Vista sleep woes- an update

16 Comments

The problem sleeping and resuming that my HP 2710p Tablet PC suddenly developed has been preoccupying my time over this long holiday period and has resonated with a lot of readers.  Some readers agree with me that Vista is to blame for my problems and others do not agree with that and believe that something I did caused the sudden onset of the problem or that some of the device drivers are to blame.  There is no answer to the cause of the problems, nothing was updated on the device nor did Vista apply some update that changed the OS.  I am very familiar with battling flaky Windows installations and I can swear that nothing was changed by me during the timeframe when this ugly problem arose.  I have spent a lot of time going over system logs to see if I can spot something that changed to no avail.  Vista doesn’t think anything changed yet my sleep problems started happening all at once and consistently so logic dictates that something changed without my knowledge.  This is in my opinion a big problem with the Windows ecosystem as it is not uncommon in my experience.  Read on to see what I’ve done so far to try and resolve this new issue.

Good friend Ed Bott, who knows more about Windows than anybody on the planet, contacted me and had a few suggestions for me to try.  I appreciate Ed’s interest in helping me resolve this and I took a lot of his advice.  First up I did a Vista system restore to the day prior to this problem occurring.  The logic here was solid and since the problems I am having started all at once the assumption that something must have changed in the system is a logical one.  The system restore went as planned and I watched the system for the next 24 hours to see if the problem went away.  I tried to force the problem to come back by sleeping and waking the system a lot, both in and out of the dock, and it worked without problem for most of a day.  I was feeling pretty good when I had another reboot in the dock when the device was going to sleep.  I was obviously disappointed at that.

Ed pointed out that one of the biggest culprits with docking and sleeping problems is the video drivers.  I was still using the video driver that shipped with the HP since I was having no issues prior to this situation.  HP hasn’t updated the video driver for the 2710p anyway so I wouldn’t have considered it a viable solution but Ed convinced me otherwise.  Vista actually offered a new Intel 965 video driver update via Windows Update recently so I went ahead and applied that update.  I watched the system for a day and had no problems until I popped the HP out of the dock to take it out with me and got the Vista Sleep of Death.  The device woke up from sleep mode running normally except the screen was dead.  The only recourse was a hard reboot since you can’t do much with a system with no active screen.  I was totally bummed out as I had only experienced this once in the months I’ve used the HP even though many have reported it under Vista.  Time to try the next thing.

Ed also had pointed out that HP has a BIOS update for the 2710p that is supposed to resolve sleep issues so next up yesterday I applied this update.  It was dated December 7 and I had seen it before but not applied it when it was released because at that time I was having no sleep problems and I didn’t see any reason to rock the boat as at that time the 2710p was running better than any other device I’ve used with Vista.  Since I am having the problem now I updated the BIOS and watched the HP for the rest of the day.  It performed flawlessly until last night when I was using it in slate mode in front the TV.  When bedtime approached I hit the Power button which put it to sleep and then popped it into the dock for the night’s charging.  It was still in slate configuration when I did this and after popping it in the dock I noticed that it was not sleeping, the hard drive was thrashing and the power light indicated the system was up and running yet I had the Sleep of Death.  The screen was off and nothing I did would bring it back as usually happens in this situation.  I had to hard reboot and waited for the reboot process to complete which took over five minutes.  When the system settled down I hit the Power button and put the HP to sleep for the night.  This morning when I awoke it was still asleep and had not rebooted during the night and the system was ready to go.

The current situation is that it doesn’t appear that my problems are resolved with the system restore to a time before the problem started happening, a video driver update and a BIOS update.  The Sleep of Death has happened twice during the past few days, something that had only happened once months ago.  I will have to watch it now to see if the HP starts rebooting in the dock during the night.  I should make clear that I realize the problem may not be the fault of Vista directly.  It could be the dock, the video driver or any other device driver that is now corrupt somehow.  I will state emphatically that I do believe ultimately it is the responsibility of the OS to prevent this sort of thing from happening.  It is great that Windows is an open OS in that it supports a lot of different hardware from many vendors, as opposed to the closed (and expensive) hardware that Macs employ.  But if the OS, in this case Vista, can’t prevent a solid running system from suddenly developing a major flaw like this then who wins?  Is the openness worth the trouble?  If you have run Windows devices for a few years how many hours have you devoted to dealing with problems like this?  How many times have you had to rebuild a Windows system from scratch to deal with weird problems?  It seems that the more things change in the Windows ecosystem the more they stay the same.

16 Comments

Marco

have you disabled and reenabled after a scandisk the hibernate file and the swap file of windows vista ? if the problem is a bad sector in hibernate file then this could resolve it

James Bailey

I might have come along late to this discssion but how have you determined that the problem is not hardware? It sure sounds like a hardware problem. I had an early MacBook Pro that randomly restarted itself. I didn’t assume it was software for more than a few hours because the problem was truly random with no pattern that I could discern.

