Timeline for Why would your stasis pods participate in your grid computing network?
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Oct 4 at 23:49 | comment | added | Questor | @DoktorJ I think intel's recent issue with their micro-code shows what can happen... Yes the CPU has a thermal sensor/governor to monitor it, but that is code written by a person and people make mistakes. | |
| Nov 17, 2017 at 18:24 | comment | added | Doktor J | @Andrey why wouldn't there be thermal sensors and governors (much like those that exist in modern day consumer and enterprise CPUs) that, if the overclocking is resulting in higher-than-expected die temperatures, the speed is throttled back until the processor cools down again? | |
| Aug 10, 2017 at 19:36 | comment | added | Andrey | @kingledion You are right, cpus idle. But cpus can also be fluidly overclocked. There could be such a system that could for short bursts push the overclock above the heat sink capabilities. If sustained it could cause the cpu to start producing errors | |
| Jan 5, 2017 at 16:32 | history | edited | Frank Cedeno | CC BY-SA 3.0 |
added 783 characters in body
|
| Jan 5, 2017 at 16:08 | comment | added | kingledion | Well, I just did 'top' and my CPU usage is 19.1%, so I doubt many readers are at 100%. When a system idle process is running (with the remaining resources), it is just executing the HLT instruction. So the CPU isn't doing any actual computation when it is halting, and this is specifically used to cut power usage. So, you aren't right to say that 'all CPU are running at 100% capacity at all times'. But it is true that a lot of big servers are running at near full capacity for months or years, so your premise is still accurate. | |
| Jan 5, 2017 at 16:00 | history | answered | Frank Cedeno | CC BY-SA 3.0 |