AMD Ryzen 9 PRO 9965X3D vs Intel Xeon w9-3575X Comparison

AMD
AMD

AMD Ryzen 9 PRO 9965X3D

CORE STATE Granite Ridge
CORE SPECS 16 Cores / 32 Threads
CLOCK SPEED 4.3 Base / 5.5 GHz Turbo
CACHE 128 MB
MAX TDP 170W
ARCHITECTURE Granite Ridge
nm
PROCESS 4 nm
LAUNCH DATE 2026
VS
Intel
INTEL

Xeon w9-3575X

CORE STATE Sapphire Rapids
CORE SPECS 44 Cores / 88 Threads
CLOCK SPEED 2.2 Base / 4.8 GHz Turbo
CACHE 97.5 MB
MAX TDP 340W
ARCHITECTURE Sapphire Rapids
nm
PROCESS 10 nm
LAUNCH DATE 2024

PERFORMANCE BENCHMARKS

passmark_data_compression
895,563
1,219,584
passmark_data_encryption
43,905
62,258
passmark_extended_instructions
70,797
109,843
passmark_find_prime_numbers
599
687
passmark_floating_point_math
157,123
273,398
passmark_integer_math
237,955
298,224
passmark_multithread
68,225
83,338
passmark_physics
4,801
6,836
passmark_random_string_sorting
92,886
134,723
passmark_single_thread
4,614
3,672
passmark_singlethread
4,614
3,672
cinebench_cinebench_r15_multicore
N/A
7,140
cinebench_cinebench_r15_singlecore
N/A
1,008
cinebench_cinebench_r20_multicore
N/A
29,751
cinebench_cinebench_r20_singlecore
N/A
4,200
cinebench_cinebench_r23_multicore
N/A
70,837

Analysis: AMD Ryzen 9 PRO 9965X3D vs Intel Xeon w9-3575X

The data is unambiguous: the Intel Xeon w9-3575X is the dominant multi-threaded workstation processor, winning 9 of 11 head-to-head benchmark comparisons against the AMD Ryzen 9 PRO 9965X3D. The AMD chip’s only victories come in single-threaded tests, where it leads by a significant 20.4%, but that advantage does little to offset the Intel processor’s massive leads in compute-heavy workloads, which range from 14.7% to 74%. For any workload that scales across cores, the Xeon w9-3575X is the clear choice, while the Ryzen 9 PRO 9965X3D is the pick for latency-sensitive, lightly-threaded tasks.

Where Each One Wins

The benchmark results split cleanly along core-count lines. The Intel Xeon w9-3575X wins every multi-threaded and parallel workload in the comparison. Its most decisive victory is in floating-point math, where it scores 273,398 against the AMD’s 157,123, a 74% advantage. This indicates a massive throughput advantage for scientific computing, engineering simulations, and any workload that relies heavily on FPU performance. Similarly, in extended instructions, the Intel chip leads by 55.2% (109,843 vs. 70,797), which points to a strong advantage in cryptography, media encoding, and specialized compute tasks that leverage AVX-512 or similar instruction sets.

The AMD Ryzen 9 PRO 9965X3D, by contrast, wins exclusively in the single-threaded domain. Its passmark_single_thread score of 4,614 is 20.4% higher than the Intel’s 3,672. This advantage is typical of a higher-clocked, lower-core-count design and matters most for legacy applications, database queries that cannot be parallelized, or interactive workloads where a single core’s response time is the bottleneck. Outside of those two single-thread tests, the AMD processor does not win a single comparison. The Intel chip also leads in data compression by 36.2%, data encryption by 41.8%, find prime numbers by 14.7%, integer math by 25.3%, multithread by 22.2%, physics by 42.4%, and random string sorting by 45%.

Architecture Differences

The two processors are built on fundamentally different design philosophies. The Intel Xeon w9-3575X is a 44-core, 88-thread behemoth based on the Sapphire Rapids architecture, manufactured on a 10 nm process at Intel’s own foundry. It features a massive die size of 4x 477 mm², which reflects its heavy multi-core focus. The AMD Ryzen 9 PRO 9965X3D is a 16-core, 32-thread chip based on the Zen 5 (Granite Ridge) architecture, built on a much denser 4 nm process at TSMC, with a die size of 2x 70.6 mm² and a total transistor count of 16,630 million.

