Sign in with
Sign up | Sign in
Your question

Microsoft to issue official patch to fix power drain bug

Last response: in CPUs
Share
February 14, 2006 4:47:43 PM

We have posted an update to our power drain bug story (http://www.tgdaily.com/2006/01/28/toms_hardware_uncover...).

Microsoft confirmed to us today that it is working onan official patch to be released to the public. This patch will consist of a regular update as well as a BIOS update.

Not really bad news, in the end.
February 14, 2006 7:56:30 PM

I think this BUG effects all computers. I have an Athlon XP desktop and running CPUIDLE Extreme and MBM to keep it cool for overclocking. About a year ago I bought an Epson R800 printer with USB2 and Firewire connection. I noticed that when I plugged in the Epson via USB2, my CPUidle efficiency dropped about 35percent and my temperatures went up about 12 degrees because the USB was like running to many cycles. I noticed this happened when I plug anything into the computer like USB digital camera. I contacted Epson and all they recommended was a new driver but I already had the latest. Then they of course recommended a reinstall of the OS, wasn't going to happen. So I tried the firewire port and everything went back to normal. This is an old problem that they are just now getting around to dealing with
February 14, 2006 8:02:05 PM

not all computers.... but more than just the Intel Core Duo... I'll agree with that :) 
February 14, 2006 8:10:04 PM

My congratulations to THG for working with Microsoft in solving that problem. :trophy:[/quote]

I was under the impression the guys at Anandtech caught it first?
February 14, 2006 8:13:33 PM

Finally!
February 14, 2006 8:13:51 PM

Just a quick clarification. I am quite sure that we never indicated that it was a problem that affected just the Core Duo.

We always were very clear that especially Intel claimed that all notebook platforms (read: AMD as well) are affected by the bug. How much at least Sonoma is affected can be seen in the article.
February 14, 2006 8:37:27 PM

Quite frankly, I really don't care, who came across the issue first, even if it is quite obvious who did.

It's important to see that there a result that makes sense to all of us. It was well worth to go the extra mile.

That said, yes, it is quite stunning how easily someone else can capitalize on someone else's efforts. I have said it before and I do it again. There were articles that violated the very basic rules journalism is built on.

I just read through a thread over engadget which praises Anandtech for driving this issue forward and trashes Tom's Hardware for delivering incomplete data. Did we? Or was Anandtech to quick to draw conclusions? Is their article complete?

No, it drilled down into a bugfix - and it only could do that because of our initial article. In our opinion, Anandtech only repeated what was known anyway, just in a more thorough way. And that is not necessarily a bad thing - even if we belive that his conclusion is incorrect - as we stated in our update article.

Solving such a problem takes a bit more than jumping on a single issue. THG invested several hundred man hours in the past weeks to reach today's announcement - not just in watching sleep states, but in engaging into discussion, opening a dialog with manufacturers, consulting system builders and having the clear vision that an update article could only be published if there is positive news for the reader. This is what we have done today.

And yes, we intentionally left out a link to Anandtech. People who haven't read the Anandtech article could care less who wrote about it. And people who know about the sorry - garbage - that was published on DailyTech and the misrepresentation of our findings on Anandtech, they know anyway where to find it. There's no need to escalate this issue from a pre-school to a kindergarden argument.
February 15, 2006 12:20:08 AM

OK, i'll join this thread then.

my question is:
If it's only an M$ bug, why is a BIOS update needed? (as explained in said article.)

BIOS update to me means that either:
1 - there's something wrong with the hardware aswell, or
2 - at the very least m$ has some trouble accessing the hardware.

if 1, then is thisforumregstrationsucks suffering from a totally different problem (having an AMD), and/or will CoreDuo with a different (non intel) chipset and Mac Powerbooks be totally immune to this?
if 2, then m$ are idiots, and/or intel made the hardware too difficult to talk to.


