Microsoft to issue patch for Core Duo power glitch

It was just a few weeks ago that the sleuths at Tom's Hardware uncovered the cause of a mysterious power drain that reduced battery life in some Core Duo-powered laptops: an uncooperative Microsoft power-management driver. Now, Microsoft has done the right thing and is developing a patch to address the problem -- though the company has yet to announce a ship date for the fix. Props to Tom's for uncovering the bug in the first place, and for doing the digging to find out what was causing it. Let's hope that Microsoft (and Intel) learn from the experience: you can run down batteries, but you can't hide.

[OK @ Feb 14th 2006 2:39PM]

Actually, anandtech did some more testing and found that it affects pentium m single core laptops as well: http://anandtech.com/mobile/showdoc.aspx?i=2693

[Joe @ Feb 14th 2006 4:14PM]

I wonder how many other power drain issues there are and how many are due to poorly designed drivers?

[Chir @ Feb 14th 2006 4:23PM]

The Anandtech article clarifys the issue, including some loose ends that Tom's Hardware was unable to resolve. We can now conclude that the problem affects all laptops based on the Sonoma and Napa platforms. Which means that for years people have been losing battery power and not knowing it!

This issue truly shows the shoddy journalism at Tom's Hardware. Anandtech not only revealed the issue in more depth, they also did the right thing as journalists.

Tom's Hardware found that the preliminary fix increases the battery life of both their Sonoma and Napa test systems (Asus W3A and W3F) even without a USB device connected. Anandtech was able to reach the correct conclusion, that the integrated USB2.0 camera was the culprit. Tom's Hardware's article, however, did not address this.

Not only that, Tom's Hardware had to rely on Anandtech's article for the nature of the problem, i.e. the technical details regarding C3-C4 sleep states. But instead of referring to their source, Tom's tried their best to hide it, referring only to a "different enthusiast site". Worse yet, Tom's found out from Anandtech their mistake from their initial testing, the use of Perfmon to inaccurately monitor C3 state times. Anandtech discovered a secret Intel tool to monitor C3 times accurately, and Tom's took this info from Anandtech's article without mentioning Anandtech once again! Shame on you, Tom's Hardware.

Anandtech, on the other hand, publically referred to the Tom's investigation in the beginning of its article.

[ch424 @ Feb 14th 2006 4:37PM]

Que 40 posts about MacBook Pro and how wonderful OSX is because it doesn't have any bugs like this. Or indeed any other bugs. At all.

[Desides @ Feb 14th 2006 5:28PM]

"4. Que 40 posts about MacBook Pro and how wonderful OSX is because it doesn't have any bugs like this. Or indeed any other bugs. At all."

I think you mean "cue," and I think you also mean "40 posts from Apple haters pointing fingers at phantom Apple fanboys."

It's getting really old, people. Almost as old as the cell phone coverage around here.

[Deathwish238 @ Feb 14th 2006 5:28PM]

This is not a core duo problem only. It is a problem with any mobile pentium processor. It is not fair to simply call it a core duo glitch.

[Patricio @ Feb 14th 2006 5:53PM]

Do this affect the new Apple notebook which is based on the intel core duo ?

[andy @ Feb 14th 2006 6:17PM]

Actually, this is not a Core Duo problem OR a Pentium M problem. This is a driver problem for a certain OS. Don't blame Intel for shoddy software run on its parts.

[kev @ Feb 14th 2006 7:09PM]

What happened to semi-objective blogging? To reiterate what other people are saying, Intel is not at fault for this, and it's quite obvious that two cores will tend to dissipate more than one core.

[andy @ Feb 14th 2006 8:39PM]

While two cores will TEND to dissapate more than one core, this is not the case for the Core Duo vs. the Pentium M. The Core Duo parts are really that good. Intel actually got its act together, and with the fixed drivers, the Dual core platform uses the same or less power as the single core platform did.

[Flav @ Feb 14th 2006 10:53PM]

RE:7
To be more specific, this probably affects ALL laptops that have USB 2.0 and the speedstep equivalent (including AMD Cool 'n quiet). So unless someone has an ancient beast without USB 2.0 and/or without Speedstep (Mobile Pentium 4 perhaps?), this problem will certainly exist. So I'm almost certain it's quite widespread.

[A @ Sep 12th 2007 9:35PM]

Has this patch ever been issued? Just wondering ...