I notice that there's lots of questions involving people asking for drivers for specific pieces of hardware. Invariably these questions are rather specific/localised and aren't really all that useful unless you have the very same piece of hardware (and unless it's an obscure bit of hardware, the first port of call should be the OEM).

I'm wondering if it's warranted to close unanswered questions of this sort (such as this) when they float back up - I doubt anyone would wait weeks or months for an answer, or to close these questions for being too localised.

I believe the same rule of thumb for shopping questions should be valid here - that unless there's a good reason that the regular ways of finding a driver don't work, asking here smacks a little of QA vampirism.

  • @Idigas strictly speaking in lolspk, it's "I CAN HAZ DRIVERZ?"
    – Sathyajith Bhat Mod
    Commented Dec 7, 2011 at 5:24
  • 1
    fixed the title ;p
    – Journeyman Geek Mod
    Commented Dec 7, 2011 at 5:37
  • @Sathya - Perfect :)
    – Rook
    Commented Dec 7, 2011 at 12:12
  • I support this.
    – Tim
    Commented Dec 8, 2011 at 16:01
  • 2
    I can't believe we are correcting people on proper LOL. Isn't the whole point of LOL is that it isn't proper??
    – surfasb
    Commented Dec 12, 2011 at 14:02
  • 2
    it srs bzns kthxbye
    – Journeyman Geek Mod
    Commented Dec 12, 2011 at 14:11

2 Answers 2


It really depends on the question. If somebody wants Windows 7 drivers for an 8 year old laptop that hardly even has the specs to support Windows 7, or conversely wants Windows 2000 drivers for a brand new chipset... probably not a very constructive question. But consider this question about finding Windows 7 drivers for a particular Dell laptop. The question has almost 16,000 views, and is one of the highest ranked questions for anonymous user feedback, with +64/-4 "was this answer helpful" votes.

There are about 275 questions with the text "find drivers" in them. Of those, about 120 have accepted answers, which is almost 70% success rate with driver questions. This tells me that these questions are useful to somebody, but a lot of them are indeed one-off sorts of things - only about half of those with accepted answers (40% of the total) have more than 200 views, and a mere 15% have over 1000 views.

So there's some data to consider. What should we do about it?

One thing to do would be the "google test". If you can find the drivers in 5 minutes by checking on google or the OEM's website, then the question should probably be closed as "too localized", kindly give the user links, and remind them that they can find things themselves as easily as we can.

If the google test returns a lot of similar "I can't find drivers for x, help" sorts of pages - especially if said pages are on forums - then I think we should keep the question. Even better if we can solve the problem! This is a case where we can actually be a useful resource, rather than being redundant.

Another thing to point users to is the Community FAQ on "how to find drivers". The method I describe in that post of looking up the vendor and hardware IDs works very well for almost all computers I've ever had to find drivers for (which is a lot of them).

Basically, "Help me find drivers" shouldn't be specifically off-topic. It's a valid computer question. But we also aren't a driver database. For specific "help me find driver" types of questions, it has to be a judgement call on whether the question should be kept around, or is too localized to be useful. If you aren't sure, then feel free to discuss with other people in our general chat room or the ask a moderator chat room. Try the google test. With this specific case, I think we should error towards helping people rather than shunning them, but that can include pointing them towards our existing resources (like the community FAQ mentioned above).

  • 2
    Why wouldn't it be constructive? The fact that drivers for old hardware do not exist, does not make the question any less ... after all, if the author knew they don't exist he wouldn't be asking. Generally, I think I agree, but am wary to say so. With that last paragraph you've made it hard to understand whether people can or cannot, in the end, ask this sort of questions. Keep it simple! Say yes or nay, don't create separate rules for every possible edge case you may encounter. The good moderators are ones that moderate in a way that makes them superfluous.
    – Rook
    Commented Dec 7, 2011 at 3:49
  • With the Google test, don't you mean general reference, not too localized? Commented Dec 9, 2011 at 4:01

I vote for keeping them alive, for several reasons:
Invariably these questions are rather specific/localised and arn't really all that useful unless you have the very same piece of hardware (and unless its a obscure bit of hardware, the first port of call should be the OEM).

That goes out for pretty much everything out there, does it not? So, it is not much of an argument.

Second, there is a lot of hardware manufacturers out there with less than poor websites in regard to driver support. Often original drivers for some (motherboards and integrated sound and graphics cards come to mind) components cannot be found at all, so people use drivers from some other components, that work equally well. Such advice can be a lifesaver, and save several days of painstaking experimenting.

Third, drivers are software in direct relation to hardware, therefore perfectly ontopic for SuperUser, whose target domains are primarily those two.

  • 1
    If there's a clear, useful answer, and there's an attempt at research, its fine. Research does not simply mean a hardware-listing dump. Also, to some extent, the main target for me are stale driver questions, as opposed to fresh ones. I have a driver-ish question currently floating around the SU system myself. Its clearly a judgement call, but its really a matter of should we let them stand, or use discretion in such cases. I think if an answer is findable in 5 minutes on google, or its a question with no answers + is churned back up by community, its less likely to be useful to anyone.
    – Journeyman Geek Mod
    Commented Dec 9, 2011 at 0:01
  • @JourneymanGeek - You contradict yourself. First, you say - "if there's an attempt at research it's fine." But what if the author couldn't find the driver (after all, that's why he's asking the question!). Then you say, a "question with no answers is not likely to be useful to anyone." That doesn't make sense! For that is the best question by definiton - the one we don't know the solution for yet. I'm not saying the answer couldn't be "driver doesn't exist", but prove it. Also, I don't like striving towards politics where a
    – Rook
    Commented Dec 9, 2011 at 0:24
  • lot of "judgment calls" are being made. Make what can and what cannot be asked clear and simple. Avoid judgment calls, for that is simply another wording for "where a few users will decide if a question is okey" (one moderator suffering from close-mania comes to mind), and that by definition violates the rule of this being a "community site" (where the majority decides).
    – Rook
    Commented Dec 9, 2011 at 0:25
  • Well, voting to close is somewhere where a few users decide if a question is fine - i'd just rather have something at MSU that says if something is or isn't fine. With the question @nhinkle gave, the question included what and where the user searched, and where he looked. . An ideal question asking for drivers would include 'i've looked at the OEM site and found/didn't find this this and this, and i've also tried the generic drivers, which failed with Y error message. I'd rather ask here and get a community consensus of some sort (as i have in the past) than go on a one man VTC spree.
    – Journeyman Geek Mod
    Commented Dec 9, 2011 at 0:45
  • @JourneymanGeek - I see less and less questions closed by five users. Research: Yes, I consider that an okey effort from the user.
    – Rook
    Commented Dec 9, 2011 at 0:52

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .