AMD Ryzen Threadripper PRO 3955WX vs Intel Xeon 6505P Comparison

AMD
AMD

AMD Ryzen Threadripper PRO 3955WX

CORE STATE Castle Peak
CORE SPECS 16 Cores / 32 Threads
CLOCK SPEED 3.9 Base / 4.3 GHz Turbo
CACHE 64 MB
MAX TDP 280W
ARCHITECTURE Zen 2
nm
PROCESS 7 nm
LAUNCH DATE 2020
VS
Intel
INTEL

Xeon 6505P

CORE STATE Granite Rapids
CORE SPECS 12 Cores / 24 Threads
CLOCK SPEED 2.2 Base / 4.1 GHz Turbo
CACHE 48 MB (shared)
MAX TDP 150W
ARCHITECTURE Granite Rapids
nm
PROCESS 5 nm
LAUNCH DATE 2025

PERFORMANCE BENCHMARKS

cinebench_cinebench_r15_multicore
3,440
3,294
cinebench_cinebench_r15_singlecore
485
464
cinebench_cinebench_r20_multicore
14,335
13,728
cinebench_cinebench_r20_singlecore
2,023
1,937
cinebench_cinebench_r23_multicore
34,131
32,687
cinebench_cinebench_r23_singlecore
4,818
4,614
geekbench_multicore
12,523
N/A
geekbench_singlecore
1,684
N/A
passmark_data_compression
601,738
480,368
passmark_data_encryption
38,058
24,458
passmark_extended_instructions
38,952
37,515
passmark_find_prime_numbers
198
217
passmark_floating_point_math
77,979
92,992
passmark_integer_math
132,456
120,456
passmark_multithread
40,155
38,456
passmark_physics
2,460
2,991
passmark_random_string_sorting
63,758
52,372
passmark_single_thread
2,679
3,187
passmark_singlethread
2,679
3,187

Analysis: AMD Ryzen Threadripper PRO 3955WX vs Intel Xeon 6505P

Head-to-Head Benchmarks

The recorded data shows a clear split between these two processors, with the AMD Ryzen Threadripper PRO 3955WX taking 12 of the 17 direct comparisons, while the Intel Xeon 6505P wins 5. The margin in the AMD victories is often substantial, but the Intel counterattacks are concentrated in specific workloads where its architecture pulls ahead.

In Cinebench testing, the AMD part is consistently ahead across every version and workload type. The multicore results show 3440 versus 3294 in R15, a 4.4% lead; 14335 versus 13728 in R20, again 4.4%; and 34131 versus 32687 in R23, also 4.4%. Single-core Cinebench results follow the same pattern: R15 at 485 versus 464 (4.5% ahead), R20 at 2023 versus 1937 (4.4%), and R23 at 4818 versus 4614 (4.4%). The consistency of that 4.4% delta across all three Cinebench generations suggests a stable architectural advantage rather than a workload-specific quirk.

The PassMark suite reveals where each chip excels. The AMD processor dominates data compression with 601738 against 480368, a 25.3% advantage. Data encryption is an even larger gap: 38058 versus 24458, a 55.6% margin, the biggest single delta in the entire comparison. Random string sorting also goes heavily to AMD at 63758 versus 52372, a 21.7% lead. Integer math shows a 10% advantage for AMD (132456 versus 120456), and extended instructions favor AMD by 3.8% (38952 versus 37515). The overall PassMark multithread score lands at 40155 for AMD versus 38456 for Intel, a 4.4% edge that mirrors the Cinebench pattern.

The Intel Xeon 6505P wins its five comparisons in more specialized areas. Floating point math is its strongest result: 92992 versus 77979, a 16.1% margin for Intel. Physics simulation also favors Intel at 2991 versus 2460, a 17.8% lead. Prime number finding goes to Intel by 8.8% (217 versus 198), and the single-thread PassMark score is notably higher for Intel: 3187 versus 2679, a 15.9% advantage. The remaining Intel win is the duplicate single-thread entry, also 3187 versus 2679. These results indicate that for scalar single-threaded work and floating-point-heavy tasks, the Intel part is the stronger performer, despite losing the Cinebench single-core tests.

