Should MS be pushing driver updates through windows update?

Associate
Joined
26 Jul 2008
Posts
2,086
Location
Cowley, Middx
This is regarding windows XP, and not 8 (and I think that I should have used forcing instead of pushing).

The long and short of it is that the yellow update shield reared its ugly head in the system tray today, and I thought that it was mucking about and downloading the malicious software removal tool again (it had done this once or twice before) until I noticed that it was taking a bit longer and what I found waiting for me was the following.

nVidia - Other Hardware - NVIDIA GeForce 6150SE nForce 430 (the driver update is from a January 2013 update of the driver)

I haven't bothered with driver updates since I built this system around six years ago (I say that if it isn't broken then you shouldn't try to fix it) and the driver list on the motherboard manufacturers website hasn't been updated since 2006 (for the graphics at least) although there are a few instances where the display driver terminated unexpectedly (the last report of this problem (I only have the event logs to go by) was a week ago).

This puts me back to my original question.

Should MS be forcing this driver update on me, or is there some problem that the update fixes? (considering the recent issues with the display driver, that might be the case but I am a bit apprehensive since the update is for a slightly higher clocked version of the GPU on my motherboard although it did replace the GeForce6100 that my motherboard had)
 
Last edited:
I think they should but if I remember correctly companies can pay Microsoft to get their driver WHQL certified. Microsoft test these and allow the company to put a WHQL sticker on their box or something like that.

Microsoft in my eyes should be doing the driver rollout but that might slowdown hardware production and push out but this method will become more stable and would improve compatibility in my eyes.
 
(I say that if it isn't broken then you shouldn't try to fix it)

there are a few instances where the display driver terminated unexpectedly

Should MS be forcing this driver update on me, or is there some problem that the update fixes?

To be brutally honest, you've probably answered your own question. Personally I'd either install the driver and see if it works, or just download it from the nVidia website.

As for Windows update 'forcing' driver updates, yes it should. If you don't want to install them then you obviously don't have to, but for the majority of users it's probably a useful thing because they'd never update them of their own accord.
 
I don't see how this is forced when driver updates are optional?

And that ugly yellow notification popping up indicates that, after well over a decade, MS is still supporting the OS. It will be missed when it's gone.
 
If they separated driver updates into a different area (a Software Update and System Update or something) then I'd have no problem. But they'd have to do more testing than they seem to be doing in the last few years, as there have been too many updates which break systems — we had the Nvidia driver originally included with Vista not allowing the OS to be installed on some systems* (admittedly caused by Nvidia, but that really should have been picked up on by Microsoft) and even this week we've had a Windows Update (which last time I checked was apparently a Windows Update-supplied SATA controller) corrupting some systems.

I've experienced on two systems now Windows 8 updates that flat refuse to install. They seem to have disappeared now thankfully, but they were getting on my nerves.

Given that you've still got AMD messing around with their mobile drivers and not integrating them all in to one driver, forcing OEMs to distribute drivers, which are inevitably ancient, I'd love for Microsoft to distribute regular driver updates, perhaps allowing optional control panels too, but they need to do far more testing than they have been recently.

On the same subject, as we've discussed many times before, I'd love it if the opt-in Microsoft Update was extended to include popular free software updates, such as for Adobe Reader, Flash, Java RE (imagine that - the prospect of Oracle's reputation going up for allowing such a move!). It would also be great if, when a new version of the Bing Toolbar (or whatever it is), or the EU-vomit-inducing browser choice window is available, that they don't keep appearing in my Windows Update options if I've previously hidden them — I've already told them where to go once, so I know I don't want them — unless they suddenly come with a free voucher for the next version of Windows, I'm fairly sure I won't have changed my mind because the version has gone from 1.1 to 1.2.

Oh and please allow Windows Defender in Windows 8 to install definitions automatically again, as Security Essentials prior to Windows 8 did, rather than displaying a "important updates are ready to be installed" note in the bottom corner of the start screen until I actually go and do it manually. Go forward Microsoft. Forward.

Oh #2, and stop occasionally changing the option from "download updates but let me choose when to install them" to "install updates automatically" randomly when I'm setting up a system for someone. I chose that option, don't try to get clever with me. Either give me no choice, or let me choose... not both depending on which mood you're in.

* My memory is faded after 6 years, but it might have been the final release candidate rather than the gold version of Vista.
 
Last edited:
We don't force driver updates you on, we simply provide them as they are certified drivers that should in theory provide better system stability / the vendor has provided new features with the driver.

This_is_gay - please provide me details of the updates (kb) that have caused corruption on your systems, thanks.
 
This_is_gay - please provide me details of the updates (kb) that have caused corruption on your systems, thanks.

