AMD Ryzen Threadripper PRO 5975WX vs Intel Xeon 674X Comparison

FAQ

Q: Which processor wins more benchmarks in the database?

A: The Intel Xeon 674X wins 13 of the 16 head-to-head benchmark comparisons, while the AMD Ryzen Threadripper PRO 5975WX takes the remaining 3.

Q: How large is the average performance gap between the two?

A: The Intel Xeon 674X holds a higher average benchmark score of 143,103 compared to the AMD chip's 124,171. The Xeon also sits at the 98th percentile of all CPUs, while the Threadripper PRO sits at the 97th percentile.

Q: In which workloads does the AMD processor outperform the Intel one?

A: The AMD Ryzen Threadripper PRO 5975WX wins in data compression by 4.4%, data encryption by 23.9%, and integer math by 14.2%.

Q: What is the single biggest performance difference in the head-to-head results?

A: The largest delta is in PassMark physics, where the Intel Xeon 674X leads by 74%. The second largest is in finding prime numbers, where Intel wins by 58.2%.

Q: Which chip has more cores and threads?

A: The AMD Ryzen Threadripper PRO 5975WX has 32 cores and 64 threads. The Intel Xeon 674X has 28 cores and 56 threads.

Q: What are the differences in memory support and PCIe generation?

A: The Intel Xeon 674X supports DDR5 memory with 409.6 GB/s bandwidth and PCIe Gen 5 with 128 lanes. The AMD chip supports DDR4 memory with 204.8 GB/s bandwidth and PCIe Gen 4 with 128 lanes.

Architecture Differences

The two processors come from fundamentally different design philosophies. The Intel Xeon 674X is built on the Granite Rapids architecture, using a 5 nm process from Intel's own foundry. It is part of the Xeon 600 series (Granite Rapids-WS) and features a dual-die layout with a combined die size of 2x 598 mm². The AMD Ryzen Threadripper PRO 5975WX uses the Zen 3 architecture under the codename Chagall PRO, manufactured on TSMC's 7 nm process. It packs 16,600 million transistors across four chiplets, each 81 mm².

Cache organization differs significantly. The Intel chip provides 112 KB of L1 cache per core and 2 MB of L2 per core, with a shared 144 MB L3 cache. The AMD processor offers 64 KB of L1 per core and 512 KB of L2 per core, with a total of 128 MB of L3. The Intel design clearly favors larger per-core caches and a bigger L3 pool, which helps explain its strong showing in cache-sensitive workloads.

Memory architecture is another major split. The Xeon 674X supports DDR5 across an eight-channel memory bus, delivering 409.6 GB/s of bandwidth. The Threadripper PRO 5975WX uses DDR4, also eight-channel, but with a peak bandwidth of 204.8 GB/s. That is exactly half the bandwidth of the Intel part, and it shows in multi-threaded throughput tests.

Both processors support ECC memory and have 128 CPU-only PCIe lanes, but the Intel chip runs PCIe Gen 5 while the AMD runs Gen 4. Clock behavior also differs: the Xeon has a base clock of 3.00 GHz and boosts to 4.90 GHz, while the Threadripper PRO has a higher base of 3.60 GHz but a lower boost of 4.50 GHz. The Intel chip is multiplier unlocked, whereas the AMD chip is locked. The Xeon also has a larger TDP envelope at 270 W versus 280 W for the AMD, which is unusual: the AMD part has more cores but slightly lower power draw.

Head-to-Head Benchmarks

The Cinebench results form a consistent block of Intel dominance. In Cinebench R15 multi-core, the Xeon scores 7,213 against 6,453 for the AMD chip, an 11.8% lead. Single-core R15 shows 1,018 versus 910, an 11.9% gap. The R20 runs mirror this: 30,057 versus 26,888 in multi-core and 4,243 versus 3,795 in single-core, both at 11.8%. Cinebench R23 multi-core continues the pattern with 71,566 versus 64,021, again 11.8%. The consistency of this 11.8% delta across all Cinebench multi-core tests suggests a stable architectural advantage, not a workload-specific quirk.

PassMark multi-thread follows the same trend: 84,196 for the Intel chip versus 75,319 for the AMD, an 11.8% win. PassMark single-thread shows 3,933 versus 3,323, an 18.4% lead for the Xeon. That single-thread margin is meaningful for lightly threaded tasks and everyday responsiveness.

The Intel chip also wins decisively in math and logic workloads. PassMark floating point math goes to the Xeon at 243,877 versus 202,363, a 20.5% lead. Extended instructions are another Intel win: 97,373 versus 82,850, a 17.5% margin. The prime number search test is not close: 693 versus 438, a 58.2% blowout. Physics simulation is the largest gap of all: 7,586 versus 4,360, a 74% advantage for the Intel part. Random string sorting is close but still goes Intel's way: 127,529 versus 123,527, a 3.2% margin.

