AMD Ryzen Threadripper PRO 9975WX vs Intel Xeon w9-3595X Comparison

AMD
AMD

AMD Ryzen Threadripper PRO 9975WX

CORE STATE Shimada Peak
CORE SPECS 32 Cores / 64 Threads
CLOCK SPEED 4 Base / 5.4 GHz Turbo
CACHE 128 MB
MAX TDP 350W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
Intel
INTEL

Xeon w9-3595X

CORE STATE Sapphire Rapids
CORE SPECS 60 Cores / 120 Threads
CLOCK SPEED 2 Base / 4.8 GHz Turbo
CACHE 112.5 MB
MAX TDP 385W
ARCHITECTURE Sapphire Rapids
nm
PROCESS 10 nm
LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
9,430
8,497
cinebench_cinebench_r15_singlecore
1,331
1,199
cinebench_cinebench_r20_multicore
39,292
35,407
cinebench_cinebench_r20_singlecore
5,546
4,998
cinebench_cinebench_r23_multicore
93,553
84,304
cinebench_cinebench_r23_singlecore
13,207
N/A
passmark_data_compression
1,644,573
1,831,962
passmark_data_encryption
85,035
92,249
passmark_extended_instructions
137,733
142,785
passmark_find_prime_numbers
620
580
passmark_floating_point_math
304,833
379,008
passmark_integer_math
461,724
473,507
passmark_multithread
104,902
99,576
passmark_physics
7,288
5,842
passmark_random_string_sorting
188,014
190,745
passmark_single_thread
4,408
3,720
passmark_singlethread
4,408
3,720

Analysis: AMD Ryzen Threadripper PRO 9975WX vs Intel Xeon w9-3595X

The Intel Xeon w9-3595X and AMD Ryzen Threadripper PRO 9975WX represent two opposing philosophies in the workstation CPU arena. The Intel part leverages a massive 60-core count to brute-force its way through heavy workloads, while the AMD chip uses a more modern, efficient Zen 5 architecture to achieve superior performance with fewer resources. Benchmark data reveals a clear split: AMD dominates in rendering and single-threaded tasks, while Intel holds the edge in specific math and data-processing workloads.

Head-to-Head Benchmarks

The most striking pattern in the head-to-head results is AMD’s sweeping victory across all Cinebench iterations. In Cinebench R15, R20, and R23, the Ryzen Threadripper PRO 9975WX wins both multi-core and single-core tests by a consistent -9.9% deltaPct relative to the Xeon. The R23 multi-core score of 93,553 for AMD versus 84,304 for Intel is a decisive 11% advantage, despite the Xeon packing nearly twice the cores. This suggests the Zen 5 architecture’s per-core efficiency is transformative. The single-core gap is even more pronounced in the Passmark suite, where AMD posts 4,408 points against Intel’s 3,720, a -15.6% deltaPct that shows a massive 18.5% raw performance lead.

AMD also wins the Passmark multithread test, scoring 104,902 against Intel’s 99,576 (a -5.1% deltaPct). This is another surprising result given the core disparity. The physics test is a blowout for AMD: 7,288 versus 5,842, a -19.8% deltaPct, demonstrating superior floating-point scheduling and cache handling. Even the find-prime-numbers test, typically a memory-latency-sensitive benchmark, goes to AMD at 620 versus 580 (-6.5%).

Intel’s wins are concentrated in specialized workloads. The largest is floating-point math, where the Xeon scores 379,008 against AMD’s 304,833, a commanding 24.3% advantage. This indicates that the Xeon’s AVX-512 and wider execution resources remain potent for scientific computing. Data compression is another Intel stronghold: 1,831,962 versus 1,644,573 (11.4%), showing superior memory bandwidth utilization in streaming workloads. Encryption also favors Intel at 92,249 versus 85,035 (8.5%), likely due to dedicated AES-NI throughput. The remaining Intel wins are narrower: extended instructions (142,785 vs 137,733, 3.7%), integer math (473,507 vs 461,724, 2.6%), and random string sorting (190,745 vs 188,014, 1.5%).