Looking at the broader database context, the AMD processor holds a 92nd percentile ranking across all CPUs with an average benchmark score of 56555. Its nearest rivals in the database are the AMD Ryzen AI 9 HX PRO 470 at 56306 (0.4% behind), the AMD Ryzen AI Max 390 at 56273 (0.5% behind), the Intel Core i9-14900HX at 56004 (1% behind), and the Intel Core 7 253PQE at 55919 (1.1% behind). The Intel Xeon 6505P sits at the 91st percentile with an average score of 53701. Its closest recorded competitors are the Intel Core i7-14700F at 53620 (0.2% behind), the Intel Xeon Phi 7290 at 53469 (0.4% behind), the Intel Core Ultra 5 245K at 54053 (0.7% ahead of the Xeon), and the AMD Ryzen 9 7900X at 53288 (0.8% behind). The overall average benchmark gap between the two reviewed parts is 56555 versus 53701, a difference of roughly 5.3% in favor of AMD.

FAQ

Q: Which processor wins more head-to-head benchmark comparisons?

A: The AMD Ryzen Threadripper PRO 3955WX wins 12 of the 17 recorded comparisons, while the Intel Xeon 6505P wins 5.

Q: Where does the Intel Xeon 6505P outperform the AMD part?

A: The Intel chip wins in PassMark floating point math (92992 versus 77979), physics simulation (2991 versus 2460), prime number finding (217 versus 198), and single-thread PassMark scores (3187 versus 2679).

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

A: The biggest delta is in PassMark data encryption, where the AMD Ryzen Threadripper PRO 3955WX scores 38058 versus 24458 for Intel, a 55.6% advantage.

Q: How do the two processors compare in Cinebench R23?

A: The AMD part scores 34131 in multicore and 4818 in single-core, while the Intel part scores 32687 and 4614 respectively. AMD leads by 4.4% in both.

Q: What are the average benchmark scores for each processor?

A: The AMD Ryzen Threadripper PRO 3955WX has an average benchmark score of 56555, and the Intel Xeon 6505P has an average of 53701.

Q: How do their database percentile rankings compare?

A: The AMD processor is in the 92nd percentile of all CPUs, while the Intel Xeon 6505P is in the 91st percentile.

Architecture Differences

The two processors come from fundamentally different design lineages. The AMD Ryzen Threadripper PRO 3955WX is built on the Zen 2 architecture, codenamed Castle Peak, from the Ryzen Threadripper 3000 series. It uses a 7 nm process node fabricated by TSMC. The chip contains 7,600 million transistors across a die size of 2x 74 mm². The Intel Xeon 6505P uses the Granite Rapids architecture, part of the Xeon 6 generation, on a 5 nm process node fabricated by Intel. The database does not record transistor count or die size for the Intel part.

Core and thread counts differ significantly. The AMD processor has 16 cores and 32 threads, while the Intel processor has 12 cores and 24 threads. Clock speeds also favor AMD: the base clock is 3.90 GHz versus 2.20 GHz, and the boost clock is 4.30 GHz versus 4.10 GHz. The AMD chip has a TDP of 280 watts, while the Intel chip has a TDP of 150 watts.

Cache layouts are structured differently. The AMD part has 64 KB of L1 cache per core, 512 KB of L2 cache per core, and 64 MB of L3 cache. The Intel part has 112 KB of L1 per core, 2 MB of L2 per core, and 48 MB of shared L3 cache. The larger per-core L2 on Intel could contribute to its single-thread PassMark advantage, while the larger total L3 on AMD may help with multi-threaded throughput.

Memory support also differs. The AMD processor uses DDR4 memory with an eight-channel bus and a recorded bandwidth of 204.8 GB/s. The Intel processor uses DDR5 memory with an eight-channel bus and a recorded bandwidth of 409.6 GB/s, double the AMD figure. Both support ECC memory. PCIe connectivity differs as well: AMD provides Gen 4 with 128 lanes (CPU only), while Intel provides Gen 5 with 88 lanes (CPU only). The Intel part has no integrated graphics, listed as N/A, and the AMD part also has no integrated graphics listed.

The sockets are incompatible: the AMD chip uses AMD Socket WRX8, while the Intel chip uses Intel Socket 4710. The AMD processor was released on 2020-07-13, and the Intel processor was released on 2025-02-23. Neither processor has an unlocked multiplier. The AMD part number is 100-000000167, and the Intel part number is SRVU7. The AMD launch MSRP is $1149, and the Intel launch MSRP is $563.

The Verdict

The data supports a straightforward choice based on workload profile. For multi-threaded productivity, content creation, and data-heavy tasks, the AMD Ryzen Threadripper PRO 3955WX is the stronger part. It leads in every Cinebench test, all by roughly 4.4%, and its wins in data compression (25.3%), data encryption (55.6%), and random string sorting (21.7%) are decisive. The 16-core, 32-thread configuration with 64 MB of L3 cache gives it a clear edge in parallel workloads that can use all available threads.

