Power Problems with new Intel Platform

[tripo]

http://www.tgdaily.com/2006/01/28/to...r_drain_issue/


More damning evidence of how the worlds largest Chip company goes about its business !! This shows a total lack of respect for the consumer,pushing buggy and flawed products.

[Áedán]

I'm a little confused - if both Microsoft and Intel are saying that it's an issue in Microsoft's Windows ACPI driver, how does that make it an Intel problem? After all, the article you linked to is titled "Microsoft driver bug saps Core Duo power"!

Besides, what silicon hasn't been buggy?

[hoinar]

Originally Posted by tripo

http://www.tgdaily.com/2006/01/28/to...r_drain_issue/


... worlds largest Chip company ...

Does Motorola make notebooks?

[tripo]

My Bad,Should have been one of the worlds largest chip manufacturer( but does it matter ).
No one knows if it is a fault with either Windows or Napa/Core Duo for sure,but given that it has surfaced with the new notebooks,i believe that the blame can be laid at Intel's doorstep for now.( Microsft may be a scapegoat
And a company like Intel is responsible for testing it's products before release and this is not a minor bug, given the widespread use of external USB devices and the effect on battery performance given that it is a mobile chipset.

P.S. Maybe their testing is still following the antiquated procedures laid down before the widespread use of USB devices.

[Áedán]

Intel has never been responsible for issues within Microsoft software. If there is a bug in the Microsoft driver, then it is a Microsoft issue, even if the bug only turns up on the new platform.

Similar issues occured when Intel released hyperthreading - There were bugs/assumptions in Microsoft's task scheduler code that only came to light when Intel released their hyperthreading CPUs. Microsoft eventually fixed them.

So, whilst Intel is responsible for testing it's products, it is not responsible for correcting bugs within Microsoft's drivers. The point is made that the Microsoft USB scheduler keeps prodding the USB device, preventing it from being able to go into power saving mode. That rather points at an issue with the Microsoft USB scheduler rather than anything that Intel has done.

If Intel finds a bug in MS's code during testing, what are they supposed to do? Hold back their product and wait a year or two for MS to fix the bug?

[hoinar]

Mate, if you read carefully the article says that's a pretty old problem, known to MS, an issue that from various reasons wasn't solved:
"...The delicate issue about this bug, provided to Microsoft partners under confidentiality guidelines as Knowledge Base article KB899179, is the fact that it has been known to be a "problem" at least since 12 July 2005. It is unclear at this time why Microsoft has not issued a patch or an advisory the public to fix or explain this bug. So far, Microsoft has not responded to our inquiries and we are still waiting for a comment on this power drain issue..."
Whatever Intel could find, I don't think they're willing to fix an important MS driver bug, my guess is that they reported it to MS and now the ball is in Redmond Park...

[tripo]

Looking at the same article

Intel has asked Microsoft "to fix the registry to make the problem go away." In the meantime, he said, Intel is working to fix the issue by itself: "We are doing additional work on our end to see if there is anything we can do to overcome the same challenge Microsoft is having

1) Since information about this bug was availabe to its partners of which I
suppose intel is a partner, they( Intel) did not take that factor into account. I do not expect them to change their entire chip layout or postpone the launch for a year. What i do expect them to do was to try and get MS to fix the bug ( they might have had like 3-4 months to do so after initial production for testing ).

2) Secondly around the launch of CoreDuo, no mention of this was made until
it was found by 3rd party testing. Just letting the consumers know about
this defect and about their efforts to get it rectified(and a temporary workaround)
would have helped.

I am not trying to provide justification to absolve MS of their crime, as MS's
attitude towrads bug fixes is well known. I only expected Intel to do better.

[Áedán]

Given that Microsoft have allegedly threatened Intel with crippling Windows when running on Intel chips if Intel continued along their software route, I'm not sure I'd consider Intel to be a "partner" in anything but the loosest sense.

In any case, Intel would not necessarily use "Windows" as a validation case. It is too poorly defined to validate hardware - instead, a dedicated hardware validation suite is used that goes through and checks every aspect of the hardware.

