AMD Ryzen Threadripper PRO 9955WX vs Intel Xeon 6731P Comparison

AMD
AMD

AMD Ryzen Threadripper PRO 9955WX

CORE STATE Shimada Peak
CORE SPECS 16 Cores / 32 Threads
CLOCK SPEED 4.5 Base / 5.4 GHz Turbo
CACHE 64 MB
MAX TDP 350W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2025
VS
Intel
INTEL

Xeon 6731P

CORE STATE Granite Rapids
CORE SPECS 32 Cores / 64 Threads
CLOCK SPEED 2.5 Base / 4.1 GHz Turbo
CACHE 144 MB (shared)
MAX TDP 245W
ARCHITECTURE Granite Rapids
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
5,996
4,522
cinebench_cinebench_r15_singlecore
846
638
cinebench_cinebench_r20_multicore
24,987
18,845
cinebench_cinebench_r20_singlecore
3,527
2,660
cinebench_cinebench_r23_multicore
59,494
44,871
cinebench_cinebench_r23_singlecore
8,399
6,334
passmark_data_compression
920,954
799,474
passmark_data_encryption
44,389
40,087
passmark_extended_instructions
76,363
65,656
passmark_find_prime_numbers
337
541
passmark_floating_point_math
156,215
157,330
passmark_integer_math
236,120
198,761
passmark_multithread
67,035
52,790
passmark_physics
4,156
7,105
passmark_random_string_sorting
99,813
88,019
passmark_single_thread
4,530
2,107
passmark_singlethread
4,530
2,107

Analysis: AMD Ryzen Threadripper PRO 9955WX vs Intel Xeon 6731P

Where Each One Wins

The AMD Ryzen Threadripper PRO 9955WX dominates the overall comparison, winning 14 of 17 benchmark tests. Its strength is concentrated in single-threaded performance and multi-threaded rendering workloads. The Threadripper's lead is especially pronounced in Cinebench tests, where it consistently outperforms the Intel Xeon 6731P by 32.6% across every single-core and multi-core iteration (R15, R20, and R23). The PassMark single-thread score shows an even larger gap, with the AMD part scoring 4,530 versus 2,107 for Intel, a 115% advantage.

The Intel Xeon 6731P wins only three tests, but those wins are meaningful for specific server workloads. It takes the PassMark prime number test with a 37.7% lead, the physics test with a 41.5% advantage, and edges out a slim 0.7% win in floating-point math. These results indicate the Xeon is better suited for compute patterns that rely on massive parallel throughput in integer-heavy or physics-simulation workloads, rather than latency-sensitive single-thread tasks.

For desktop workstation users, the Threadripper is the clear choice. It wins all Cinebench tests, all PassMark multithread tests, and all single-thread tests. The Xeon's wins are narrow in floating-point math, and its prime number and physics wins point to specific HPC or scientific computing scenarios. If the workload involves rendering, video encoding, data compression, or general productivity, the AMD part wins nearly everywhere. If the workload is dominated by prime-number sieving or physics simulation, the Intel part has a distinct edge.

FAQ

Q: Which CPU is faster in Cinebench R23 multi-core?

A: The AMD Ryzen Threadripper PRO 9955WX scores 59,494 versus 44,871 for the Intel Xeon 6731P, a 32.6% advantage for AMD.

Q: Does the Intel Xeon 6731P win any benchmark tests?

A: Yes, it wins three tests: PassMark find prime numbers (541 vs 337), PassMark physics (7,105 vs 4,156), and PassMark floating-point math (157,330 vs 156,215).

Q: How much faster is the AMD part in single-threaded PassMark?

A: The Threadripper scores 4,530 while the Xeon scores 2,107, giving AMD a 115% advantage in that test.

Q: What are the core and thread counts?

A: The AMD Ryzen Threadripper PRO 9955WX has 16 cores and 32 threads. The Intel Xeon 6731P has 32 cores and 64 threads.