The overall win count is 6 for Intel and 10 for AMD. However, the magnitudes matter more than the tally. AMD’s losses are mostly under 9%, while its wins in physics and single-thread are 19.8% and 15.6% respectively. Intel’s one outsized win (24.3% in floating-point) is offset by AMD’s consistency across the board.

Architecture Differences

The two processors are built on fundamentally different silicon. Intel’s Xeon w9-3595X uses the Sapphire Rapids architecture on a 10 nm process from Intel’s own foundries, with a massive die configuration of 4x 477 mm². AMD’s Ryzen Threadripper PRO 9975WX uses Zen 5, codenamed Shimada Peak, fabricated by TSMC on a 4 nm node, with a smaller 4x 70.6 mm² die setup and 33,260 million transistors.

Core counts diverge sharply: Intel fields 60 cores / 120 threads, while AMD uses 32 cores / 64 threads. Clock speeds tell the rest of the story. AMD’s base clock is 4.00 GHz and boost reaches 5.40 GHz, whereas Intel’s base is just 2.00 GHz with a 4.80 GHz boost. This explains the per-core benchmark disparity — AMD’s higher clocks and newer architecture deliver more work per cycle.

Cache hierarchies differ in structure. Intel allocates 80 KB L1 and 2 MB L2 per core, with a shared 112.5 MB L3. AMD uses 64 KB L1 and 1 MB L2 per core, but a larger 128 MB L3 pool. The larger L3 on AMD helps mitigate the core-count deficit in memory-heavy workloads.

Memory bandwidth is a major differentiator: AMD’s eight-channel DDR5 interface delivers 409.6 GB/s, versus Intel’s 307.2 GB/s on the same eight-channel setup. This 33% bandwidth advantage likely drives AMD’s wins in multithreaded and physics tests. PCIe connectivity also favors AMD with 128 lanes (Gen 5, CPU only) versus Intel’s 112 lanes, making the Threadripper more expandable for multi-GPU configurations.

Power envelopes are close but distinct: Intel’s TDP is 385 W, exceeding AMD’s 350 W. Both support ECC memory and have unlocked multipliers, but the efficiency gap is clear — AMD delivers higher performance at a lower TDP. Release timing also favors AMD’s newer design: Intel launched on 2024-08-23, while AMD arrived nearly a year later on 2025-07-22.

FAQ

Q: Which processor has more cores and threads?

A: The Intel Xeon w9-3595X has 60 cores and 120 threads, while the AMD Ryzen Threadripper PRO 9975WX has 32 cores and 64 threads.

Q: Why does the 32-core AMD chip beat the 60-core Intel chip in Cinebench R23 multi-core?

A: The AMD chip scores 93,553 versus Intel’s 84,304. This is due to AMD’s higher boost clock of 5.40 GHz (versus 4.80 GHz) and its 4 nm process node, which allows more instructions per clock than Intel’s 10 nm Sapphire Rapids design.

Q: In which benchmark does Intel have its largest lead?

A: Intel wins Passmark floating-point math by 24.3%, scoring 379,008 against AMD’s 304,833. This indicates superior SIMD and floating-point execution resources.

Q: What memory bandwidth does each processor support?

A: AMD provides 409.6 GB/s over eight-channel DDR5, while Intel provides 307.2 GB/s over the same eight-channel configuration. AMD’s bandwidth is roughly 33% higher.

Q: Which processor offers more PCIe lanes?

A: The AMD Ryzen Threadripper PRO 9975WX has 128 lanes of Gen 5 PCIe (CPU only), compared to Intel’s 112 lanes of Gen 5.

Q: Are both processors unlocked for overclocking?

A: Yes, both the Intel Xeon w9-3595X and the AMD Ryzen Threadripper PRO 9975WX have multiplier unlocked status.

