AMD EPYC 9565 vs Intel Xeon w9-3595X Comparison

AMD
AMD

AMD EPYC 9565

CORE STATE Turin
CORE SPECS 72 Cores / 144 Threads
CLOCK SPEED 3.15 Base / 4.3 GHz Turbo
CACHE 384 MB (shared)
MAX TDP 400W
ARCHITECTURE Zen 5
nm
PROCESS 4 nm
LAUNCH DATE 2024
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
11,585
8,497
cinebench_cinebench_r15_singlecore
1,635
1,199
cinebench_cinebench_r20_multicore
48,273
35,407
cinebench_cinebench_r20_singlecore
6,814
4,998
cinebench_cinebench_r23_multicore
114,937
84,304
cinebench_cinebench_r23_singlecore
16,226
N/A
passmark_data_compression
2,579,631
1,831,962
passmark_data_encryption
141,936
92,249
passmark_extended_instructions
209,595
142,785
passmark_find_prime_numbers
2,422
580
passmark_floating_point_math
549,422
379,008
passmark_integer_math
717,948
473,507
passmark_multithread
135,221
99,576
passmark_physics
18,036
5,842
passmark_random_string_sorting
291,941
190,745
passmark_single_thread
3,696
3,720
passmark_singlethread
3,696
3,720

Analysis: AMD EPYC 9565 vs Intel Xeon w9-3595X

Head-to-Head Benchmarks

The benchmark data presents a strikingly one-sided contest. Across the sixteen recorded head-to-head tests, the AMD EPYC 9565 claims fourteen wins, while the Intel Xeon w9-3595X manages just two. The margins are not subtle. In Cinebench R23 multi-core, the EPYC 9565 scores 114,937 against the Xeon's 84,304, a 36.3% advantage. That gap is nearly identical across the entire Cinebench suite: R15 multi-core shows 11,585 versus 8,497 (36.3%), R20 multi-core shows 48,273 versus 35,407 (36.3%), and even single-core tests follow the same pattern. The EPYC 9565 posts 1,635 in Cinebench R15 single-core to the Xeon's 1,199, a 36.4% lead, and 6,814 versus 4,998 in R20 single-core, another 36.3% margin.

The PassMark results widen the divide further in several specialized workloads. The most dramatic gap appears in PassMark find prime numbers, where the EPYC 9565 scores 2,422 versus the Xeon's 580, a 317.6% advantage. PassMark physics tells a similar story: 18,036 versus 5,842, a 208.7% lead. These are not incremental differences; they represent a dominant performance class separation. Data encryption shows a 53.9% edge (141,936 versus 92,249), random string sorting a 53.1% edge (291,941 versus 190,745), and integer math a 51.6% edge (717,948 versus 473,507). Extended instructions land at 46.8% ahead (209,595 versus 142,785), and floating point math at 45% ahead (549,422 versus 379,008). Data compression rounds out the major wins with a 40.8% margin (2,579,631 versus 1,831,962).

The Xeon's only victories come in PassMark single-thread and its duplicate PassMark singlethread entry, both scoring 3,720 against the EPYC's 3,696. That is a 0.6% margin, a narrow but consistent win for Intel in this specific metric. It is importantly the EPYC 9565 wins the Cinebench single-core tests by over 36%, so the PassMark single-thread result appears to measure a different aspect of performance, likely a lighter-weight workload where the Xeon's higher boost clock can assert itself.

Looking at the broader database context, the EPYC 9565 holds an average benchmark score of 285,471, placing it in the 99th percentile of all CPUs. Its nearest rivals include the Intel Xeon 696X at 286,102 (0.2% higher), the AMD EPYC 9555P at 287,066 (0.6% higher), the Intel Xeon 6780E at 280,438 (1.8% lower), and the AMD Ryzen Threadripper 9970X at 279,778 (2% lower). The Xeon w9-3595X, by contrast, averages 209,881, also in the 99th percentile, but its nearest rivals sit notably lower: the AMD EPYC 9455P at 217,854 (3.7% higher), the Intel Xeon 6741P at 194,901 (7.7% lower), the AMD EPYC 9335 at 194,228 (8.1% lower), and the Intel Xeon 678X at 193,477 (8.5% lower). The EPYC 9565's average score is 36% higher than the Xeon w9-3595X's, which aligns almost exactly with the Cinebench deltas observed head-to-head.

