EPYC 7302 vs Xeon W-3275M
Aggregate performance score
Xeon W-3275M outperforms EPYC 7302 by a significant 20% based on our aggregate benchmark results.
Primary details
Comparing Xeon W-3275M and EPYC 7302 processor market type (desktop or notebook), architecture, sales start time and price.
| Place in the ranking | 196 | 260 |
| Place by popularity | not in top-100 | not in top-100 |
| Cost-effectiveness evaluation | 4.96 | 11.23 |
| Market segment | Server | Server |
| Series | Intel Xeon W | AMD EPYC |
| Power efficiency | 11.55 | 12.72 |
| Architecture codename | Cascade Lake (2019−2020) | Zen 2 (2017−2020) |
| Release date | 3 June 2019 (5 years ago) | 7 August 2019 (5 years ago) |
| Launch price (MSRP) | $7,453 | $978 |
Cost-effectiveness evaluation
Performance per price, higher is better.
EPYC 7302 has 126% better value for money than Xeon W-3275M.
Detailed specifications
Xeon W-3275M and EPYC 7302 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 | 28 (Octacosa-Core) | 16 (Hexadeca-Core) |
| Threads | 56 | 32 |
| Base clock speed | 2.5 GHz | 3 GHz |
| Boost clock speed | 4.6 GHz | 3.3 GHz |
| Bus type | DMI 3.0 | no data |
| Bus rate | 4 × 8 GT/s | no data |
| Multiplier | 25 | 30 |
| L1 cache | 1.75 MB | 1 MB |
| L2 cache | 28 MB | 8 MB |
| L3 cache | 38.5 MB | 128 MB (shared) |
| Chip lithography | 14 nm | 7 nm, 14 nm |
| Die size | no data | 192 mm2 |
| Maximum core temperature | 76 °C | no data |
| Number of transistors | no data | 4,800 million |
| 64 bit support | + | + |
| Windows 11 compatibility | + | + |
| Unlocked multiplier | - | + |
Compatibility
Information on Xeon W-3275M and EPYC 7302 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 (Uniprocessor) | 2 (Multiprocessor) |
| Socket | FCLGA3647 | TR4 |
| Power consumption (TDP) | 205 Watt | 155 Watt |
Technologies and extensions
Technological solutions and additional instructions supported by Xeon W-3275M and EPYC 7302. You'll probably need this information if you require some particular technology.
| Instruction set extensions | Intel® AVX-512 | no data |
| AES-NI | + | + |
| AVX | + | + |
| vPro | + | no data |
| Enhanced SpeedStep (EIST) | + | no data |
| Speed Shift | + | no data |
| Turbo Boost Technology | 2.0 | no data |
| Hyper-Threading Technology | + | no data |
| TSX | + | - |
| Turbo Boost Max 3.0 | + | no data |
| Precision Boost 2 | no data | + |
| Deep Learning Boost | + | - |
Security technologies
Xeon W-3275M and EPYC 7302 technologies aimed at improving security, for example, by protecting against hacks.
| TXT | + | no data |
| EDB | + | no data |
Virtualization technologies
Virtual machine speed-up technologies supported by Xeon W-3275M and EPYC 7302 are enumerated here.
| AMD-V | - | + |
| VT-d | + | no data |
| VT-x | + | no data |
| EPT | + | no data |
Memory specs
Types, maximum amount and channel quantity of RAM supported by Xeon W-3275M and EPYC 7302. Depending on the motherboard, higher memory frequencies may be supported.
| Supported memory types | DDR4-2933 | DDR4 Eight-channel |
| Maximum memory size | 2 TB | 4 TiB |
| Max memory channels | 6 | 8 |
| Maximum memory bandwidth | 140.8 GB/s | 204.763 GB/s |
| ECC memory support | + | + |
Peripherals
Specifications and connection of peripherals supported by Xeon W-3275M and EPYC 7302.
| PCIe version | 3.0 | no data |
| PCI Express lanes | 64 | no data |
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 | 25.86 | 21.54 |
| Recency | 3 June 2019 | 7 August 2019 |
| Physical cores | 28 | 16 |
| Threads | 56 | 32 |
| Chip lithography | 14 nm | 7 nm |
| Power consumption (TDP) | 205 Watt | 155 Watt |
Xeon W-3275M has a 20.1% higher aggregate performance score, and 75% more physical cores and 75% more threads.
EPYC 7302, on the other hand, has an age advantage of 2 months, a 100% more advanced lithography process, and 32.3% lower power consumption.
The Xeon W-3275M is our recommended choice as it beats the EPYC 7302 in performance tests.
Should you still have questions on choice between Xeon W-3275M and EPYC 7302, ask them in Comments section, and we shall answer.
Similar processor comparisons
We picked several similar comparisons of processors in the same market segment and performance relatively close to those reviewed on this page.
