EPYC 9654 vs Celeron J4025
Aggregate performance score
EPYC 9654 outperforms Celeron J4025 by a whopping 8061% based on our aggregate benchmark results.
Primary details
Comparing Celeron J4025 and EPYC 9654 processor market type (desktop or notebook), architecture, sales start time and price.
| Place in the ranking | 2533 | 7 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 2.62 | 1.37 |
| Market segment | Desktop processor | Server |
| Series | no data | AMD EPYC |
| Power efficiency | 8.77 | 19.88 |
| Architecture codename | Gemini Lake Refresh (2019) | Genoa (2022−2023) |
| Release date | 4 November 2019 (5 years ago) | 10 November 2022 (2 years ago) |
| Launch price (MSRP) | $107 | $11,805 |
Cost-effectiveness evaluation
Performance per price, higher is better.
Celeron J4025 has 91% better value for money than EPYC 9654.
Detailed specifications
Celeron J4025 and EPYC 9654 basic parameters such as number of cores, number of threads, base frequency and turbo boost clock, lithography, cache size and multiplier lock state. These parameters indirectly say of CPU speed, though for more precise assessment you have to consider their test results.
| Physical cores | 2 (Dual-core) | 96 |
| Threads | 2 | 192 |
| Base clock speed | 2 GHz | 2.4 GHz |
| Boost clock speed | 2.9 GHz | 3.7 GHz |
| Multiplier | no data | 24 |
| L1 cache | 56 KB (per core) | 64K (per core) |
| L2 cache | 4 MB (shared) | 1 MB (per core) |
| L3 cache | no data | 384 MB (shared) |
| Chip lithography | 14 nm | 5 nm, 6 nm |
| Die size | 93 mm2 | 12x 72 mm2 |
| Maximum core temperature | 105 °C | no data |
| Number of transistors | no data | 78,840 million |
| 64 bit support | + | + |
| Windows 11 compatibility | + | no data |
Compatibility
Information on Celeron J4025 and EPYC 9654 compatibility with other computer components: motherboard (look for socket type), power supply unit (look for power consumption) etc. Useful when planning a future computer configuration or upgrading an existing one. Note that power consumption of some processors can well exceed their nominal TDP, even without overclocking. Some can even double their declared thermals given that the motherboard allows to tune the CPU power parameters.
| Number of CPUs in a configuration | 1 | 2 |
| Socket | Intel BGA 1090 | SP5 |
| Power consumption (TDP) | 10 Watt | 360 Watt |
Technologies and extensions
Technological solutions and additional instructions supported by Celeron J4025 and EPYC 9654. You'll probably need this information if you require some particular technology.
| AES-NI | + | + |
| AVX | - | + |
| Enhanced SpeedStep (EIST) | + | no data |
| Precision Boost 2 | no data | + |
Virtualization technologies
Virtual machine speed-up technologies supported by Celeron J4025 and EPYC 9654 are enumerated here.
| AMD-V | - | + |
| VT-d | + | no data |
| VT-x | + | no data |
Memory specs
Types, maximum amount and channel quantity of RAM supported by Celeron J4025 and EPYC 9654. Depending on the motherboard, higher memory frequencies may be supported.
| Supported memory types | DDR4 | DDR5-4800 |
| Maximum memory size | no data | 6 TiB |
| Maximum memory bandwidth | no data | 460.8 GB/s |
Graphics specifications
General parameters of integrated GPUs, if any.
| Integrated graphics card | Intel UHD Graphics 600 (250 - 700 MHz) | no data |
Peripherals
Specifications and connection of peripherals supported by Celeron J4025 and EPYC 9654.
| PCIe version | 2.0 | 5.0 |
| PCI Express lanes | 6 | 128 |
Synthetic benchmark performance
Various benchmark results of the processors in comparison. Overall score is measured in points in 0-100 range, higher is better.
Combined synthetic benchmark score
This is our combined benchmark performance rating. We are regularly improving our combining algorithms, but if you find some perceived inconsistencies, feel free to speak up in comments section, we usually fix problems quickly.
Passmark
Passmark CPU Mark is a widespread benchmark, consisting of 8 different types of workload, including integer and floating point math, extended instructions, compression, encryption and physics calculation. There is also one separate single-threaded scenario measuring single-core performance.
GeekBench 5 Single-Core
GeekBench 5 Single-Core is a cross-platform application developed in the form of CPU tests that independently recreate certain real-world tasks with which to accurately measure performance. This version uses only a single CPU core.
GeekBench 5 Multi-Core
GeekBench 5 Multi-Core is a cross-platform application developed in the form of CPU tests that independently recreate certain real-world tasks with which to accurately measure performance. This version uses all available CPU cores.
Pros & cons summary
| Performance score | 0.92 | 75.08 |
| Recency | 4 November 2019 | 10 November 2022 |
| Physical cores | 2 | 96 |
| Threads | 2 | 192 |
| Chip lithography | 14 nm | 5 nm |
| Power consumption (TDP) | 10 Watt | 360 Watt |
Celeron J4025 has 3500% lower power consumption.
EPYC 9654, on the other hand, has a 8060.9% higher aggregate performance score, an age advantage of 3 years, 4700% more physical cores and 9500% more threads, and a 180% more advanced lithography process.
The EPYC 9654 is our recommended choice as it beats the Celeron J4025 in performance tests.
Note that Celeron J4025 is a desktop processor while EPYC 9654 is a server/workstation one.
Should you still have questions on choice between Celeron J4025 and EPYC 9654, ask them in Comments section, and we shall answer.
Other comparisons
We've compiled a selection of CPU comparisons, ranging from closely matched processors to other comparisons that may be of interest.
