Chen LinCai陈林材

内存容量拉满和速度拉满的问题

clcon 4 days ago


内存容量和速度是个矛盾体,128G内存插满达到最大内存,却损失了内存的速度,如果是剪辑视频,ai的对内存的容量要求大于速度,毕竟主要是用显存来处理计算和画图和内存用来加载和临时存储。目前ddr5插两根内存可以最大化内存的速度,可以xmp超频,但是内存容量最高也就是48+48,一共96G,而且好的内存一般只是32G一条,那样64G内存就只能在速度上满足,对画图和渲染不友好。 The central node would state "Memory Trade-Offs." From there, two branches would extend: one labeled "Capacity" and the other "Speed." The "Capacity" branch would have sub-nodes like "128GB Maximum," "Video Editing," "AI Tasks," and "VRAM for Computation." The "Speed" branch would detail "Reduces with Full Slots," "XMP Overclocking," and "DDR5 Optimizations." Additional branches from "Capacity" would note the max of "96GB with 2 Sticks" and "Commonly 32GB per Stick." Another branch might illustrate the conclusion: "64GB for Speed vs. Graphics and Rendering."