and the BIOS update needed, will the consumer download it directly from m$, or will they send it to the mobo makers who will incorporate it into a bios flash? i don't like the idea of m$ going anywhere near my bios, who knows what kind of drm crap they'll install.thankfully, i'm never going to get a laptop, or anything else, with any kind of m$ anything preinstalled. i'm just worried about Joe Consumer
February 15, 2006 12:26:09 AM

good point, I don't trust M$... but I would definitely install the patch... but the BIOS flash... I'd wait for a update from the mobo manufacturer. I don't think the problem is the Intel chipset, cuz apparently other computers suffer from the same problem. I think it is mainly a M$ problem with perhaps a bit of a hardware/firmware problem.
February 15, 2006 4:44:39 AM

I'm experiencing a similar problem (May not be related) with my new Vantec 2.5" and 60 Toshiba Drive. I'm not using it on a notebook and i'm convinced it is a USB port problem not giving enough power.

The HDD will not format. If formatted it will copy the data very slowly and and the data is corrupt.

hope to hear from M$ soon

Theo
February 15, 2006 4:53:19 AM

Why the bios update? It would seem likea reasonable spot to place information you want included, when the system comes out of speed mode.
February 15, 2006 12:23:40 PM

I agree that Anand's articals have been in very poor taste. I hope MS will have a patch, and the damage done to core duo can be undone.
February 15, 2006 1:31:06 PM

Conroe,

I believe it has been mentioned multiple times. No one at Tom's Hardware ever claimed that is is a bug that only relates to Core Duo. The information that this bug may relate to aother platforms - read: not just Sonoma, but AMD as well - was published in our original article.

The damage has been done, quite frankly over at Anandtech, which either has not read every page of our article, or simply misrepresented our findings. Anyone reading our articles and Anand's articles will see what is happening here.

Wolfgang
February 15, 2006 2:10:06 PM

I dunno windshear, in my case over a year ago the system was AMD athlon XP 3000+ barton core, on DFI Lanparty KT400A desktop system. Some overclockers running cpuidle or waterfall or similar software to keep CPU cooler may have noticed this "bug" a long time ago but couldn't figure out what the deal was and it wasn't as important on a desktop PC as opposed to a notebook where every bit of juice counts.

I would be interested to see if anyone would test the effects of the USB on even desktops especially tuned PC's being overclocked to see what impact this has on their efficiency and temperatures. This is another reason Firewire is better than USB.
February 15, 2006 2:51:46 PM

Quote:
Conroe,

I believe it has been mentioned multiple times. No one at Tom's Hardware ever claimed that is is a bug that only relates to Core Duo. The information that this bug may relate to aother platforms - read: not just Sonoma, but AMD as well - was published in our original article.

The damage has been done, quite frankly over at Anandtech, which either has not read every page of our article, or simply misrepresented our findings. Anyone reading our articles and Anand's articles will see what is happening here.

Wolfgang


I think it's a flame war. Sad, realy. The two best acting like kids.

I did read the artical, and the new all over the web was "core duo bug."
February 15, 2006 3:59:36 PM

Good job guys. That's the proper way to do it! ;-)

-mpjesse
February 15, 2006 4:19:41 PM

Quote:
The damage has been done, quite frankly over at Anandtech, which either has not read every page of our article, or simply misrepresented our findings. Anyone reading our articles and Anand's articles will see what is happening here.
The only article I've read at Anand is this one, so maybe I've missed something. But what I've read doesn't even support your pity party here. Can you please point me to where "the damage has been done", because I've read both articles and I don't see what you see.

What I see is that TG published a very preliminary (and as is always the case as such, incomplete) article about the problem, laid more blame on Intel than was deserved, and though not stated, did imply that the duo proc was a major factor.

Anandtech took that knowledge, went a lot further and into more detail with it, and came out with more fair and reasonable conclusions that countered this mis-implication.

Then TG got around to their own more in-depth continuation article that just wasn't as good as Anand's, and continued to try to equally blame Intel and MS, even though admitting that it is a Microsoft driver issue, and finding that, to quote from the second TG article, "Intel was restrained by a confidentiality agreement with Microsoft".