I know that Vista has less of a reputation for sleep reliability than OS X but still it seems to me that random problems that start without any software changes point to hardware issues.

bluespapa

We’re used to thinking problems are somebody’s fault in something manufactured (as opposed to the snow covering my car that you Texas boys don’t have to deal with). We’re also used to dividing up the teams in ways that are no longer accurate especially in computing. Used to be the IBM software guys would blame the hardware, and the hardware guys would say it’s a software problem.

Now the team is no longer parts of a company, they are teams that stretch across companies. I certainly know less about this than you who work in IT, but the results are dependent on people around the world these days, in companies that do their part. Microsoft decided they want to be the center of the software action, and get disproportionate blame because they are coordinating the teams in other companies as well by the structure they provide. Except that, as Josh points out, Intel has their role at the center. There’s no center, really; more like motion that everyone is part of.

And unfortunately that points to how complex troubleshooting it all is. Could be this, could be that. For starters.

My experience with Apple isn’t much better, fwiw. My experience is also that James is pretty insightful, and that Josh is about the most responsive software company I’ve ever dealt with. But Microsoft wrote most of the software on James’ computer, and gets to hear most about James’ problems with it.

Bashing Vista? James has been ecstatic with computer (software, hardware), was one of the earliest of early adopters of Vista beta, loves to show how cool his stuff is, and is reporting on his perceptions of the conversation of this set of problems. He’s steered more biz to Microsoft than their average marketing exec. Partly by reporting on his experiences and problems, as well as pleasures.

Woadan

When I used to work for the big telco, users blamed us, they blamed Microsoft, or they blamed Darth Vader. (The voice actor for DV was James Earl Jones, and he was a spokesman from the Bell Atlantic days.)

When it isn’t working, most users don’t care. They just want it fixed. They don’t want to wait until Monday let alone next week. If a Mac user isn’t experiencing it (and given they are 5% of the market, problems don’t echo as loudly, even when a fair number of users are experiencing the issue), then they want you to be more like a Mac.

The bottom line is that to end users, the vast mojority of whom are trying to get to http://www.girlygirlygirls.com/, it just needs to be like a toaster. Put your bread in once, and if it’s too dark, adjust, pop in bread again, and if it’s too light, adjust once more, and then… Ah! Nirvana. (At least until the toaster dies.)

Yes, I know a computer is vastly more complex than a toaster. But the popular media and news outlets have convinced us it can be that easy. And in the always-on, give-it-to-me-yesterday world that we live in, we want it. Nope, sorry, we NEED it!

Happy New Year everyone!

Woadan

Josh Einstein

Meh, maybe you ought to look and see how widespread Intel stacks are. There’s more Intel CPU/Video/Network stacks in operation right now than Windows Vista. And guess what, I’ve had problems with XP resuming from standby every now and then too.

In fact, my HP2710P has been running flawlessly as well and only recently I’ve been experiencing infrequent bluescreens since popping in a Verizon EVDO card if I leave it in while I put it in standby. Maybe that’s Vista’s fault too. Maybe it’s Vista’s fault that my hinge is loose. Call it sandbagging, call it whatever you want. I call it being rational.

Oh and James, “bash Vista” were your words before I used them. :)

Meh

Josh is also sandbagging. Given how widespread the sleep of death problem is across a number of different laptops, it’s entirely fair to suspect the common element in these situations, which is, unfortunately, Vista.

Kevin C. Tofel

I completely understand Josh’s reasoning as well, but I think there’s a perception management issue that’s worth noting. While we and most of our readers are what I’d generally classify as “tech saavy”, the majority of consumers are going to perceive this type of problem as a Microsoft issue. They’re not going to zero in on the component manufacturers and look for solutions in the drivers. They’re simply going to get fed up because “Vista is causing the issue”. I’m not saying that Vista itself is the problem here, but the perception of the consumer is going to be that it is. That’s a totally different issue than the technical problem of course, but it’s no less of an challenge in my book.

Bill

Just to keep this from a full on Vista bitch fest- I have had a HP TC4400 w/ Vista for about 6 months now- I use the dock and each day I dock it at the end of the day, press power and it goes to sleep, in the morning it wakes up fine. I also have 2 other vista laptops that run flawlessly. I hope you can get your problem resolved- but I think the majority of users are doing fine.. it is the small percentage of those who have issues who make the most noise.

James Kendrick

Josh, I understand what you’re saying, believe me. I don’t intend to “bash Vista”, I am simply reporting what my system began to do with no provocation. I know that given my experience with Windows I will eventually resolve this, with or without MS help. What drives me to report this stuff is how the many, many users who don’t have that ability to resolve something like this who will be screwed when they run into problems like this. Like it or not the OS controls the environment (or should) and to the end-user like those I just mentioned it doesn’t matter if a driver suddenly exhibited problems or the like. There is no logic behind a system that shut down fine one night and then had a consistent problem the next morning, no matter how you slice it.