The AMD processor has its own territory. Data compression goes to the Threadripper PRO at 1,293,784 versus 1,236,272, a 4.4% edge. Data encryption is a much bigger win: 80,453 versus 61,195, a 23.9% lead. Integer math also favors AMD: 360,156 versus 308,968, a 14.2% margin. These three wins suggest the AMD chip has strong integer execution and encryption acceleration capabilities, even though it trails in most other measured workloads.

Specification Differences

The two CPUs diverge on nearly every specification field. The Intel Xeon 674X has 28 cores and 56 threads, while the AMD Ryzen Threadripper PRO 5975WX has 32 cores and 64 threads. The AMD chip offers 4 more cores and 8 more threads. Base clocks differ: 3.00 GHz for Intel versus 3.60 GHz for AMD. Boost clocks flip the order: 4.90 GHz for Intel versus 4.50 GHz for AMD.

Power envelopes are close but not identical. The Intel part has a TDP of 270 W, the AMD part 280 W. Process nodes differ by generation: Intel uses 5 nm, AMD uses 7 nm. The Intel die is 2x 598 mm², while the AMD design uses 4x 81 mm² chiplets with 16,600 million transistors. Transistor count for Intel is not recorded in the database.

Cache sizes are a clear differentiator. Intel provides 112 KB of L1 per core, 2 MB of L2 per core, and 144 MB of shared L3. AMD offers 64 KB of L1 per core, 512 KB of L2 per core, and 128 MB of L3. The Intel chip has both larger per-core caches and a larger L3 pool.

Memory support is a generation apart. Intel runs DDR5 with 409.6 GB/s bandwidth; AMD runs DDR4 with 204.8 GB/s. Both are eight-channel and both support ECC. PCIe generation differs: Intel is Gen 5, AMD is Gen 4, both with 128 CPU-only lanes. The Intel chip is multiplier unlocked, the AMD chip is not. Release dates sit years apart: Intel is dated 2026-02-01, AMD is dated 2022-03-07. The Intel launch MSRP is $2199; the AMD launch MSRP is $3299.

The Verdict

The data points to the Intel Xeon 674X as the stronger overall processor. It wins 13 of 16 benchmark comparisons, holds a higher average score (143,103 versus 124,171), and reaches the 98th percentile versus the AMD chip's 97th. The Xeon is ahead by 11.8% across all Cinebench multi-core tests, 18.4% in single-thread, 20.5% in floating point math, and 74% in physics. These are not marginal wins; they are substantial across rendering, simulation, and general compute.

The AMD Ryzen Threadripper PRO 5975WX is not without merit. It has 4 more cores and 8 more threads, and it wins decisively in data encryption (23.9% ahead) and integer math (14.2% ahead). It also edges out the Intel chip in data compression. For workloads that hammer integer operations or encryption, the AMD part is the better choice. However, the overall benchmark picture favors Intel.

The launch MSRP difference is notable: the Intel chip launched at $2199, the AMD at $3299. That means the Intel part is both faster in most tests and carried a lower launch price. The AMD chip's higher launch MSRP combined with fewer benchmark wins makes it harder to recommend for general workstation use, despite its core-count advantage.

Where Each One Wins

The Intel Xeon 674X is the pick for rendering workloads, physics simulation, floating point math, and single-threaded performance. The Cinebench R23 multi-core score of 71,566 versus 64,021 makes it the clear choice for 3D rendering and video encoding. The 74% lead in physics and 20.5% lead in floating point math point to scientific computing and simulation tasks. The 18.4% single-thread advantage means it also handles interactive workloads, compilation, and general desktop use more smoothly.

The AMD Ryzen Threadripper PRO 5975WX wins in encryption, integer math, and data compression. The 23.9% lead in data encryption makes it the stronger option for cryptographic workloads, secure communications, and disk encryption. The 14.2% lead in integer math suits database processing, financial modeling, and code compilation. The 4.4% edge in data compression helps with archival and backup tasks.

The core count difference matters for highly parallel workloads that scale linearly, but the benchmark data shows that the Intel chip's architecture overcomes the AMD core deficit in most multi-threaded tests. The Xeon wins multi-thread by 11.8% despite having 4 fewer cores. That is the headline finding of this comparison: architectural efficiency beats raw core count in the recorded measurements.

For a mixed-use workstation, the Intel Xeon 674X is the safer choice. For encryption-heavy or integer-heavy environments where the AMD wins are concentrated, the Threadripper PRO 5975WX has a specific role. The data is clear: Intel wins the majority of tests, often by double-digit margins, and does so at a lower launch MSRP.