This article is rated Stub-class on Wikipedia's content assessment scale. It is of interest to the following WikiProjects:

Computing This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles ??? This article has not yet received a rating on the project's importance scale. Microsoft Windows: Computing Mid‑importance This article is within the scope of WikiProject Microsoft Windows, a collaborative effort to improve the coverage of Microsoft Windows on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.Microsoft WindowsWikipedia:WikiProject Microsoft WindowsTemplate:WikiProject Microsoft WindowsMicrosoft Windows articles Mid This article has been rated as Mid-importance on the project's importance scale. This article is supported by WikiProject Computing.

Does it also tell you how much memory is idle too, or just cpu? Either way, it should be added to this page JayKeaton 08:46, 19 May 2007 (UTC)Reply[reply]

Are you asking if it tells you how much memory is currently unused? If so, then no, that is part of the memory manager, not the scheduler. --Android Mouse 03:08, 24 May 2007 (UTC)Reply[reply]

The idle process doesn't really "tell" you anything. It just executes. Task Manager, Performance Monitor, and other tools will tell you how much the Idle process is executing, and that tells you how much idle CPU time you have. These tools also reveal free memory and other performance info, but Idle process has nothing to do with those. Jeh (talk) 22:21, 22 July 2012 (UTC)Reply[reply]

To which file really belongs the code of this process? NTOSKRNL?--Dojarca 19:10, 12 June 2007 (UTC)Reply[reply]

I just scanned the threads section in Windows Internals 4th edition and it didn't say specifically. Wherever the scheduling code is located is likely where the idle process code is located. For NT it's likely ntoskrnl, for 9x, who knows. --Android Mouse 19:26, 14 June 2007 (UTC)Reply[reply]

I don't know if it exists on 9x (Does it?), . 68.39.174.238 16:21, 16 September 2007 (UTC)Reply[reply]

The code running in the idle thread is mostly in ntoskrnl.exe, with some calls into the HAL. If you look at the idle process with a kernel debugger you'll find the thread functions are all nt!KiIdleLoop, nt being the debugger's abbreviation for ntoskrnl. Jeh (talk) 15:38, 11 October 2009 (UTC)Reply[reply]

As for 9x, 1) this article is about the NT family; and 2) who cares, really? Jeh (talk) 16:49, 11 October 2009 (UTC)Reply[reply]

What nonsensical foreign gibberish. Still no explanation to anyone's questions for years! No explanation about why System Idle Process consumes 94%+ of resources during mid-day peak usage times. — Preceding unsigned comment added by Fmpoor1234 (talk • contribs) 04:52, 10 February 2011 (UTC)Reply[reply]

(sigh) Yes, the article does explain that - or rather, it shows that that's a mistaken concept to begin with. The idle process doesn't "consume" anything, it shows you how much CPU time is not being used at the moment. Jeh (talk) 18:05, 17 July 2012 (UTC)Reply[reply]

This is easily one of the most poorly-written articles on Wikipedia. Please, someone with a good knowledge of this subject, rewrite this article!

--Agreed, it's shocking. I don't need to be told the same thing in five different ways.

I just rewrote the article from scratch. It still needs some improvement, but I hope it's a bit more acceptable this way. Feel free to revise it. Dolda2000 (talk) 06:33, 27 March 2008 (UTC)Reply[reply]

I came here to understand what a System Idle Process is but I understand very, very little from this article unfortunately. —Preceding unsigned comment added by 128.175.197.246 (talk) 00:55, 10 January 2009 (UTC)Reply[reply]

On the off chance that the anon decides to return, System Idle Process is basically PC-speak for "nothing". Assuming that everything is working properly, the amount of CPU time that Windows says System Idle Process is consuming -- that's actually the amount of CPU time that nothing is consuming. It's a placeholder -- if you're a gamer, you may better understand it by comparing it to MissingNo..

Of course, I've had many an experience where one or more programs were trying to do RAM- or CPU-intensive tasks, but the SIP was hogging the damn CPU (actually, both of them). The programs that needed the CPU were getting one or two percent of it, while the SIP kept the rest! Not sure how "nothing" can use everything -- probably just one of so very many programming errors Microsoft has made.

TL;DR: System Idle Process = Nothing. 68.48.239.52 (talk) (DavidJCobb on Wikia) 03:49, 14 January 2010 (UTC)Reply[reply]

No, it is not a "programming error." The System Idle Process cannot hog the CPU. Period, end of discussion. Its threads are scheduled below the priority that any other threads can possibly reach - even though they may show up at priority zero, they're really scheduled as if they had priority -1; it's a special case in the scheduler: If NO ordinary threads want to run on a free CPU, then and only then does the scheduler select that CPU's idle thread for execution. Whenever you see the idle process consuming almost all of your CPU time that is proof that nothing else wants any CPU time. If something else should be using CPU time, then the something else must be blocked (perhaps a deadlock). It is also possible that you have a device generating an interrupt storm and all of the system's time is being spent servicing the interrupts; enable the "kernel times" feature of Task Manager to see. But the Idle Process absolutely cannot be the cause of these situations. Jeh (talk) 04:35, 14 January 2010 (UTC)Reply[reply]

