AMD EPYC 4004 CPUs Launched: Up To 16 Cores At 5.7 GHz For $699, Crushing Intel Xeon In Perf & Efficiency, 2.8x Faster In Crypto Mining

AMD has launched its EPYC 4004 CPUs for the AM5 platform, offering big gains in performance and efficiency versus past EPYC & Intel Xeon chips.

AMD Brings EPYC 4004 "Zen 4" CPUs To AM5 Platform: Deliver Big Gains In Value, Performance & Efficiency While Also Boosting Crypto Mining By 2.8x

The AMD EPYC 4004 Desktop CPUs are centered around the AM5 socket which we have seen on the consumer-oriented Ryzen platforms. The socket has so far supported Ryzen 7000 & the Ryzen 8000G CPUs and will soon get Ryzen 9000 "Zen 5" CPU support too. The Zen 4-powered EPYC 4004 Desktop CPUs come in a range of options, scaling from 4 cores to 16 cores and offering up to 128 MB of L3 cache with the 3D V-Cache-boosted "X" variants. These chips are designed to offer strong value, performance, and efficiency.

Related Story Microsoft Intros Copilot+ PCs For The “AI PC” Era: Snapdragon X First, Intel Lunar Lake & AMD Strix Later With Faster NVIDIA RTX & AMD Radeon GPUs

In performance comparisons, AMD is pitting the chips against the Xeon E-2400 "Raptor Lake" CPU family which scales from just 4-8 cores. These chips are the only mainstream server platform that is offered by Intel so it makes sense to compare the two lineups. AMD touts a 1.7x increase in perf, 1.5x increase in perf/$, 1.7x increase in perf/system$, and 1.5x increase in perf/Watt. Comparing the top Intel 8 core with its own 8-core EPYC 4004 CPU, the company reveals that at a $277 lower cost, you get 80% faster performance and 50% better efficiency.

A range of performance comparisons are showcased but the one that intrigued me the most is the Cryptography and Cryptocurrency "CPU Mining" performance slide which shows that the new EPYC 4004 CPUs offer a 2.4x performance per $ gain and 60% better efficiency. AMD's Ryzen, EPYC, and Threadripper CPUs are being used for mining purposes in specific algorithms that favor their large number of cores and cache and offer great efficiency and performance in those tasks.

Press Release: AMD