In this video I’m testing a question that comes up a lot in Chia farming: does running eight drives as a RAID0 array actually use more power than farming the same eight drives as individual disks? I’m in the middle of replotting for newer formats (like Gigahorse/Mad Max), and I didn’t want to build out the rest of my rack only to find out I picked the wrong disk layout. The logic says RAID should pull more power because reads get striped across all drives, so more heads are moving more often—but I wanted real numbers. To keep the test clean, I didn’t use a wall power meter. I powered a block of eight 18TB SAS drives with a Pico ATX supply and measured DC input power through a shunt (Batrium shunt block), sampling about every 1.5 seconds and logging it to MySQL. Under normal farming load, eight individual disks averaged about 51W total (6.4W/drive). The same eight drives in RAID0 averaged about 57W total (7.1W/drive), roughly 52 kWh/year more per 8 drives—about $9/year at my ~$0.17/kWh. When you scale that to a full rack, it adds up, and that’s why I’m sticking with individual drives: lower long-term power cost and way easier drive identification and replacement when something fails. RAID0 is still tempting for fast plotting/offload workflows, but for my farm, the tradeoffs don’t pencil out.