The reason the article says the same thing five different ways is that when we had fewer explanations, people kept coming here and asking "why does the Idle process only allow other tasks to use two or three percent of the CPU?!?" or "why is the Idle process using all of my CPU time?" and similar signs that they did not understand the concept. So, we added alternative explanations, trying to explain the same thing but in different ways. Very often different ways of explaining things "work" for different people. It seems to have worked as we haven't had one of these "idle process is hogging my CPU!" questions for quite a while, so I would suggest leaving the apparently redundant explanations. Perhaps some headings could be added to make the boundaries between them more clear. Jeh (talk) 22:27, 22 July 2012 (UTC)Reply[reply]

It seems to me that System Idle Process is simply an OS-specific version of Idle (CPU), and the two should be merged. This would be good for Idle (CPU), too, because it's currently rather short. Plus then we could treat how other OSes handle idle time as well. Thoughts? Objections? —DragonHawk (talk|hist) 15:52, 10 February 2010 (UTC)Reply[reply]

• Keep. There is considerably more detail here than belongs in an article about the general principle of a CPU "idling." Not every OS handles this by an actual context switch to a designated idle process, either. If other OSs are to be covered (and of course they should be), let them have their own articles. Jeh (talk) 19:13, 10 February 2010 (UTC)Reply[reply]

• • Maybe. On second thought... This article is only about 4K bytes. Even if 10 different OSs were covered in Idle (CPU) to this degree the total would still be well under the article length guideline. Of course a redirect would be left behind here. Are there other OS-specific Idle articles? If not, maybe the addition of Windows-specific details to Idle (CPU) would prompt other OS's SMEs to contribute there. Jeh (talk) 20:59, 22 March 2010 (UTC)Reply[reply]
  • Keep. it is a windows process and is very informitive --75.89.2.70 (talk) 23:33, 20 September 2010 (UTC)Reply[reply]
  • Not merge. Sorry for my English. It is good Windows-oriented article. Really, clear and informative. There is no any objective reasons to merge it with something else about the processor idle. Because this is article about Windows Idle Process. This in itself is interesting. Besides the article has been categorized as Windows Architecture article. In which one may find information about this concrete component of Windows Architecture. Whom is needed to terminate or deface good article? —Preceding unsigned comment added by 92.112.137.244 (talk) 03:26, 10 December 2010 (UTC)Reply[reply]

If you added information about how different operating systems treat the idle CPU idle then merging the articles would make sense, but it doesn't make sense to merge the articles without mentioning how other operating systems treat the idle CPU as well. Not all computers are windows. Absolute Relativity (talk) 02:47, 5 September 2010 (UTC)Reply[reply]

Agreed. Let's see individual articles, or material in other articles, about how Linux, BSD, MacOS X, VMS, etc., handle idling. Then once that material exists the articles can be merged if it still seems like a good idea. Merging now would just create an article with an OS-neutral title but with Windows-heavy content. Jeh (talk) 04:13, 5 September 2010 (UTC)Reply[reply]

I came here tryimg to find out exactly how this works and the article hasn't really made this clear - on XP, over dozens of installations, including relatively fresh ones I have assumed that it is a '% of CPU going spare', yet the system can have slowed to a halt with over 50% available even though there is still plenty of RAM (and HDD space) available as well. I've been trying to figure out why this would happen if it is just a marker for how much of the CPU is not being used and the article doesn't really explain this. According to the article, this scenario shouldn't exist - yet it does. Maybe it's down to sloppiness of how the NT OSes are built - but I feel that this should have a mention so that others can understand why this happens. groovygower (talk) 22:04, 13 September 2012 (UTC)Reply[reply]

I get that you repeat the same information multiple times is so that no one misses it, but this is tantamount to raising your voice and repeating the same thing when the other person doesn't understand you. If they didn't get it the first time, saying it louder won't help.

What would be useful is a little background here. Why is the idle process necessary? When was it created and what happened without it? If people are given a chain of logic which makes sense and which leads back to the information you wish to sink in, not only will they understand better but you then have a legitimate reason to repeat yourself. Thetrellan (talk) 17:36, 9 December 2015 (UTC)Reply[reply]

I'm not seeing the excessive repetition that you're apparently seeing. There is an early statement of general principles, followed by refinements.

Necessary: it isn't strictly speaking necessary, but it's convenient. The article already covers this - it avoids a special case that would occur when there's no ordinary thread to switch to.

When - NT had this from its beginning. NT's "spiritual parent", DEC's VMS, did also. So in the context of these operating systems there was never any time "without it" to compare to. Jeh (talk) 17:50, 9 December 2015 (UTC)Reply[reply]

I took another look at it and removed some repetition. What do you think? Jeh (talk) 19:31, 9 December 2015 (UTC)Reply[reply]

Looks like it. Still there in process hacker... 2A00:1370:8184:1765:2320:789C:D12F:3C7A (talk) 08:15, 15 November 2022 (UTC)Reply[reply]