Author: Robert Hyatt
Date: 14:19:54 01/25/00
Go up one level in this thread
On January 25, 2000 at 16:28:31, Georg v. Zimmermann wrote: >Hi, > >you can switch it on and off once per hour without effect on your hardware >during normal PC-lifetime. There are two issues here. 1. 99% of failures _do_ occur on power up (or on power down, but you don't detect them until power up.) This is caused by thermal expansion. When you power them off, heat dissipates differently from different places on (say) a CPU chip, or an IDE controller chip, or whatever. And that means different rates of expansion/contraction that produce significant stress levels (mechanical stress) everywhere. From that perspective, powered up all the time is better, since the parts in the cpu really don't wear out. 2. Monitor failures are different. The CRT has a definite lifetime, as a filament is heated to produce electrons to illuminate the phosphor dots on the CRT face. Over time, the electron production falls off. Some newer monitors actually nearly power themselves down on inactivity, which prevents this. For older monitors, powering them off is likely good, although this thermally stresses the HV power supply in the thing... We have some gateway machines that are 7 years old in a lab. They run 24 hours a day, 7 days a week. Students power the monitors off, but the PCs are running NT and stay up constantly. Machines that are on/off every day don't seem to do that well... > >Actually if it doesn't do anything at night, you _should_ switch it off IMHO. >Consider the enviromental issue... its our and our childrens world. > >Regards, >Georg v. Zimmermann Energy is another issue altogether of course. :)
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.