Q9550 overclock shows different speed in BIOS than..

[de8212]

it does in Windows.


I've got it set at 8.5 * 448 in BIOS. When booting it shows 3.8GHz. In W7 under the System Properties, it says "Q9550 @ 2.83GHz 3.40GHz"

Is Windows confused?

 

[BonzaiDuck]

Originally posted by: de8212
it does in Windows.


I've got it set at 8.5 * 448 in BIOS. When booting it shows 3.8GHz. In W7 under the System Properties, it says "Q9550 @ 2.83GHz 3.40GHz"

Is Windows confused?

A few opening questions -- deferring to others with more extensive experience with RC1/Windows-7. I've only installed it to get "first impressions" -- wasn't happy with the default support for my old CRT Viewsonic monitor (E70f), or the availability of drivers for some more dated hardware like my HighPoint RocketRaid 464 PCI card.

Do you have EIST and/or C1E enabled in the motherboard BIOS?

I didn't notice any inconsistencies with Windows 7 (64-bit) reporting CPU speed for a Wolfdale E8400 @ 3.6 Ghz. But I had the power-saving features turned off in BIOS.

Also, I'd need to refresh myself at the Intel Spec-Finder, so I'll ask here instead: Is 8.5 the default (and highest) multiplier for the 9550? Or did you lower the multiplier to get that host frequency of 448 Mhz? I've seen more than one case (first-hand) where lowering the multiplier results in the processor speed being incorrectly reported in certain software.

 

[de8212]

Bonzai - I really just used the same settings as in this youtube overclocking guide since it is the same mobo/cpu as mine. Those are the only settings I clanged.

http://www.youtube.com/watch?v...Bfjy6o&feature=related

 

[BTRY B 529th FA BN]

OP refer to this

 

[BonzaiDuck]

Good point, BTRY B 529th FA BN.

There's something else, though.

Keep in mind I'm not yet that familiar with the P45 chipset. But I took a look at the YouTube video de8212 has been watching. I don't care how long he (the You-Tuber) went in PRIME95 (he says 17 hours) -- or anything else. As far as I know at the moment, it is VERY UNWISE to leave "PCI-E Frequency" on "Auto" -- for any C2D/C2Q motherboard or chipset.

I checked the Q9550 spec, and indeed, the stock multiplier is 8.5. As I said, reducing the multiplier below stock has shown this sort of behavior he describes, but such is apparently not the issue.

I don't think the You-Tuber did anything with the power-saving EIST/C1E features (I must be so old now, that I just forgot what I saw five minutes ago). So I think we're pointing the OP in the right direction.

 

[de8212]

Thanks for the link BTRY and the additional info Bonzai. Unfortunately I am still a bit confused on the answer to my original question.

 

[BonzaiDuck]

Originally posted by: de8212
Thanks for the link BTRY and the additional info Bonzai. Unfortunately I am still a bit confused on the answer to my original question.

Keep in mind first of all that Windows 7 RC1 is "pre-release." I can't find drivers for a rather common PCI RAID controller. I can't seem to find a monitor-driver for my Viewsonic CRT. There may be all sorts of minor glitches and benign bugs in the OS.

If the power-saving features of the motherboard (EIST or "speedstep," C1E, etc.) are turned "on," this could explain the behavior consistent with your observation: that it shows the correct processor speed at bootup as the BIOS posts, but it reports a slower speed in Windows when the system is idling. The powersaving features drop the CPU multiplier and reduce the effective speed at idle, and raise the multiplier to its default level under load. This might account for the inconsistencies you see.

 

[de8212]

Thanks you Bonzai. Disabling EIST did the trick. I read through the other links but quite frankly just got more confused the more I read.

 

[dguy6789]

Originally posted by: BonzaiDuck

As far as I know at the moment, it is VERY UNWISE to leave "PCI-E Frequency" on "Auto" -- for any C2D/C2Q motherboard or chipset.

Could you elaborate?