The cache hierarchies also differ dramatically. Both CPUs have 80 KB of L1 cache per core and 2 MB of L2 cache per core, but the Intel chip has a smaller 97.5 MB L3 cache, while the AMD chip packs a massive 128 MB L3 cache. This larger L3 cache on the AMD part is a key feature for its single-thread performance, as it can hold more working data closer to the cores. The Intel chip compensates with sheer core count. Memory support also diverges: the Intel Xeon uses an eight-channel DDR5 memory bus with a theoretical bandwidth of 307.2 GB/s, while the AMD Ryzen uses a dual-channel DDR5 bus with just 89.6 GB/s. The Intel processor also offers 112 PCIe Gen 5 lanes, compared to only 24 on the AMD, making the Xeon vastly superior for multi-GPU or high-density storage configurations. Both support ECC memory and are classified as Server/Workstation parts.

The Verdict

For any user whose primary concern is multi-threaded throughput, the Intel Xeon w9-3575X is the only rational choice from this data. It is ahead in every parallel benchmark, often by substantial margins. Its 44 cores and 88 threads provide a raw execution capacity that the 16-core AMD part cannot match, and its eight-channel memory bus ensures that data can be fed to those cores at a rate 3.4 times faster than the AMD’s dual-channel configuration. The data shows a 22.2% lead in the multithread benchmark, a 36.2% lead in data compression, and a 74% lead in floating-point math. If your workload is rendering, simulation, code compilation, or heavy data processing, the Xeon w9-3575X wins.

The AMD Ryzen 9 PRO 9965X3D is the better pick for single-threaded performance. Its 20.4% lead in the single-thread test is not trivial; it means faster response times in applications that cannot use more than one or two cores. Its higher base clock of 4.30 GHz and boost clock of 5.50 GHz, compared to the Intel’s 2.20 GHz and 4.80 GHz, contribute to this advantage. However, its lower TDP of 170 watts versus 340 watts suggests it will be easier to cool in a standard workstation chassis, though the data does not quantify a performance-per-watt advantage. The verdict is clear: pick the Intel for throughput, pick the AMD for single-thread responsiveness.

FAQ

Q: Which processor is faster in multi-core workloads?

A: The Intel Xeon w9-3575X wins all multi-threaded benchmarks. It leads by 22.2% in passmark_multithread (83,338 vs. 68,225), by 36.2% in data compression, by 41.8% in data encryption, and by 74% in floating-point math.

Q: Does the AMD Ryzen 9 PRO 9965X3D win any benchmark?

A: Yes, the AMD chip wins the passmark_single_thread test with a score of 4,614, which is 20.4% higher than the Intel’s 3,672. It also wins the duplicate passmark_singlethread test with the same margin.

Q: How do the core counts compare?

A: The Intel Xeon w9-3575X has 44 cores and 88 threads, while the AMD Ryzen 9 PRO 9965X3D has 16 cores and 32 threads.

Q: What is the memory bandwidth difference?

A: The Intel Xeon w9-3575X supports eight-channel DDR5 memory with a theoretical bandwidth of 307.2 GB/s. The AMD Ryzen 9 PRO 9965X3D supports dual-channel DDR5 memory with a theoretical bandwidth of 89.6 GB/s.

Q: Which processor has a larger L3 cache?

A: The AMD Ryzen 9 PRO 9965X3D has a 128 MB L3 cache, which is larger than the Intel Xeon w9-3575X’s 97.5 MB L3 cache.

Q: What is the average benchmark score for each?

A: The Intel Xeon w9-3575X has an average benchmark score of 144,323, while the AMD Ryzen 9 PRO 9965X3D has an average score of 143,735, a difference of 0.4% in favor of Intel.

Head-to-Head Benchmarks

The single largest victory for the Intel Xeon w9-3575X is in floating-point math, where it scores 273,398 against the AMD’s 157,123. This 74% delta is the most lopsided result in the comparison and highlights the Xeon’s immense FPU throughput, which is critical for scientific and engineering applications. The second-largest win comes in extended instructions, where the Intel chip scores 109,843 versus 70,797, a 55.2% lead. This suggests a strong advantage in cryptography and media codecs that use advanced instruction sets.