Q: Which CPU has a higher boost clock?

A: The AMD part boosts to 5.40 GHz, while the Intel part boosts to 4.10 GHz. The AMD base clock is 4.50 GHz versus 2.50 GHz for Intel.

Q: Do both support DDR5 memory?

A: Yes, both support DDR5 with eight-channel memory buses and 409.6 GB/s memory bandwidth. Both also support ECC memory.

Head-to-Head Benchmarks

The biggest win for AMD is in PassMark single-thread, where the Threadripper PRO 9955WX scores 4,530 versus 2,107 for the Xeon 6731P. That 115% delta is the largest single gap in the entire comparison. This is not just a clock-speed story, though the AMD base clock of 4.50 GHz versus 2.50 GHz for Intel is significant. The single-thread results carry over to every Cinebench single-core test, where AMD wins by exactly 32.6% in R15, R20, and R23.

In multi-core Cinebench, the AMD part maintains that same 32.6% lead across all three versions. The R23 multi-core score of 59,494 versus 44,871 is particularly telling because it represents sustained all-core rendering performance. The Xeon has double the cores (32 vs 16) and double the threads (64 vs 32), yet it still loses by nearly a third. This indicates the AMD Zen 5 architecture is far more efficient per core in these workloads.

PassMark multithread shows a 27% win for AMD (67,035 vs 52,790). Data compression goes to AMD by 15.2% (920,954 vs 799,474). Integer math favors AMD by 18.8% (236,120 vs 198,761). Extended instructions go AMD's way by 16.3% (76,363 vs 65,656). Random string sorting is 13.4% faster on AMD (99,813 vs 88,019). Data encryption is 10.7% faster on AMD (44,389 vs 40,087).

The Intel wins are narrower in absolute terms but still notable. The prime number test shows 541 for Intel versus 337 for AMD, a 37.7% lead. Physics simulation gives Intel 7,105 versus 4,156, a 41.5% advantage. Floating-point math is nearly a tie, with Intel at 157,330 versus AMD at 156,215, a 0.7% margin. These results suggest the Xeon's higher core count pays off in specific parallel compute patterns, even though it loses in most general-purpose tests.

Specification Differences

The core and thread counts differ substantially. AMD offers 16 cores and 32 threads, while Intel offers 32 cores and 64 threads. Clock speeds favor AMD: base 4.50 GHz versus 2.50 GHz, boost 5.40 GHz versus 4.10 GHz. The TDP values also differ, with AMD rated at 350 watts and Intel at 245 watts.

Socket compatibility is split: AMD uses Socket sTR5, Intel uses Socket 4710. The AMD part has an unlocked multiplier, while the Intel part is locked. PCIe lane counts differ, with AMD providing Gen 5 with 128 lanes (CPU only) and Intel providing Gen 5 with 136 lanes (CPU only). The launch MSRP for AMD is $1649, and the launch MSRP for Intel is $2700.

Release dates differ as well. The AMD part launched on 2025-07-22, while the Intel part launched earlier on 2025-02-23. Market segment classification is also different: AMD is listed as Desktop, while Intel is Server/Workstation.

Architecture Differences

The AMD Ryzen Threadripper PRO 9955WX uses the Zen 5 architecture with the Shimada Peak codename, built on a 4 nm process at TSMC. It has 16,630 million transistors across a die size of 2x 70.6 mm². Cache layout includes 64 KB L1 per core, 1 MB L2 per core, and 64 MB L3 total.

The Intel Xeon 6731P uses Granite Rapids architecture (also the codename), built on a 5 nm process at Intel. Its die size is 598 mm², and it has no listed transistor count. Cache layout includes 112 KB L1 per core, 2 MB L2 per core, and 144 MB shared L3. Notably, the Intel part has more than double the L3 cache of the AMD part (144 MB versus 64 MB).