James A. Morman

I have a 2710p also, and I have it set to go to sleep after inactivity so it goes to sleep at least 8 times per day. I have not had this problem. The only sleep problem I have is when it falls asleep in portrait mode it re-awakes in landscape. I have to hit the little rotate button on the side to fix it everytime.

Josh Einstein

James, Except that it is not an operating system’s job (nor within its power) to make sure that each component of the system does its job properly. An operating system’s job is to make it so that software has a unified API for interacting with the various hardware devices. Mac OSX can’t keep the video card or network card from bombing a machine any more than Windows can.

What you are expecting Windows to do *is* theoretically possible using the same kind of techniques that Microsoft .NET or Java uses to provide a fully managed environment. The only problems are 1) no hardware maker in their right minds would go along with it and 2) it would be slow as balls.

You closed your last post about this by saying “Thanks for nothing, Vista”. How about “thanks for providing an automatic feature that reports these errors so that people at Microsoft can find and fix bugs that are sometimes Microsoft’s fault and sometimes other software or hardware makers’ faults.”

It just really bugs me that you and alot of other bloggers jump on the trash Vista bandwagon when you even admit that you don’t know if Vista is really the problem – all without giving any credit for the great lengths Microsoft is going to to make these problems go away. Don’t think for a second that many of those mysterious Windows updates that suddenly make things like sleep/resume work better are all Windows bugs that were fixed. There are lots of:

if ( crappyDriverInstalled )
FixStackCorruption();

Workarounds floating around just to make end users happy because driver makers are not always so quick to fix problems because they know users will just play it safe and blame Vista.

James Kendrick

GoodThings2Life, good points about the power management and something I had meant to do but forgot. I have now turned off PM for the 7 USB root hubs showing in Device Manager and the two network devices as well. Any of these could easily be the culprits who are waking the computer in the night and causing a crash so here’s hoping. Thanks!

Meh

To answer the question: In the years I used a laptop with Win 2K and Windows XP I had sometimes to adapt usage of sleep/hibernate to some idiosyncracies of the lappy in question, but overall I never bothered to do a reinstall because things mostly worked.

On my (nearing 9 months old now) Vista Sony I am considering doing a reinstall, not because of the sleep problems outlined above (I think those problems are a bug in Vista, personally) but because of intermittent BSODs, which is a definite downscale in quality from W2K and XP.

It’s a shame, because Vista, Office 2007 and particularly OneNote 2007 have been real improvements for my working otherwise.

GoodThings2Life

I have found, over the years, that vendors like HP and Gateway, etc. are not in a very good habit of keeping drivers posted on their site up to date.

Many of the Dell’s where I work have also started randomly blue-screening on people, and I was able to track down that a Windows Update broke some driver compatibility. The solution? Update ALL drivers regardless of whether they were previously working. I agree with Ed– video is the biggest culprit, but I’ve seen a lot of Intel chipset drivers cause this type of issue as well as network drivers.

I use http://www.driveragent.com for checking to see if new drivers are needed. You have to be careful about sound drivers since even though they may use the same chipset for sound, that chipset driver is heavily customized for each vendor. Otherwise, most things are generally pretty accurate. Just be aware that I’ve also seen Intel chipset drivers skip installation on IDE/SATA controllers and USB controllers, so you may have to manually update those from Device Manager.

OK, last comment, though I’m sure you’ve seen it before– in Device Manager, verify that the Power Management settings for your network/mobile devices are unchecked and do the same for USB Root Hubs. This has always been a huge “sleep” breaker in my experience.

In any case, good luck on resolving your issues. I look forward to hearing the final solution. :)

Meh

The depressing thing for me is that my old Sony laptop with XP had the odd glitch with sleep/hibernations, but the new one with Vista seems to have many more, particularly Sleep of Death. I’ve only had Insomnia happen a couple of times, but now I am paranoid so on the move I always try to make the time to wait until the hibernate has switched power off, because I don’t want my laptop bag to catch fire…

Scott_H

Now that I think about it, the last time I actually had to re-install an OS due to corruption was about 6 years ago. My w…uh, someone turned off my laptop without shutting Windows 2000 down properly. It wouldn’t boot up at all, so I had to re-install it. Every other installation of Windows I’ve done since then has been “voluntary.”

Strange, huh?

I’ve commented at length in other posts here about my Vista problems — mostly BSODs. I never did find out what the culprit was, but I suspect it was either video or wireless adapter driver related. At first, I used to get BSODs just at shutdown. Later, it started happening upon waking from Sleep.

I ended up replacing Vista on my laptop with XP (voluntarily, of course). Over the weekend, however, I installed Vista in a VM on my laptop — just for grins, mostly, but also to keep up with the “technology.” I’m using VMWare 6 and haven’t had any problems so far in the very short time I’ve been using it.

One obvious, but probably unhelpful, suggestion would be not to use Sleep — just do a complete shutdown, especially if it’s in the dock.

Comments are closed.