In data compression, the Intel chip scores 1,219,584 against 895,563, a 36.2% lead. For data encryption, the Intel chip scores 62,258 against 43,905, a 41.8% lead. In random string sorting, the Intel chip scores 134,723 versus 92,886, a 45% lead. These results underscore the Xeon’s superiority in data manipulation and I/O-heavy tasks. The physics benchmark shows a 42.4% lead for Intel (6,836 vs. 4,801), which is relevant for simulation and physics engines. In integer math, the Intel chip leads by 25.3% (298,224 vs. 237,955), and in find prime numbers, it leads by 14.7% (687 vs. 599).

The multithread benchmark is a key overall indicator, and the Intel chip wins it with 83,338 versus 68,225, a 22.2% lead. This metric confirms that the Xeon’s 44-core design delivers more aggregate throughput than the AMD’s 16-core design. The only AMD wins are in the single-thread test, where it scores 4,614 versus 3,672, a 20.4% advantage. This means that for any single-threaded task, the AMD chip will feel snappier, but for everything else, the Intel chip is faster.

Specification Differences

The two processors differ in nearly every core specification. The Intel Xeon w9-3575X has 44 cores and 88 threads, while the AMD Ryzen 9 PRO 9965X3D has 16 cores and 32 threads. The Intel base clock is 2.20 GHz with a boost of 4.80 GHz; the AMD base clock is 4.30 GHz with a boost of 5.50 GHz. The Intel TDP is 340 watts, while the AMD TDP is 170 watts. The Intel chip uses the Intel Socket 4677, while the AMD chip uses the AMD Socket AM5. The process nodes differ: Intel uses 10 nm, while AMD uses 4 nm. The Intel chip has a die size of 4x 477 mm², while AMD has a die size of 2x 70.6 mm².

L2 cache per core is 2 MB on the Intel and 1 MB on the AMD, while L3 cache is 97.5 MB on Intel and 128 MB on AMD. The memory bus is eight-channel on Intel and dual-channel on AMD, with bandwidths of 307.2 GB/s and 89.6 GB/s, respectively. PCIe lanes are 112 on Intel and 24 on AMD, both Gen 5. The Intel chip has no integrated graphics, while the AMD chip includes Radeon Graphics. The Intel chip has an unlocked multiplier, while the AMD chip does not. The Intel launch MSRP is $3789; the AMD has no listed launch MSRP.

DETAILED SPECIFICATIONS

SPECIFICATION
9 PRO 9965X3D
w9-3575X
Core Specs
Cores
16
44 +175.0%
Threads
32
88 +175.0%
Base Clock (GHz)
4.3
2.2 -48.8%
Boost Clock (GHz)
5.5
4.8 -12.7%
Frequency (GHz)
4.3
2.2 -48.8%
Turbo Clock (GHz)
5.5
4.8 -12.7%
Multiplier
43
22 -48.8%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
80 KB (per core)
80 KB (per core)
L2 Cache
1 MB (per core)
2 MB (per core)
L3 Cache
128 MB
97.5 MB
Power
TDP (W)
170
340 +100.0%
PPT
230 W
—
Architecture
Codename
Granite Ridge
Sapphire Rapids
Generation
Ryzen 9 (Zen 5 (Granite Ridge))
Xeon W (Sapphire Rapids)
Process Size
4 nm
10 nm
Transistors
16,630 million
—
Die Size
2x 70.6 mm²
4x 477 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR5
Memory Bus
Dual-channel
Eight-channel
Memory Bandwidth
89.6 GB/s
307.2 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket AM5
Intel Socket 4677
Chipsets
X870E, X870, B850, B840, X670E, X670, B650E, B650, A620
—
PCIe
Gen 5, 24 Lanes(CPU only)
Gen 5, 112 Lanes(CPU only)
DMI
—
4.0 x8
AMD Multi-Die
IO Process Size
6 nm
—
Graphics
Integrated Graphics
Radeon Graphics
—
Other
Market
Server/Workstation
Server/Workstation
Production Status
Active
Active
Launch Price
—
$3789
Part Number
100-000001999
SRN72
Package
FC-LGA1718
FC-LGA16A
Tj Max
95°C
—
Bundled Cooler
None
—
View Ryzen 9 PRO 9965X3D Details View Xeon w9-3575X Details