So to recap, TG spearheaded the discovery, which is good. But then failed to properly distribute blame where it was clearly due in an attempt to be, what, unbiased? And Anandtech's article was clearer, better written, and went into considerably more technnical detail than TG's equivalent follow-up article. So TG editors whined, bitched, and moaned. As a pro-TG-biased long-standing member of THGC, that's what I see. And if as biased as I am towards TG, I see it that way, imagine how unbiased people see it.

In the end though my biggest disappointments are really that neither site bothered testing with Linux, MacOS, or prior versions of Windows to definitively identify to consumers which OS's are affected, neither site has explained how MS can update a BIOS, and neither site explained why a BIOS update is part of the fix for what so clearly seems to be an OS issue. Which is more disappointing from the TG end because the editors claim to be ever so more concerned for the consumer. So then why aren't such simple questions looked into, for the benefit of the consumer?

And my only guess to why a BIOS update is needed is that it has something to do with handling the exit of a stand-by state. Which I could only even get after reading Anandtech's article. It would be nice if this matter could be cleared up so that I don't have to guess.
February 15, 2006 4:48:15 PM

silvr_phoenix,

let me just repeat one more time that we really don't care who finds these things first. the goal is the result. At least that is our interest.

Yes, our first article was a preliminary article. Without that initial research, Microsoft would not have publically conceded the existence of this bug. Without this article, Anandtech would not have known where to look at and would not even had access to that very patch we outlined.

Second, Anandtech and Dailytech mentioned that we blamed Core Duo. This is plain false and damaging in my personal view. We simply stated that the bug surfaces only in the Core Duo notebook we had available at the time, but that there is credible indication that all notebooks supporting USB 2.0 and including XP SP2 are affected.

The quote you mention does not blame Intel in any way. However, we do question Intel's approach to not force Microsoft to release a bugfix earlier.

Yes, Anand did thorough testing. However, in our opinion, his conclusions were not completely accurate (which you can see in our second article) and incomplete. As you can see in our article, C2 and C3 may be just one variable of the equation, but not the only one. It's easy to say it, but we had the C2 and C3 findings Anand presented in his article one week earlier. But they are simply not sufficient to present a solution to the user. So, if you believe Anand - do you believe us?

I do agree that Anand brought additional explanations and technical information earlier and in a more comprehensive way, but quite frankly, it did not change anything to the conclusion we drew in our first article. Both texts say that the driver is at fault and that multiple platforms can be affected.

In our opinion, such an article can only be released, if there is a solution for every reader. Simply saying that there is a problem, does not help very much. Please read our statement, if you you'd like to know why we released the first article.

Sorry for using some harsh words here, but I am just getting a bit tired of being accused that we have missed the mark. Anand can praise the Core Duo platform in his initial review, change his statement in his blog by stating that our findings were nothing new and that he has been researching the same matter for some time as well - and then change it again by stating in his article that we uncovered that issue.

It is beyond my comprehension that you would complain and interpret our defence as whining. The result of this issue here is a good thing in the end for all of us. Perhaps this is what we all should care about and not who had one more piece of data that explained something along the way.
February 15, 2006 4:49:50 PM

Quote:
So to recap, TG spearheaded the discovery, which is good. But then failed to properly distribute blame where it was clearly due in an attempt to be, what, unbiased? And Anandtech's article was clearer, better written, and went into considerably more technnical detail than TG's equivalent follow-up article.


Let's take a look at what EXACTLY was said in the original article.

Quote:

Many things are possible, but if we're willing to believe our well-informed sources, a problem with the USB 2.0 controller may be at fault. We can neither predict nor yet prove if this behavior will also apply to production notebooks in this series - remember, ours was a prototype. But we are convinced that Intel and other parties involved will be highly motivated to find and fix this problem as quickly as they can. It could even be the early stepping model of the chipset (NB Rev 03, SB Rev 02) that Asus built into our test machine.


Quote:
Then too, there are rumors that earlier versions of the ICH7M had problems with the USB 2.0 controller that caused excessive consumption of battery power. As the old saying goes, "there's some germ of truth in every rumor." For our part, we intend to dig into a dual core system with an integrated graphics core and a 945GM chipset to help us determine where that excess power is going.