Where Each One Wins

The EPYC 9565 dominates in every multi-threaded and compute-heavy category. Rendering workloads, represented by the Cinebench R15, R20, and R23 multi-core tests, all show the same 36.3% advantage. This makes the EPYC 9565 the clear choice for CPU-based rendering, video encoding, and any workload that scales across many cores. The PassMark integer and floating point math scores reinforce this: 51.6% and 45% leads respectively indicate strong performance in scientific computing, financial modeling, and engineering simulations. The 53.9% encryption lead matters for database workloads, VPN gateways, and any application relying on cryptographic operations. Data compression at 40.8% ahead benefits storage servers, backup systems, and large-scale data pipelines.

The two outlier tests deserve special attention. PassMark find prime numbers, where the EPYC leads by 317.6%, suggests a massive advantage in prime-number generation and related number-theoretic tasks. PassMark physics, with a 208.7% lead, points to a similarly large gap in physics simulation workloads, common in game physics, particle simulations, and certain scientific applications. These are not typical office tasks, but for users running these specific workloads, the EPYC 9565 is in a different performance tier entirely.

The Xeon w9-3595X wins only the PassMark single-thread test, and by a razor-thin 0.6%. This means for lightly threaded, short-duration tasks that rely on a single core's peak speed, the Xeon holds a marginal edge. However, the Cinebench single-core results contradict this pattern, with the EPYC winning by 36.4% in R15 and 36.3% in R20. A reasonable interpretation is that the Xeon's 4.80 GHz boost clock gives it an edge in very short, bursty single-thread workloads, while the EPYC's higher sustained single-core performance wins in longer single-threaded renders or calculations. For practical purposes, the Xeon's single-thread win is too narrow to outweigh its substantial losses everywhere else.

Architecture Differences

The two processors come from fundamentally different design philosophies. The AMD EPYC 9565 is built on the Zen 5 architecture, codenamed Turin, and belongs to the EPYC 9005 series. It uses a 4 nm process at TSMC and packages 99,780 million transistors across twelve chiplets, each with a die size of 70.6 mm². The Intel Xeon w9-3595X uses the Sapphire Rapids architecture on Intel's 10 nm process, with four dies of 477 mm² each. The process node difference is significant: 4 nm versus 10 nm gives AMD a substantial density and efficiency advantage.

Core counts diverge sharply. The EPYC 9565 provides 72 cores and 144 threads, while the Xeon w9-3595X offers 60 cores and 120 threads. That is a 20% core advantage for AMD, which directly contributes to its multi-threaded leads. Cache hierarchies also differ. Both processors use 80 KB of L1 cache per core, but the L2 cache differs: the EPYC has 1 MB per core, while the Xeon has 2 MB per core. The L3 cache tells the opposite story: the EPYC 9565 has 384 MB shared L3, while the Xeon w9-3595X has 112.5 MB. The EPYC's L3 cache is more than three times larger, which helps with data-heavy workloads and large working sets.

Memory subsystems show a similar divergence. The EPYC 9565 uses twelve-channel DDR5 memory with a bandwidth of 576.0 GB/s. The Xeon w9-3595X uses eight-channel DDR5 with 307.2 GB/s. The EPYC's memory bandwidth is 87.5% higher, a critical factor for memory-bound workloads. Both support ECC memory, but the EPYC's wider memory bus gives it a clear throughput advantage. PCIe connectivity also favors AMD: the EPYC offers Gen 5 with 128 lanes (CPU only), while the Xeon offers Gen 5 with 112 lanes (CPU only). That is 16 additional lanes for the EPYC, which matters for systems with many GPUs or NVMe drives.

Clock speeds present a mixed picture. The EPYC 9565 has a base clock of 3.15 GHz and a boost clock of 4.30 GHz. The Xeon w9-3595X has a lower base clock of 2.00 GHz but a higher boost clock of 4.80 GHz. The Xeon's higher boost likely explains its narrow PassMark single-thread win, while the EPYC's higher base clock and greater core count drive its multi-threaded dominance. Thermal design power is close: 400 watts for the EPYC versus 385 watts for the Xeon. The Xeon is multiplier-unlocked, while the EPYC is not, indicating overclocking headroom on the Intel part. The Xeon also uses Intel Socket 4677, while the EPYC uses AMD Socket SP5, so platform choice will determine compatibility.

FAQ

Q: Which processor is faster in multi-core Cinebench benchmarks?

A: The AMD EPYC 9565 wins all three Cinebench multi-core tests by a 36.3% margin. In Cinebench R23 multi-core, it scores 114,937 versus the Intel Xeon w9-3595X's 84,304.

Q: Does the Intel Xeon w9-3595X win any benchmark?