Specification Differences

| Specification | Intel Xeon w9-3595X | AMD Ryzen Threadripper PRO 9975WX |

|---|---|---|

| Cores | 60 | 32 |

| Threads | 120 | 64 |

| Base Clock | 2.00 GHz | 4.00 GHz |

| Boost Clock | 4.80 GHz | 5.40 GHz |

| TDP | 385 W | 350 W |

| Process Node | 10 nm | 4 nm |

| Foundry | Intel | TSMC |

| Die Size | 4x 477 mm² | 4x 70.6 mm² |

| Transistors | Not specified | 33,260 million |

| L1 Cache | 80 KB (per core) | 64 KB (per core) |

| L2 Cache | 2 MB (per core) | 1 MB (per core) |

| L3 Cache | 112.5 MB | 128 MB |

| Memory Bandwidth | 307.2 GB/s | 409.6 GB/s |

| PCIe Lanes | 112 (Gen 5, CPU only) | 128 (Gen 5, CPU only) |

| Release Date | 2024-08-23 | 2025-07-22 |

| Launch MSRP | $5889 | $4099 |

| Codename | Sapphire Rapids | Shimada Peak |

The Verdict

The data points to a clear choice for most users: the AMD Ryzen Threadripper PRO 9975WX is the superior processor. It wins 10 of 16 benchmarks, including all Cinebench tests, multithread, physics, and single-thread performance. Its -15.6% deltaPct in single-thread and -19.8% in physics are decisive margins that will be felt in everyday use and simulation workloads. The 409.6 GB/s memory bandwidth and 128 PCIe lanes make it more capable for data-intensive and multi-GPU workstation builds, all while consuming 35 W less power than the Intel part.

The Intel Xeon w9-3595X is not without merit. Its 24.3% lead in floating-point math and 11.4% edge in data compression make it the pick for scientific computing and specific enterprise workloads that leverage those instructions. The 112.5 MB L3 cache and 60 cores provide raw throughput for embarrassingly parallel tasks that scale linearly. However, its lower clock speeds and older 10 nm process place it behind in nearly every interactive or latency-sensitive benchmark.

For content creators, engineers running CAD, or developers compiling code, the Threadripper PRO 9975WX is the data-backed recommendation. Its combination of higher clocks, newer architecture, and greater bandwidth delivers wins where it matters most — rendering, physics, and single-thread responsiveness. The Intel part remains relevant only for users whose software is heavily optimized for floating-point math or data compression, where its 379,008 and 1,831,962 scores respectively remain class-leading. AMD’s launch MSRP of $4099 versus Intel’s $5889 further underscores the value proposition, though performance per dollar was not explicitly benchmarked. The verdict is unambiguous: AMD’s Zen 5 design out-executes Intel’s core-count strategy in the vast majority of real-world workloads.

DETAILED SPECIFICATIONS

SPECIFICATION
Threadripper PRO 9975WX
w9-3595X
Core Specs
Cores
32
60 +87.5%
Threads
64
120 +87.5%
Base Clock (GHz)
4
2 -50.0%
Boost Clock (GHz)
5.4
4.8 -11.1%
Frequency (GHz)
4
2 -50.0%
Turbo Clock (GHz)
5.4
4.8 -11.1%
Multiplier
40
20 -50.0%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
64 KB (per core)
80 KB (per core)
L2 Cache
1 MB (per core)
2 MB (per core)
L3 Cache
128 MB
112.5 MB
Power
TDP (W)
350
385 +10.0%
Architecture
Architecture
Zen 5
—
Codename
Shimada Peak
Sapphire Rapids
Generation
Ryzen Threadripper (Zen 5 (Shimada Peak))
Xeon W (Sapphire Rapids)
Process Size
4 nm
10 nm
Transistors
33,260 million
—
Die Size
4x 70.6 mm²
4x 477 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR5
Memory Bus
Eight-channel
Eight-channel
Memory Bandwidth
409.6 GB/s
307.2 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket sTR5
Intel Socket 4677
Chipsets
WRX90, TRX50, Pro 695
—
PCIe
Gen 5, 128 Lanes(CPU only)
Gen 5, 112 Lanes(CPU only)
DMI
—
4.0 x8
AMD Multi-Die
IO Process Size
6 nm
—
Other
Market
Server/Workstation
Server/Workstation
Production Status
Active
Active
Launch Price
$4099
$5889
Part Number
100-000000723
SRN71
Package
FC-LGA4844
FC-LGA16A
Tj Max
95°C
—
Bundled Cooler
None
—
View Ryzen Threadripper PRO 9975WX Details View Xeon w9-3595X Details