There... you see. THG acknowledges the problem possibly existed before Napa! And this was before Anand or THG did any in depth articles on the issue!

Remember that the focus of the original article was not on this bug, but rather the performance of Core Duo/Napa. What THG discovered was this issue and they mentioned it. (whether it was right or wrong to associate "rumors" with the article is debatable) The focus of the article was to write it and get it out so people can read about how the new Core Duo laptops perform. Instead the focus was shifted to this frigin bug by communities, Anandtech, etc. Had THG squarely focused on this bug in the original article it would have taken weeks to publish. (it took Anandtech nearly a month to write their article after THG published this) What Anandtech did was a little "monday morning quarterbacking" in my opinion.

It's important to remember the intent of the article; Core Duo performance.

Additionally I think there's something to be said for THG working with both M$ and Intel on this. Not once did anandtech ever communicate with M$ or Intel on this issue- so the statement that anandtech had a "better written" article is a little misleading I think. Believe you me, having phone conferences with both Intel and M$ isn't a "one day thing." It probably tooks weeks of communication with THG, M$, and Intel on this issue.

In the end, THG did it the right way. Anandtech did it the fast and easy way- and slandered THG in the process.

BTW, here's the Anandtech news article that's a major point of contention in this whole matter: http://www.dailytech.com/article.aspx?newsid=775

EDIT: posted wrong Anandtech blog article. Can't find it now... lol

-mpjesse
February 15, 2006 6:26:10 PM

Quote:
Second, Anandtech and Dailytech mentioned that we blamed Core Duo. This is plain false and damaging in my personal view. We simply stated that the bug surfaces only in the Core Duo notebook we had available at the time, but that there is credible indication that all notebooks supporting USB 2.0 and including XP SP2 are affected.
Which was barely made clear in the first article. Heck, it's not made much clearer in the second article. The overall feel of the articles both very much imply that Intel is as equally at fault as Microsoft, and that Core Duo is much more of a factor than it really is. (Which, in fact, Core Duo is in no way a factor to this bug at all. It is just impacted more by it, likely because with two cores it's using that much more electricity when it can't enter the proper sleep states.) But you have to pick out tiny statements amongst huge amounts of text just to figure that out. And even then, there's a definate overall feel that just doesn't support those tiny statements. It leaves them feeling more like legal taglines in fine print than the major decisive points that they should be.

From what I've read from AT (which I admit isn't much at this point in time, so I may be missing something), I've got to fully agree with them setting these records straight, because looking at TG's articles you really only get this opinion if you read very carefully and go through with a fine toothed comb. Where as AT takes the effort to make these important points quite clear.

Is it damaging? Sure. Is it false though? Not in my opinion. From what I've read so far, it's quite deserved.

Quote:
However, we do question Intel's approach to not force Microsoft to release a bugfix earlier.
So tell me, what exactly do you think Intel could have done differently that wouldn't have resulted in costing them dearly? How do you think it could have been handled better? It's easy to question something and put someone down when you don't have to give any answers. I'm not saying that you don't have the right to question. Heck, what would be the point of life if we couldn't question? But from a professional, to weigh so heavily and yet not suggest something better? It's pretty pathetic IMHO. Us unimportant little readers can get away with that kind of stuff because we're not the professionals. We're not supposed to have those kinds of answers. But from the 'professional' TG editors? Sad.

Quote:
In our opinion, such an article can only be released, if there is a solution for every reader. Simply saying that there is a problem, does not help very much.
Thank you for just proving one of my points so well. So instead of providing a solution that will help a large number of people to minimize their problems, even if it isn't a complete or perfect solution, you'd rather just leave everyone to suffer without helping at all? That's like not feeding hundreds of starving people, even though you have a pantry full of food to give them, just because a handful of them might have food allergies. God forbid you just help people as much as you can as quickly as possible to minimize the overall suffering while making any dangers clear and obvious so that those suffering can make their own educated choice on whether or not to take a risk instead of you making that choice for them.

