AMD EPYC 9654 vs Intel Xeon Gold 6146

Comparative analysis of AMD EPYC 9654 and Intel Xeon Gold 6146 processors for all known characteristics in the following categories: Essentials, Performance, Memory, Compatibility, Peripherals, Security & Reliability, Advanced Technologies, Virtualization. Benchmark processor performance analysis: PassMark - Single thread mark, PassMark - CPU mark, Geekbench 4 - Single Core, Geekbench 4 - Multi-Core.

AMD EPYC 9654

Intel Xeon Gold 6146

Differences

Reasons to consider the AMD EPYC 9654

84 more cores, run more applications at once: 96 vs 12
168 more threads: 192 vs 24
A newer manufacturing process allows for a more powerful, yet cooler running processor: 5 nm vs 14 nm
Around 20% better performance in PassMark - Single thread mark: 2792 vs 2333
3.9x better performance in PassMark - CPU mark: 147784 vs 38055

Specifications (specs)
Number of cores	96 vs 12
Number of threads	192 vs 24
Manufacturing process technology	5 nm vs 14 nm
Benchmarks
PassMark - Single thread mark	2792 vs 2333
PassMark - CPU mark	147784 vs 38055

Reasons to consider the Intel Xeon Gold 6146

Around 1954% higher maximum core temperature: 76°C vs 3.7 GHz
2.2x lower typical power consumption: 165 Watt vs 360 Watt

Maximum core temperature	76°C vs 3.7 GHz
Thermal Design Power (TDP)	165 Watt vs 360 Watt

Compare benchmarks

CPU 1: AMD EPYC 9654
CPU 2: Intel Xeon Gold 6146

PassMark - Single thread mark

CPU 1

CPU 2

2792

2333

PassMark - CPU mark

CPU 1

CPU 2

147784

38055

Name	AMD EPYC 9654	Intel Xeon Gold 6146
PassMark - Single thread mark	2792	2333
PassMark - CPU mark	147784	38055
Geekbench 4 - Single Core		4964
Geekbench 4 - Multi-Core		35971

Compare specifications (specs)

	AMD EPYC 9654	Intel Xeon Gold 6146
Essentials
Architecture codename	Zen 4	Skylake
Launch date	10 Nov 2022	Q3'17
Launch price (MSRP)	$11805
OPN Tray	100-100000789
Place in performance rating	12	26
Processor Number		6146
Series		Intel® Xeon® Scalable Processors
Status		Launched
Vertical segment		Server
Performance
Base frequency	2.4 GHz	3.20 GHz
Die size	72 mm²
L1 cache	6 MB
L2 cache	96 MB
L3 cache	384 MB
Manufacturing process technology	5 nm	14 nm
Maximum core temperature	3.7 GHz	76°C
Number of cores	96	12
Number of threads	192	24
Transistor count	78840 million
Unlocked
Maximum frequency		4.20 GHz
Number of Ultra Path Interconnect (UPI) Links		3
Memory
ECC memory support
Max memory channels	12	6
Supported memory types	DDR5-4800	DDR4-2666
Maximum memory size		768 GB
Supported memory frequency		2666 MHz
Compatibility
Configurable TDP	320-400 Watt
Max number of CPUs in a configuration	2
Package Size	FC-LGA6096	76.0mm x 56.5mm
Sockets supported	SP5	FCLGA3647
Thermal Design Power (TDP)	360 Watt	165 Watt
Low Halogen Options Available
Peripherals
Max number of PCIe lanes	128	48
PCI Express revision	5.0	3.0
Scalability		S4S
Security & Reliability
Execute Disable Bit (EDB)
Intel® Run Sure Technology
Intel® Trusted Execution technology (TXT)
Mode-based Execute Control (MBE)
Advanced Technologies
Enhanced Intel SpeedStep® technology
Instruction set extensions		Intel® SSE4.2, Intel® AVX, Intel® AVX2, Intel® AVX-512
Intel 64
Intel® AES New Instructions
Intel® Hyper-Threading technology
Intel® Optane™ Memory Supported
Intel® TSX-NI
Intel® Turbo Boost technology
Intel® Volume Management Device (VMD)
Intel® vPro™ Platform Eligibility
Number of AVX-512 FMA Units		2
Speed Shift technology
Virtualization
Intel® Virtualization Technology (VT-x)
Intel® Virtualization Technology for Directed I/O (VT-d)
Intel® VT-x with Extended Page Tables (EPT)