectively, could translate into real time savings. But for everything else, the AMD part is the data-driven pick.
The percentile rankings are identical at 98, which means both are elite parts. But within that elite tier, the AMD processor's average score of 182700 versus 158540 represents a 15.2% overall advantage. That is not a marginal difference; it is a generational gap in performance.
FAQ
Q: Which processor wins more benchmarks?
A: The AMD Ryzen Threadripper PRO 9975WX wins 14 of 16 head-to-head tests. The Intel Xeon 676X wins only 2: find prime numbers and physics.
Q: What is the largest performance gap between the two?
A: The biggest delta is in random string sorting, where the AMD part scores 188014 versus 137976, a 36.3% lead. The next largest is in extended instructions at 30.9% and integer math at 30.1%.
Q: Are there any workloads where the Intel Xeon 676X is faster?
A: Yes, in the PassMark find prime numbers test, the Xeon scores 738 versus 620, a 16% lead. In the physics test, it scores 8281 versus 7288, a 12% advantage.
Q: How do the Cinebench scores compare?
A: The AMD processor leads by 20.8% in R15 multi-core, R20 multi-core, and R23 multi-core. Single-core margins are also around 20.8% in R15 and R20. The R23 multi-core score is 93553 for AMD versus 77447 for Intel.
Q: What are the average benchmark scores?
A: The AMD Ryzen Threadripper PRO 9975WX has an average benchmark score of 182700. The Intel Xeon 676X averages 158540. Both are in the 98th percentile of all CPUs.
Q: How close are the single-thread scores?
A: The AMD part leads with 4408 versus 4015, a 9.8% margin. This is one of the smaller gaps, suggesting the Xeon's single-thread performance is competitive but not sufficient to overcome the AMD lead in multi-thread tests.
Specification Differences
The two processors share several fundamental specifications. Both have 32 cores and 64 threads, both support DDR5 memory in an eight-channel configuration, and both offer Gen 5 PCIe with 128 lanes (CPU only). Both have ECC memory support, no integrated graphics, and target the server/workstation market segment. Both are currently active production parts.
The differences start with clock speeds. The AMD Ryzen Threadripper PRO 9975WX runs at 4.00 GHz base and 5.40 GHz boost. The Intel Xeon 676X runs at 2.80 GHz base and 4.90 GHz boost. That is a substantial 1.20 GHz base clock deficit and a 0.50 GHz boost deficit for the Intel part.
Thermal design power differs as well. The AMD processor has a 350 W TDP, while the Intel part is rated at 275 W. The sockets are incompatible: AMD uses Socket sTR5, while Intel uses Socket 4710.
The AMD part is unlocked for overclocking, as is the Intel part. Both have multiplier unlocked status. The release dates differ, with the AMD part launching on 2025-07-22 and the Intel part on 2026-02-01.
The launch MSRP for the AMD Ryzen Threadripper PRO 9975WX is $4099. The Intel Xeon 676X has a launch MSRP of $2499. The part numbers are 100-000000723 for AMD and SA2CY for Intel.
Architecture Differences
The architectural divide is significant. The AMD Ryzen Threadripper PRO 9975WX uses the Zen 5 architecture with the codename Shimada Peak, built on a 4 nm process at TSMC. The Intel Xeon 676X uses the Granite Rapids architecture, also codenamed Granite Rapids, built on a 5 nm process at Intel's own foundry.
Transistor counts are only provided for the AMD part: 33,260 million. The die size also differs, with AMD using 4x 70.6 mm² chiplets, while Intel uses 2x 598 mm² dies. This reflects fundamentally different design approaches: AMD's chiplet strategy versus Intel's monolithic (or dual-die) design.
Cache hierarchies diverge in meaningful ways. The AMD L1 cache is 64 KB per core, while Intel provides 112 KB per core. AMD's L2 is 1 MB per core, while Intel doubles that to 2 MB per core. The L3 cache is 128 MB for AMD versus 144 MB shared for Intel. The larger per-core L2 on the Intel part likely contributes to its wins in the prime number and physics tests.
Memory bandwidth is identical at 409.6 GB/s for both, given the same eight-channel DDR5 configuration. The memory bus is also the same at eight channels.
The AMD part belongs to the 9000 series (Ryzen Threadripper generation), while the Intel part is from the Xeon 600 generation (Granite Rapids-WS). The AMD codename is Shimada Peak; the Intel codename is Granite Rapids. The process node advantage goes to AMD at 4 nm versus Intel's 5 nm, which helps explain the clock speed and efficiency differences.
Where Each One Wins
The AMD Ryzen Threadripper PRO 9975WX is the clear winner for rendering workloads. The Cinebench R15, R20, and R23 results all show a 20.8% advantage, making this the go-to part for 3D rendering, video encoding, and any multi-threaded content creation task. The 21.3% lead in data compression and 25.7% lead in data encryption also make it the stronger choice for file archiving, database workloads, and secure communications.
For general compute and development, the AMD part's 30.9% lead in extended instructions and 30.1% lead in integer math indicate superior performance for compilation, scientific computing, and AVX-heavy code. The 15.1% lead in the PassMark multi-thread test reinforces that this is a better all-round workstation processor. The 36.3% lead in random string sorting is particularly relevant for data processing and text-heavy analytics.
The Intel Xeon 676X has a narrower but real niche. The 16% win in find prime numbers and the 12% win in physics suggest that workloads involving number theory, cryptography (some forms), or physics simulation with rigid-body dynamics could favor the Intel part. The larger 2 MB per-core L2 cache likely helps these specific integer loops fit entirely in fast memory. For a user running a dedicated prime-number search or a physics engine that mirrors the PassMark physics test, the Xeon 676X is the data-backed choice.
The overall average score gap favors AMD by roughly 15%, but the Intel part's lower TDP (275 W versus 350 W) and lower launch MSRP ($2499 versus $4099) are specification facts that may matter to system integrators. The data, however, is unambiguous: for the vast majority of workloads, the AMD Ryzen Threadripper PRO 9975WX delivers more performance. Only in two very specific compute patterns does the Xeon 676X pull ahead.