AMD Ryzen Threadripper PRO 9975WX vs Intel Xeon 6745P Comparison

Head-to-Head Benchmarks

The benchmark comparison between the AMD Ryzen Threadripper PRO 9975WX and the Intel Xeon 6745P is remarkably one-sided. Out of 16 recorded head-to-head tests, the AMD processor wins 15, with the Intel chip taking a single victory. The margins are not trivial either; several workloads show double-digit percentage advantages for the Threadripper part.

Starting with the Cinebench suite, the AMD chip dominates across all six tests. In Cinebench R15 multicore, the Threadripper scores 9430 against the Xeon’s 7214, a 30.7% advantage. The single-core R15 test shows the same 30.7% delta, with 1331 versus 1018. This identical percentage repeats in Cinebench R20 multicore (39292 vs 30062), R20 single-core (5546 vs 4244), and R23 multicore (93553 vs 71578). The consistency of that 30.7% figure across every Cinebench workload is striking, suggesting a fundamental per-core performance advantage rather than a scaling quirk.

PassMark results paint a similar picture with some variation. Integer math shows the largest gap: 461724 for AMD versus 336926 for Intel, a 37% lead. Random string sorting follows closely at 40.8% (188014 vs 133528). Data encryption favors the Threadripper by 27.6% (85035 vs 66665), while extended instructions show a 27.1% delta (137733 vs 108326). Single-thread performance also leans AMD, with 4408 versus 3450, a 27.8% edge. Data compression, floating-point math, multithread, and physics tests all show AMD leads ranging from 14% to 24.6%.

The lone Intel win comes in the find prime numbers test, where the Xeon scores 681 versus AMD’s 620, a 9% advantage. This is a curious result, as prime-number generation often stresses integer throughput and cache behavior differently than other workloads. Still, it stands as the only category where Intel’s architecture gains ground.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD Ryzen Threadripper PRO 9975WX records an average benchmark score of 182700, while the Intel Xeon 6745P averages 154858. The AMD part sits in the 98th percentile of all CPUs, matching the Xeon’s percentile ranking.

Q: How does the Threadripper compare to its nearest rivals?

A: The AMD chip’s closest competitor is the AMD EPYC 8534P with an average score of 185092, putting the Threadripper 1.3% behind. It runs 1.5% ahead of the AMD EPYC 7763 (179916) and 3.7% ahead of the Intel Xeon 6740P (176227), while trailing the Intel Xeon 6740E (187718) by 2.7%.

Q: What are the Xeon 6745P’s closest competitors?

A: The Intel part’s nearest rival is the Intel Xeon 676X with an average score of 158540, placing the 6745P 2.3% behind. It also trails the AMD EPYC 9355P by 3.4% (160358), the AMD EPYC 7663 by 4.4% (161973), and the AMD EPYC 9375F by 4.7% (162497).

Q: Does the Intel processor win any benchmark category?

A: Yes, the Xeon 6745P wins the PassMark find prime numbers test with 681 points versus AMD’s 620, a 9% margin. This is the only test among 16 where Intel takes the lead.

Q: What is the largest performance gap between the two?

A: The biggest delta appears in PassMark random string sorting, where the Threadripper leads by 40.8% (188014 vs 133528). Integer math follows closely with a 37% gap (461724 vs 336926).

Q: Do both processors support ECC memory?

A: Yes, both the AMD Ryzen Threadripper PRO 9975WX and the Intel Xeon 6745P support ECC memory, and both use DDR5 with an eight-channel memory bus and 409.6 GB/s of memory bandwidth.

Architecture Differences

The architectural gap between these two processors is substantial. The AMD Ryzen Threadripper PRO 9975WX uses the Zen 5 architecture under the codename Shimada Peak, built on a 4 nm process at TSMC. The chip contains 33,260 million transistors across a die size of 4x 70.6 mm². In contrast, the Intel Xeon 6745P uses Granite Rapids architecture (also the codename) on a 5 nm process fabricated by Intel itself, with a die size of 2x 598 mm². Intel’s datasheet does not list a transistor count.

Cache configurations differ notably. The AMD part allocates 64 KB of L1 cache per core and 1 MB of L2 per core, with a shared 128 MB L3 cache. The Intel chip provides 112 KB of L1 per core and 2 MB of L2 per core, but its L3 cache is substantially larger at 336 MB shared. This L3 difference may explain some workload-specific behaviors, particularly in the prime number test where Intel wins.

