OSNews

I was a bit confused by your post, thinking "Why would you be swearing bullets if your PC rebooted without failure?" Then I realized you must've meant "sweating bullets" rather than "swearing bullets". lol :p

Edited 2007-08-15 20:48

Yeah I read it three times and still didn't get it, so I moved on.

I got it first time. Maybe I am confused ?

Lol! It's a shame I can't edit the title, would be nice to be able to fix that.

Although, come to think of it, after all the time I spend restoring my system, analyzing the kernel dumps and looking into possible solutions, 'swearing' might be a more literal portrait. ;-)

How do you know they're badly tested? Unless you're part of the testing team or know exactly how it's done, you're really in no position to make such assertions.

Never had problem with critical updates, but recommended updates often mess things up. Wireless network disappear, usb ports missing etc., and its not just once and not just on one type of machine, thats why I always told people to disable automatic updates and use autopatcher instead.

Not a windows user myself but people come to me for windows problem

i think it's because some of the updates broke his windows installation in some way.

In the same way i know that badly coded applications which crash and generally don't work properly haven't been properly tested, without having to be part of the QA team.

Or maybe his computer is broken... Or overclocked... Or has some unwanted programs that got installed when he was visiting bad sites and clicking OK to anything that's presented to him... etc.

While bad QA certainly can lead to trouble, trouble is not a certain sign of bad QA.
Unless of course you subscribe to the notion that QA should also test on broken/overclocked/spywareridden/etc. configurations to make sure it works in these settings too.

Good point but I can assure you that my system is not overclocked or spyware riden. Neither do I install everything I can. I have a lean, stable self built system which happens to have crashed so badly after three separat updates that I had to restor my system fron backup.

I have a stable video card and Windows XP installation, but for some reason I cannot upgrade my video drivers beyond 81.98 - the screen just goes black upon reboot and no forum help has worked so far.

I could blame nVidia, Microsoft or Leadtek..

Alternatively I could look at myself and ask what I may have done. The problem does happen with others, but it also doesn't happen with hundreds of thousands more, so who's fault is it?

Probably nVidia... Especially if it's an older graphics card.

Some time ago I had to install old TNT2 cards on a machine...
Went to nVidia and downloaded the drivers that was supposed to have support for that old cards...

Installer couldn't find a compatible card.
Checked through the INF and discovered that there was NO TNT2 entries left in there.

I ended up searching around for older drivers to get something that would install at all.

It's just a 6600GT.

I'm just saying it could be a lot of things, and not necessarily MS fault (though they do deserve blame for some stuff).

It's pretty obvious that if the same set of patches fail on many systems there is either a problem with the testing or a problem with the QA or both.

Or it could be due to a real weird software/hardware combination - it isn't as though Microsoft hips patches that a broken deliberately for shits and giggles. They test it as much as they can given the environment they have.

The simple fact is, if there is a driver or an application which uses undocumented hacks for the sake of so-called 'performance' and the update breaks it, I'd say the onus lies on the software vendor NOT Microsoft. These companies should not be using interfaces which are deemed either unsafe, undocumented or worse depreciated for many years.

"Or it could be due to a real weird software/hardware combination"

Note that I said "many systems". Obviously failure on a smaller amount could very well be due to unexpected hardware/software combos.
If it fails for me on my system does not really say anything about the quality of MS products just like how failure of Linux/BSD/Whatever to work on individual systems doesn't say anything about their quality.

When your perfectly working system no longer boots after three separate updates, and you have to restore from a backup on all three ocasions, the updates are badly tested. It's called logic, you should try it some time.

Three separate updates? Maybe it's just something to do with your computer? It's possible.. but of coarse, lets just blame Microsoft. It has to be their fault. It couldn't be an incompatibility caused by software or drivers on your system, now could it?

We're all so Jedi at keeping our Windows systems in perfect working order...

The bottom line is that it could be either.. it could be Microsoft's poor testing, but it could also be something to do with your system. Without knowing the reason why it happened, it's really hard to just blame either cause.

Edited 2007-08-16 12:26

Look, if it had to do with an incompatibility with one of my system components then I can understand the system breaking, after all there are thousands of possible PC configurations out there. The thing is, I have some very standard components. nForce 590 SLI Asus Mobo, GeForce 7600 GPU, 2 GB DDR2 and 2 maxtor 300gb in a RAID0 configuration.

When 1 optional and 2 recommended updates break my system, from a fresh install as well as from backup,then MS has not really tested these updates enough. The chipset drivers I use are provided by MS update, and each and every time I install one, two or all three updates in whatever order, my system breaks beyond being able to either boot or repair. The kernel dump I get from one of these updates tells me it has to do with not being able to access the HDs so that seems to me a chipset driver issue. Yet, as mentioned above, my drivers are provided by the MS update service.

So, in conclusion. if MS can't test they're own updates for compatibility with each other, then Yes, I blame MS for my system breaking.

I don't know who you have been talking to, but I certainly investigate before I start making claims.

Edit: Typos

Edited 2007-08-16 14:03 UTC

Windows updates used to crash my PC (until I switched to Linux). I remember having to reinstall because of them twice in 3 years. Usually it was a combination of 3rd party apps + updates that triggered the problems.

nLite supports it (I think).

Instructions and some details here:
http://www.msfn.org/articles.php?action=show&showarticle=49

Make that vLite:
http://www.vlite.net/index.html