My suspicion is that MS said they'd fix the bug, and never got around to it. Having seen bugs in software reported (by the company I work for), and literally having to wait over a year to see them fixed (In this case, security bugs), I would not be surprised if MS put it down as a minor issue and did not consider it a priority.

In terms of Intel - what can they do at that point? Wait on MS who may or may not fix it at some point? Or perhaps, they release it, and then subtly suggest to one or two sites that they might test some things in order to put public pressure on MS? After all, we are talking about prototypes here, not production systems.

[tripo]

Do you not end up testing your hardware with the software that it is supposed to run as Windows XP makes up a large share of users ?( i am a bit naive when it comes to testing matters ..and this is just for my knowledges sake) You would use a hardware validation suite first and then test it out using standardised/known additional equipment( power supply,graphics,hard drive) running Windows XP.

The only thing we can hope for is that there is atleast some sort of an improvement in the attitude and workings of the companies involved after this incident.

[Áedán]

Which version of Windows would you use to test with? Windows 2000, SP1, SP2, SP3, SP4, XP, XP SP1, XP SP2? What hotfixes would you test with? What drivers would you test with, what graphics cards/network cards/BIOS/etc?

Why Windows XP, when so many corporates use Windows 2000? They're by far the biggest market.

The whole point of the hardware validation suite is that you can ensure that the hardware works exactly as it's specified to work, without having to worry about the 10,000,000 permutations of software/drivers/BIOS.

[tripo]

I know that they are a lot of versions,variants of Windows lying around ..but you would not expect them to test it out for all the versions..only the supported ones with all the fixes applied ( how many does that make.) and the intel core duo .This is not supposed to be an exhaustive test,something of an extra just to make sure and given the resources it has well i donot think it could have been that difficult.
Then if the drain had turned up on older software or unsupported software, it might have been justified for them to was their hands off it .

[Áedán]

The list of supported ones with fixes applied makes: Windows NT4 SP6, Windows 2000 SP4 (and SP5 when it comes out), Windows XP, XP SP1, SP2, XP MCE, XP Tablet PC, XP x64, 2003 R2 SE, 2003 EE, 2003 DE, 2003 WE,2003 SBS. None of the x64 OSes would be supported by the Core Duo, but might be supported at a later date.

Well, all of the ones I posted up there are supported one way or another. I am aware that there are a number of products being sold on the market that have Windows NT 4 embedded in them. You wouldn't necessarily realise that NT 4 was running on them devices though. Online support for Windows NT4 doesn't disappear until 2007.

Any idea how long it takes to validate a single OS, with a single driver configuration?

[tripo]

Been trying to edit the previous post but timed out ahh....
My point was not really valid in the sense that you would not expect a hardware maker to test for software compatibility, but i was just hoping they would to a minimal extent.
And how long does it take to validate a single OS with a single driver ? ( i dont know..youll have to tell me )

Certainly the lack of the fix to what is supposedly a simple registry error will give license to conspiracy theorists [ me ]to claim it is a problem with the chip itself and not Windows XP SP2

Let us see how the story unfolds ...

[Favu]

This is the same chip used in the new Intel "Mac Books" right? Does it have the power issues there?

[tripo]

They are still trying to verify that fact !!!

[Áedán]

Well, last company I knew of that was doing a full and complete validation of a server OS took nearly two years over it. They were validating that NT4 would be suitable, in terms of stability and reliability, for them to move their banking applications on to.

[weldzilla]

just another uninformed person wanting to trash intel cuz they think they will get a "Cool" response and it is the thing to do. Makes me ill quite frankly.

WZ

[Seperah]

I had a feeling that was coming from somebody sooner or later...

[tripo]

I did not expect you guys to take this so seriously..was just looking for some light hearted banter and comments...If i had wanted to trash Intel,there are a lot of other real things that they havent done right!

Maybe i should have posted it under a different heading or topic

[Áedán]

Originally Posted by tripo

More damning evidence of how the worlds largest Chip company goes about its business !! This shows a total lack of respect for the consumer,pushing buggy and flawed products.

Those are serious allegations you made against Intel, with nothing to back them up with. Remember, when you post some here, you are publishing your views in the same way as if you had written a newspaper article and had it published.

Using phrases such as "damning evidence" and "a total lack of respect" indicate that you are being serious, not lighthearted.