Can anyone Explain how this is Possible?

DJG said:

Cool! Let's have a Top Ten Reasons You Have Negative CPU Usage Contest

- Too close to an event horizon?
- Neighbor is sucking your CPU through your wireless network?
- Using a CPU Usage Meter that doesn't work right?

We need some more

portal anomalies...maybe
(i would thought by now that we would have taken care of those.....)

zigzag3143 said:

Um Its in power saver mode

Ken

actually its balance mode...
unless i misread your comment

holo88 said:

your cpu is so unfathomably awesome that is uses compressed time and space to preform tasks, instead of the cpu, itself? i want one!

And its a AMD (take that intel!...)

Antman said:

No. You can return a negative value with an improperly configured calculation. Are you playing with WMI repository? What app is showing that?

I think you are playing with us.

hmm i didnt use WMI or even know that you can change your cpu usuage
so this would be a signed float number correct?

ciphernemo said:

My bet is on poorly written software.

i never thought of that of mark russinovich...

Antman said:

We coders prefer underdeveloped. Undocumented feature rich.

lol i hear that...

DJG said:

Oooh! I want one of those too!

you need a portal device to and the right moment to activate to create a anomaly inside your cpu

You can't be serious!

yep alot of software is written with so called "undocumented features that can cause errors among other things....



as for the real reason why this is happening is because of disk corruption (which is spanning across the disk for some reason...)
this is why the interrupts are peaked (the OS is retrying and retrying to read a sector and by that time there is another sector to be read and so on...)
although this i cannot explain unless these are the app requesting to be written on to (or it might be too many portals....)

Cindy said:

Have you tried another CPU reading app to compare darkassain ?

actually yes i have core temp and this one is off the charts....


task manager shows me 0% although you can clearly see that there is some usage


i guess i do live up to my saying...

Cindy said:

How strange !

yep like i said i do live up to my name...
now that you mention it though i compared both procexp and from what i can see core one is behaving itself while core 2 is erratic...




but what i want to know is why is this happening...

It seems that you are playing with a new codec.

Is the problem on the machine with the HD3200 or with the integrated video?

I found a utility to identify supported DXVA modes, but it is in Japanese. Have not examined it. Yes, I understand that you have probably investigated your supported modes already. I am looking for a way to isolate what is making the call/calculation.

I'm getting confused here - the problem is on the tablet, no?

OMG! 84 processes??! I take it you're running all the out-of-the-box bloatware. I'm a tablet + workstation dude too, but I'm only running 56-58 processes on my Lenovo tablet, which out-of-the-box also shows up with 80+ processes, probably including one that lets you know when you should go to the bathroom ...

I just run the stuff that enables the special buttons, screen rotation, the power manager, plus I use Ritepen instead of the TIP.

Edit: In your system specs some of the entries are flipped between what's in the tablet & the home server ... I take it you didn't somehow embed an ATI HD3200 PCI-E card into your tablet, right? Cause that could really make your CPU flip out .

Not useful to this conversation but with all the mention of poor coders I wanted to add that if you really want to get an accurate CPU utilization use Native APIs (yes undocumented). WMI is great for quick scripts but I wouldn't use it for a real-time monitoring software.

Edit: This is the Native API ZwQuerySystemInformation