A: Yes, it wins the PassMark single-thread test with a score of 3,720 versus the EPYC 9565's 3,696, a 0.6% margin. This is also reflected in the duplicate PassMark singlethread entry.

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

A: The largest gap is in PassMark find prime numbers, where the EPYC 9565 scores 2,422 against the Xeon's 580, a 317.6% advantage. PassMark physics shows the second-largest gap at 208.7%.

Q: How do their average benchmark scores compare?

A: The EPYC 9565 has an average benchmark score of 285,471, while the Xeon w9-3595X averages 209,881. The EPYC's nearest rival is the Intel Xeon 696X at 286,102, while the Xeon's nearest rival is the AMD EPYC 9455P at 217,854.

Q: Which processor has more cores and threads?

A: The AMD EPYC 9565 has 72 cores and 144 threads. The Intel Xeon w9-3595X has 60 cores and 120 threads.

Q: How do their memory bandwidth figures differ?

A: The EPYC 9565 supports twelve-channel DDR5 memory with 576.0 GB/s bandwidth. The Xeon w9-3595X supports eight-channel DDR5 with 307.2 GB/s bandwidth.

Specification Differences

| Specification | AMD EPYC 9565 | Intel Xeon w9-3595X |

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

| Cores | 72 | 60 |

| Threads | 144 | 120 |

| Base Clock | 3.15 GHz | 2.00 GHz |

| Boost Clock | 4.30 GHz | 4.80 GHz |

| TDP | 400 W | 385 W |

| Socket | AMD Socket SP5 | Intel Socket 4677 |

| Architecture | Zen 5 | Sapphire Rapids |

| Codename | Turin | Sapphire Rapids |

| Process Node | 4 nm | 10 nm |

| Foundry | TSMC | Intel |

| Transistors | 99,780 million | Not specified |

| Die Size | 12x 70.6 mm² | 4x 477 mm² |

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

| L3 Cache | 384 MB (shared) | 112.5 MB |

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

| Memory Bandwidth | 576.0 GB/s | 307.2 GB/s |

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

| Launch MSRP | $10486 | $5889 |

| Multiplier Unlocked | No | Yes |

| Part Number | 100-000001447 | SRN71 |

| Release Date | 2024-10-09 | 2024-08-23 |

The specification sheet reinforces the benchmark picture. The EPYC 9565 leads in core count, process node, L3 cache, memory bandwidth, and PCIe lanes. The Xeon w9-3595X counters with a higher boost clock, larger L2 cache per core, an unlocked multiplier, and a lower launch MSRP. For workloads that scale across cores, memory bandwidth, or cache capacity, the EPYC 9565 is the stronger part. For single-threaded burst workloads or overclocking scenarios, the Xeon w9-3595X offers specific advantages. The recorded data, however, shows the EPYC 9565 winning fourteen of sixteen head-to-head tests, often by substantial margins.

DETAILED SPECIFICATIONS

SPECIFICATION
EPYC 9565
w9-3595X
Core Specs
Cores
72
60 -16.7%
Threads
144
120 -16.7%
Base Clock (GHz)
3.15
2 -36.5%
Boost Clock (GHz)
4.3
4.8 +11.6%
Frequency (GHz)
3.15
2 -36.5%
Turbo Clock (GHz)
4.3
4.8 +11.6%
Multiplier
31.5
20 -36.5%
SMP CPUs
2
1 -50.0%
Cache
L1 Cache
80 KB (per core)
80 KB (per core)
L2 Cache
1 MB (per core)
2 MB (per core)
L3 Cache
384 MB (shared)
112.5 MB
Power
TDP (W)
400
385 -3.8%
Configurable TDP
320-400 W
—
Architecture
Architecture
Zen 5
—
Codename
Turin
Sapphire Rapids
Generation
EPYC (Zen 5 (Turin))
Xeon W (Sapphire Rapids)
Process Size
4 nm
10 nm
Transistors
99,780 million
—
Die Size
12x 70.6 mm²
4x 477 mm²
Foundry
TSMC
Intel
Memory
Memory Support
DDR5
DDR5
Memory Bus
Twelve-channel
Eight-channel
Memory Bandwidth
576.0 GB/s
307.2 GB/s
ECC Memory
Yes
Yes
Platform
Socket
AMD Socket SP5
Intel Socket 4677
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
—
Interconnect
CXL
Gen 2.0
—
Other
Market
Server/Workstation
Server/Workstation
Production Status
Active
Active
Launch Price
$10486
$5889
Part Number
100-000001447
SRN71
Package
FC-LGA6096
FC-LGA16A
View EPYC 9565 Details View Xeon w9-3595X Details