new KB959233: Speeding up Windows Vista

[raskren]

Someone should email this article to Mark Russinovich. I bet he can get to the bottom of it.

[Panda X]

And when I posted it here and on AskVG.com long time back, a few ppl told that its a hoax and doesnt work. ;)

It is. It's been confirmed from Microsoft themselves that that doesn't do anything. It already uses your full amount of cores on boot except for a few seconds where it doesn't need it and it wouldn't benifit from having both+ enabled. It's a hoax 100%

[stevember]

may be Windows is able to autodetect the number of cores only after the boot is ended, so this option allows the boot process to be informed about the number of cores and start the boot process in parallel mode?

but why Microsoft didn't enable this option automatically?

why the boot process doesn't read the number of cores from the hardware config saved in the registry?

This is a mistery...

No, any detecting software needs to be loaded and run therefore slowing booting, so best presume one core get on with it, if chance of predefining based on later detection great idea.

[raskren]

I sent Mark an email and pointed him at this forum topic. We'll see...

[mad_onion]

yes this needs to be cleared up, last I heard from anyone who actually knows what their talking about this has no affect on boot time.

[Mercellus]

I sent Mark an email and pointed him at this forum topic. We'll see...

This should be interesting.

[raskren]

Mark Russinovich replied to my email. He requested that I not post the entire email exchange verbatim, so I'll paraphrase.

In my email I suggested that these tweaks were totally useless. Mark agreed and added that they are "incorrect and potentially damaging guidance" and that the "[series] will be pulled."

[-Nic-]

Mine was already set to 2?

[kaiwai]

I have a quad core, set it to 4 processors and rebooted. Didnt notice any increased performance, but I let it stay like that anyway..thanx

I'd probably say that one will get alot more benefits with Windows 7 given that they've parallelised alot more of the start up sequence. I guess one will only really see major benefits when it comes to systems with alot of services loading at boot up.

[Mercellus]

Mark Russinovich replied to my email. He requested that I not post the entire email exchange verbatim, so I'll paraphrase.

In my email I suggested that these tweaks were totally useless. Mark agreed and added that they are "incorrect and potentially damaging guidance" and that the "[series] will be pulled."

Haha, nice! I figured he'd say something like that, he is a very smart fellow.

[Yusuf M. Veteran]

If it's a hoax then why is it still up? Of course, I'm not implying that it's true/untrue. I'm simply asking.

[Brandon Live Veteran]

This is very embarassing that something so blatantly wrong got up on a microsoft.com server. I (and others like MarkRus) are contacting the owners of this content to get it removed.

This setting has NO IMPACT on boot performance, except to limit it to make your system artifically SLOWER by disabling processors. This exists for debugging purposes only (for example, developers use this setting to test their applications/drivers in a simulated single core environment if they have a multi-core CPU).

[Mercellus]

This is very embarassing that something so blatantly wrong got up on a microsoft.com server. I (and others like MarkRus) are contacting the owners of this content to get it removed.

This setting has NO IMPACT on boot performance, except to limit it to make your system artifically SLOWER by disabling processors. This exists for debugging purposes only (for example, developers use this setting to test their applications/drivers in a simulated single core environment if they have a multi-core CPU).

What I'd love to know is how this was published in the first place; was it written by someone who manages the knowledge base itself or was it written by an MVP?

[Brandon Live Veteran]

What I'd love to know is how this was published in the first place; was it written by someone who manages the knowledge base itself or was it written by an MVP?

I don't know, but I'm sure the appropriate people will look into how this happened and take steps to prevent similar errors in the future. In the meantime, please don't alter this or any related settings as you could potentially hinder the operation of your machine.

[darkz]

WTF am i special or what?

i couldn't live with the fact i'm proven wrong, so i double checked.

when advanced boot options looks like this (deselected checkbox, showing 1, like it was when i first opened it) my system boots in 1min

when it looks like that it boots in 51secs. offcourse these times are not entirely accurate, cause i still have to activate/deactivate the chronometer, but it still *boots faster*

someone could perform the same test and videotape it

[Shadrack]

Remember that during boot, the major bottleneck is the hard drive, as it needs to load a heck of a lot of data from that.

Getting a faster hard drive will give you a much bigger boost than using more than one core.

exactly.

[ViperAFK]

mine boots exactly the same speed with this...

[Ci7]

This is very embarassing that something so blatantly wrong got up on a microsoft.com server. I (and others like MarkRus) are contacting the owners of this content to get it removed.

This setting has NO IMPACT on boot performance, except to limit it to make your system artifically SLOWER by disabling processors. This exists for debugging purposes only (for example, developers use this setting to test their applications/drivers in a simulated single core environment if they have a multi-core CPU).

mine boots exactly the same speed with this...

....

[Brandon Live Veteran]

WTF am i special or what?

i couldn't live with the fact i'm proven wrong, so i double checked.

when advanced boot options looks like this (deselected checkbox, showing 1, like it was when i first opened it) my system boots in 1min

when it looks like that it boots in 51secs. offcourse these times are not entirely accurate, cause i still have to activate/deactivate the chronometer, but it still *boots faster*

someone could perform the same test and videotape it

Boot time can be inconsistent. Any difference you measured is likely random, and certainly has nothing to do with this setting.

[darkz]

Boot time can be inconsistent. Any difference you measured is likely random, and certainly has nothing to do with this setting.

then why does it happen two times in a row, with about exactly the same improvements?

did you measure by yourself?

[ViperAFK]

you do realize Brandon Live works for microsoft and probably knows what he's talking about right?

[Smigit]

you do realize Brandon Live works for microsoft and probably knows what he's talking about right?

the guys that wrote this article probably work at ms too :p

I do have faith in Brandon but, however it may be possible that the setting is working for that user.

[wellofsouls]

you do realize Brandon Live works for microsoft and probably knows what he's talking about right?

And this article is from a Microsoft KB.

Sometimes it seems some guys from Microsoft don't really know what they are talking about, one way or another, including Bill Gates and Steve Ballmer :laugh:

Personally, since I mostly keep up my Vista machine 24/7, so any increase in boot time doesn't really matter to me anyway. :p

[darkz]

you do realize Brandon Live works for microsoft and probably knows what he's talking about right?

and that means he's an unbeatable ****ing authority right?

better go measure it yourself and report the results, i don't really give a **** where any of you work as long as you don't care about me

[Mercellus]

I am confident that Brandon knows what he is talking about.