PCIe lane counts also diverge. The Threadripper offers Gen 5 with 128 lanes (CPU only), while the Xeon provides Gen 5 with 88 lanes (CPU only). Both target the server/workstation market segment, and neither includes integrated graphics. The AMD processor has an unlocked multiplier, whereas the Intel part is locked. The AMD chip’s release date is later, 2025-07-22, compared to Intel’s 2025-02-23.

Specification Differences

Both processors feature 32 cores and 64 threads, so core count does not separate them. The AMD part runs a base clock of 4.00 GHz with a boost clock of 5.40 GHz, while the Intel chip operates at a base of 3.10 GHz and a boost of 4.30 GHz. Thermal design power differs as well: AMD lists 350 watts, Intel lists 300 watts.

Sockets are incompatible: AMD uses Socket sTR5, Intel uses Socket 4710. Process nodes differ (4 nm for AMD, 5 nm for Intel), and the foundries differ (TSMC versus Intel). The AMD chip reports 33,260 million transistors; the Intel datasheet omits this figure. Die sizes are constructed differently, with AMD’s 4x 70.6 mm² arrangement versus Intel’s 2x 598 mm². Cache hierarchy shows AMD’s 64 KB L1 and 1 MB L2 per core against Intel’s 112 KB L1 and 2 MB L2 per core, with L3 totals of 128 MB versus 336 MB.

Memory support is identical in type (DDR5), channel count (eight-channel), and bandwidth (409.6 GB/s). Both support ECC. PCIe generation is the same (Gen 5), but lane counts differ: 128 for AMD, 88 for Intel. Neither has integrated graphics. The AMD part is unlocked, the Intel part is not. Release dates differ by roughly five months. The AMD chip carries part number 100-000000723, while Intel’s is SRWPAQ7L9.

Where Each One Wins

The AMD Ryzen Threadripper PRO 9975WX is the clear winner for compute-heavy, multithreaded workloads. Its 30.7% lead across every Cinebench R15, R20, and R23 test positions it strongly for rendering, 3D animation, and video encoding tasks that scale with multicore performance. The 37% edge in integer math and 40.8% lead in random string sorting suggest advantages in data processing, scripting, and algorithmic tasks that rely on rapid sorting and integer operations. The 27.6% encryption advantage and 27.1% extended instructions lead indicate strength in security-related workloads, cryptography, and specialized instruction-heavy code. The 14% floating-point lead covers scientific computing, simulations, and financial modeling. The 24.6% multithread advantage and 18.6% physics lead reinforce its utility for simulation and physics-based rendering. For single-threaded responsiveness, the 27.8% single-thread lead means snappier day-to-day interactions in software that remains partially single-threaded.

The Intel Xeon 6745P wins only the prime number finding test, with a 9% lead. This narrow victory hints at an advantage in workloads that stress prime-number generation, which can appear in certain encryption key generation routines or mathematical research. The larger 336 MB L3 cache may contribute here, though the data does not specify the mechanism. Beyond that single test, the Xeon does not lead anywhere else in the recorded benchmarks. Its 300 watt TDP may make it more manageable in thermal-constrained environments, and its earlier release date means it has been available longer, but the performance data shows no other wins.

The Verdict

For almost every benchmark category, the AMD Ryzen Threadripper PRO 9975WX is the superior choice. The data shows a minimum 14% advantage across all PassMark tests and a consistent 30.7% lead across all Cinebench tests. The Threadripper’s higher boost clock (5.40 GHz versus 4.30 GHz), newer process node (4 nm versus 5 nm), and larger PCIe lane count (128 versus 88) align with its benchmark dominance. The unlocked multiplier adds flexibility for users who wish to push beyond stock settings, while the Xeon’s locked multiplier offers no such headroom.

The Intel Xeon 6745P does have a few qualitative merits. Its 336 MB L3 cache is more than double the AMD’s 128 MB, which may benefit certain cache-sensitive workloads, and the prime number test win suggests some niche applicability. The lower 300 watt TDP could be relevant in dense server deployments where power and cooling budgets are strict. The earlier release date also means the platform has been available for a longer period.

However, the benchmark record is unambiguous. A buyer needing maximum performance across rendering, computation, data encryption, and general multithreaded tasks should select the AMD Ryzen Threadripper PRO 9975WX. A buyer with a specialized workload centered on prime number discovery, or one constrained by the Xeon’s lower TDP and earlier availability, might consider the Intel Xeon 6745P, but that user would accept a 27.6% to 40.8% disadvantage in most other measured categories. The data says AMD wins this matchup decisively.