Both support DDR5 memory with eight-channel buses and 409.6 GB/s bandwidth, and both support ECC memory. Neither has integrated graphics. The AMD part is listed as part of the 9000 series and Ryzen Threadripper generation, while the Intel part is part of the Xeon 6 generation (Granite Rapids-SP). The AMD part has an unlocked multiplier, which is unusual for a workstation processor, while the Intel part is locked.

The Verdict

The data clearly favors the AMD Ryzen Threadripper PRO 9955WX for most workloads. Its 32.6% lead across every Cinebench test, 115% lead in PassMark single-thread, and 27% lead in PassMark multithread make it the better choice for rendering, content creation, and general productivity. The fact that it wins 14 of 17 benchmarks, despite having half the cores and threads of the Intel part, shows that per-core performance is far more important for those tasks than raw core count.

The Intel Xeon 6731P makes sense for specific compute patterns. Its 41.5% win in physics simulation and 37.7% win in prime number finding indicate strengths in parallel integer and physics workloads. The slim 0.7% floating-point win is negligible in practice. For users running simulation software, scientific computing, or prime-number-heavy algorithms, the Xeon's higher core count (32 cores, 64 threads) and larger L3 cache (144 MB versus 64 MB) can be leveraged effectively.

For a desktop workstation, the AMD part is the practical recommendation. It has a lower launch MSRP ($1649 versus $2700), an unlocked multiplier for tuning, and it wins the majority of benchmarks. The Intel part is competitive only in niche server workloads where its 245-watt TDP and higher core count matter more than raw clock speed. Choose AMD for most tasks, choose Intel if the workload specifically benefits from physics simulation or prime-number processing.

The database percentile rankings reflect this split: AMD sits at the 97th percentile among all CPUs, while Intel sits at the 96th. Both are excellent processors, but the AMD part is more versatile, more efficient per clock, and more consistent across diverse benchmarks. The Intel part is a specialist, not a generalist, and that distinction matters for most buyers.

DETAILED SPECIFICATIONS

SPECIFICATION
Threadripper PRO 9955WX
6731P
Core Specs
Cores
16
32 +100.0%
Threads
32
64 +100.0%
Base Clock (GHz)
4.5
2.5 -44.4%
Boost Clock (GHz)
5.4
4.1 -24.1%
Frequency (GHz)
4.5
2.5 -44.4%
Turbo Clock (GHz)
5.4
4.1 -24.1%
Multiplier
45
25 -44.4%
SMP CPUs
1
1 0.0%
Cache
L1 Cache
64 KB (per core)
112 KB (per core)
L2 Cache
1 MB (per core)
2 MB (per core)
L3 Cache
64 MB
144 MB (shared)
Power
TDP (W)
350
245 -30.0%
Architecture
Architecture
Zen 5
Granite Rapids
Codename
Shimada Peak
Granite Rapids
Generation
Ryzen Threadripper (Zen 4 (Storm Peak))
Xeon 6 (Granite Rapids-SP)
Process Size
4 nm
5 nm
Transistors
16,630 million
Die Size
2x 70.6 mm²
598 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR5
Memory Bus
Eight-channel
Eight-channel
Memory Bandwidth
409.6 GB/s
409.6 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket sTR5
Intel Socket 4710
Chipsets
WRX90, TRX50, Pro 695
PCIe
Gen 5, 128 Lanes(CPU only)
Gen 5, 136 Lanes(CPU only)
AMD Multi-Die
IO Process Size
6 nm
10 nm
Interconnect
CXL
Gen 2.0, 64 Lanes (Shared with PCI-E)
Other
Market
Desktop
Server/Workstation
Production Status
Active
Active
Launch Price
$1649
$2700
Part Number
100-000000725
SRVNR
Package
FC-LGA4844
FC-LGA18N
Tj Max
95°C
102°C
Bundled Cooler
None
None
View Ryzen Threadripper PRO 9955WX Details View Xeon 6731P Details