They haven't on mine, thank you. But I've linked the thread to the SATA one which seems to be affecting many systems, and said thread handily also provides a link to the article on your site, KB2823324

Oh, and I wouldn't be quite as patronising if you stopped calling me gay. I'm Gav. As in Gavin. Thank you.
 
That's one of Microsoft's own drivers so Windows Update is the only viable mode of delivery.

I disagree that they issue too many updates that break systems. It's actually very rare.
 
I disagree that they issue too many updates that break systems. It's actually very rare.

Thankfully it is rare. But this is an operating system, not a random program getting updated. I know there are countless combinations of hardware and software, but there should never be a system rendered useless by an update.
 
They haven't on mine, thank you. But I've linked the thread to the SATA one which seems to be affecting many systems, and said thread handily also provides a link to the article on your site, KB2823324

Oh, and I wouldn't be quite as patronising if you stopped calling me gay. I'm Gav. As in Gavin. Thank you.

Sorry Gavin - I didn't read your username properly as was in a rush replying (I did think this_is_gay would've been a strange username choice but this is the web so who knows) - so apologies :) I was only trying to offer assistance regarding the update in case it has been proven / turns out to be a common issue (it's great for us to report on these things internally when they crop up and really helps our testing). I see its the same KB that we've realised a V2 on though so that's cool :)
 
Thankfully it is rare. But this is an operating system, not a random program getting updated. I know there are countless combinations of hardware and software, but there should never be a system rendered useless by an update.

You're right, there shouldn't, but for that to be a realistic expectation Microsoft would have to write perfect software and to test every single known hardware and software combination. Not only is it impossible but in many cases would defeat the purpose of issuing the patch in the first place.

So then, you have to look at how frequent the occurrence is. By your own admission it is rare, so we can all be thankful for that.

Next, it's not as if the update process is entirely out of your hands on x86/64 system. You have control over how updates are downloaded and installed. If the rare exceptions aren't rare enough for you, you can choose to hold off until the updates have had some time in the wild. For businesses and more mission critical networks there is the whole WSUS infrastructure.

Finally, if you accept the fact that no process is perfect you then look at how the vendor responds. Within 48 hours they had both pulled the update from general distribution and issued a fix for those affected.

They aren't 100% right 100% of the time but as far as issuing reliable updates they're one of the best in the business. They have to be, otherwise customers would go elsewhere.
 
You're right, there shouldn't, but for that to be a realistic expectation Microsoft would have to write perfect software and to test every single known hardware and software combination. Not only is it impossible but in many cases would defeat the purpose of issuing the patch in the first place.

So then, you have to look at how frequent the occurrence is. By your own admission it is rare, so we can all be thankful for that.

Next, it's not as if the update process is entirely out of your hands on x86/64 system. You have control over how updates are downloaded and installed. If the rare exceptions aren't rare enough for you, you can choose to hold off until the updates have had some time in the wild. For businesses and more mission critical networks there is the whole WSUS infrastructure.

Finally, if you accept the fact that no process is perfect you then look at how the vendor responds. Within 48 hours they had both pulled the update from general distribution and issued a fix for those affected.

They aren't 100% right 100% of the time but as far as issuing reliable updates they're one of the best in the business. They have to be, otherwise customers would go elsewhere.
But, taking this week's issue, that seems to affect quite a few systems - why wasn't it picked up in testing? It's not a niggly patch which rolls back and keeps trying to install — it replaces a key system file and cripples affected systems.

It's not even the first serious WU issue this year. We also had one in February. Thankfully that one wasn't as serious and didn't prevent users getting into their system, though Joe Average wouldn't have a clue what to do to avoid the issue.
 
I don't know what to say without sounding like an apologist.

If Microsoft introduced a bug or violated their own spec, then it's their fault. If third party software is interacting with those components in an undocumented way (which antivirus is notorious for) then the blame lies squarely with the 3rd party vendor.

The fact is in spite of it being widely reported as a 'botched update from Microsoft' we don't actually know yet.
 
I don't see how this is forced when driver updates are optional?

Then can you explain why the update system downloaded that driver update?

I can at least possibly solve the problem as I went through the microsoft update web interface, and in the hardware optional section where it shows another driver update for the GPU (this time for March although information from the nVidia website tells me that the last update was in February) so it might have been a glitch as there's no mention of it in the hidden updates list (all that is there is the latest malicious software removal tool and that stupid EEA browser choice program) although I'll probably have to update in the near future since the on-board GPU doesn't cut the mustard with Phantasy Star Online 2 (that is as long as updating to the latest driver doesn't cause any issues when I get the new card and that also depends on if a low end GT600 card would work with my motherboard).
 
Back
Top Bottom