Sorry for using some harsh words here, but I am just getting a bit tired of TG editors pretending to care about the end consumers when all that they really seem to care about is their own image.

Quote:
It is beyond my comprehension that you would complain and interpret our defence as whining.
Perhaps your ability to comprehend is a part of the problem that I see then? I'd love for you to prove me wrong on that. I'd love for TG to put its money where its mouth is for a change. So far though, I'm just not seeing it. I used to love THG dearly, but it has really changed, and not for the better. TG has gotten whiney, untechnical, money/hit-oriented, and image-proud. It's a far cry from the investigative, helpful, technical place that it used to be. And so far all that you've done is reinforce that.

Did Anandtech and Dailytech take undeserved pot shots at TG? I don't know. I don't read blogs. And as I said, because I'm so much of a TG fan, I only have the one AT article to go by. I haven't kept up with AT's articles. So maybe I'm missing some well deserved reason to bitch. Even still, what I have seen from TG is not impressive, and most certainly not rising above all that to be professional. Hell, TG removed an op-ed rant because they so clearly jumped right into the mud. Instead of proving that you're human, leaving the op-ed piece up, and editing it with an appology for it, you just take it down and pretend that it never happened. Because the gods forbid you be human, or worse, be humble.

You really deserve a pat on the back for finding the bug, for working hard at trying to find answers (even if you haven't really provided any of those answers, nor answered any of the other questions that were raised), and for forcing M$ to get off it's fat arse and do something for a change.

But you also deserve such a kick in the arse for everything else that it really more than washes at this point IMHO.

Where as AT may deserve a kick for sniping TG more than necessary and for not having the 'complete' answer that you don't have yet either, but the pat on the back that they deserve for answering some of the more basic questions, helping sooner, and just in general doing better writing on it IMHO puts them in a better position than TG is at this point in time. Which is sad, because I'd really like to see TG on top.

And yes, I'm probably being overly harsh at this point. But if someone doesn't say it, if someone doesn't take you to task for it, then you're just going to be all Microsoft about it. :lol: 

How's that for fair play? :wink:
February 15, 2006 6:31:07 PM

Quote:
What Anandtech did was a little "monday morning quarterbacking" in my opinion.
And I don't argue that. I completely agree there. But I disagree pretty much everywhere else, as I've clearly stated in my previous post, so I won't go into a second time. IMHO both sides of the TG/AT kerfluffle here deserve both a switch kick in the pants and a pat on the back, each for their own reasons. But when I tally up the pros and cons on both ends so far, I really have to say that AT comes out smelling better than TG. But I also deeply stress the so far part, because there's no reason that TG can't still come out on top. It's what I'd like to see.
February 16, 2006 1:30:12 AM

I like ur new signature BTW.

-mpjesse
February 16, 2006 1:33:55 AM

You really like italic. lol
February 27, 2006 6:38:38 AM

Quote:
And yes, we intentionally left out a link to Anandtech. People who haven't read the Anandtech article could care less who wrote about it. And people who know about the sorry - garbage - that was published on DailyTech and the misrepresentation of our findings on Anandtech, they know anyway where to find it. There's no need to escalate this issue from a pre-school to a kindergarden argument.


uhh... Let me quote your EIC, two days after your post:

http://omid.tomshardware.com/2006/02/kubickis_hormon.ht...

Quote:
Kristopher Kubicki, Editor in Chief of gossip rag Daily Tech, is about to experience hormone replacement therapy, according to rumors out of nowhere. Usually, I wouldn't give credence to this kind of talk, but Kubicki has proved himself a master of the untruth, and the gods of innuendo want to have a chat with his soul.


Wolfgang, didn't you used to write for Die Spiegel or something? [/quote]

Kristopher
February 27, 2006 11:30:10 AM

FYI the registry patch works on desktop PC's also, mine being the Athlon XP Barton running DFI Lanparty KT400A motherboard.

With the patch CPUidle extreme software shows drop in CPU core temperature back to normal with USB plugged in. Before, the CPU would run around 12% CPU cycles with the USB plugged in no matter what.
!