For users prioritizing single-thread response or floating-point math, the Intel Xeon 6505P is the better choice. Its PassMark single-thread score of 3187 is 15.9% higher than the AMD score of 2679, and it wins floating point math by 16.1% and physics by 17.8%. The higher per-core boost clock behavior in those specific tests, combined with the larger per-core L2 cache, appears to translate into tangible wins in those measured scenarios.

The average benchmark gap of 56555 to 53701, with AMD at the 92nd percentile and Intel at the 91st, reinforces the overall AMD advantage. The Intel part does sit closer to its own nearest rivals, within 0.2% to 0.8% of the AMD Ryzen 9 7900X and Intel Core Ultra 5 245K cluster, whereas the AMD part sits in a slightly higher performance tier next to recent mobile HX and high-end desktop parts. The data does not indicate that the Intel chip closes the gap to the AMD chip in overall terms, but it does offer a different balance of strengths and weaknesses that may suit specific applications differently.

Specification Differences

| Field | AMD Ryzen Threadripper PRO 3955WX | Intel Xeon 6505P |

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

| Cores | 16 | 12 |

| Threads | 32 | 24 |

| Base Clock | 3.90 GHz | 2.20 GHz |

| Boost Clock | 4.30 GHz | 4.10 GHz |

| TDP | 280 W | 150 W |

| Socket | AMD Socket WRX8 | Intel Socket 4710 |

| Architecture | Zen 2 | Granite Rapids |

| Codename | Castle Peak | Granite Rapids |

| Generation | Ryzen Threadripper (Zen 2 (Castle Peak)) | Xeon 6 (Granite Rapids-SP) |

| Process Node | 7 nm | 5 nm |

| Foundry | TSMC | Intel |

| Transistors | 7,600 million | Not recorded |

| Die Size | 2x 74 mm² | Not recorded |

| L1 Cache | 64 KB (per core) | 112 KB (per core) |

| L2 Cache | 512 KB (per core) | 2 MB (per core) |

| L3 Cache | 64 MB | 48 MB (shared) |

| Memory Support | DDR4 | DDR5 |

| Memory Bus | Eight-channel | Eight-channel |

| Memory Bandwidth | 204.8 GB/s | 409.6 GB/s |

| PCIe | Gen 4, 128 Lanes (CPU only) | Gen 5, 88 Lanes (CPU only) |

| Integrated Graphics | None listed | N/A |

| Market Segment | Desktop | Server/Workstation |

| Release Date | 2020-07-13 | 2025-02-23 |

| Launch MSRP | $1149 | $563 |

| Part Number | 100-000000167 | SRVU7 |

DETAILED SPECIFICATIONS

SPECIFICATION
Threadripper PRO 3955WX
6505P
Core Specs
Cores
16
12 -25.0%
Threads
32
24 -25.0%
Base Clock (GHz)
3.9
2.2 -43.6%
Boost Clock (GHz)
4.3
4.1 -4.7%
Frequency (GHz)
3.9
2.2 -43.6%
Turbo Clock (GHz)
4.3
4.1 -4.7%
Multiplier
39
22 -43.6%
SMP CPUs
1
2 +100.0%
Cache
L1 Cache
64 KB (per core)
112 KB (per core)
L2 Cache
512 KB (per core)
2 MB (per core)
L3 Cache
64 MB
48 MB (shared)
Power
TDP (W)
280
150 -46.4%
Architecture
Architecture
Zen 2
Granite Rapids
Codename
Castle Peak
Granite Rapids
Generation
Ryzen Threadripper (Zen 2 (Castle Peak))
Xeon 6 (Granite Rapids-SP)
Process Size
7 nm
5 nm
Transistors
7,600 million
Die Size
2x 74 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR4
DDR5
Memory Bus
Eight-channel
Eight-channel
Memory Bandwidth
204.8 GB/s
409.6 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket WRX8
Intel Socket 4710
PCIe
Gen 4, 128 Lanes(CPU only)
Gen 5, 88 Lanes(CPU only)
AMD Multi-Die
IO Process Size
14 nm
10 nm
Interconnect
UPI Links
3 x24 24 GT/s
CXL
Gen 2.0, 64 Lanes (Shared with PCI-E)
Other
Market
Desktop
Server/Workstation
Production Status
Active
Active
Launch Price
$1149
$563
Part Number
100-000000167
SRVU7
Package
sWRX8
FC-LGA18N
Tj Max
97°C
Bundled Cooler
None
View Ryzen Threadripper PRO 3955WX Details View Xeon 6505P Details