I'm struggling to wrap my head around the following thing:

  1. I have a video adapter (NVidia). It was bought new about a year ago and had the newest video drivers at the time
  2. I disabled all the NVidia updates, stopped all their services / consoles / helpers / whatnot
  3. Let's say I play an older (2015) video-game - all good, the device breaks absolutely no sweat over it on best graphical settings, everything is stable
  4. The game doesn't get any updates, nor does any configuration of my environment change.
  5. After some time (around half a year maybe?) the performance worsens - and noticeably so. After a bit more time (a month or two?) the game starts to randomly crash. Luckily, I manage to get the crash call stack pointing to nvwgf2umx.dll all the times
  6. So.. NVidia drivers, huh? Alright. Go to their site - and check if there are drivers update. Indeed there are. Installing, restarting.. TADA! It works - great performance, no crashes.

And this leaves me scratching my head, because I do not understand.. :

  • .. How does the system "know" that the drivers are outdated?
  • .. Even if the system knows the drivers are outdated, why does anything change over time? If "the game is the same", "the system/environment is the same", why does the mere fact NVidia released a new driver causes my hardware (or an existing driver which was working fine just some month ago) to suddenly throw a tantrum on me?

The only logical conclusion I can arrive to here is that the manufacturer somehow can control the device and cause it to .. well, "underperform" to "encourage" the user to update their drivers.

  • 1
    "the mere fact NVidia released a new driver causes my hardware ... to suddenly throw a tantrum on me?" - I think you are confusing correlation with causation. Just because Nvidia has released a new driver (which it regularly does?) and your system also showed strange behavior (which systems sometimes do) at about the same time does not mean that your system showed strange behavior because a new driver was released. But technically of course a vendor might actually make its driver underperform some time after installation - which does not means though that it actually does. Jan 11 at 19:12
  • 2
    "nor does any configuration of my environment change" - does this mean that you got no kind of software updates on the system, especially not the normal OS updates? Because these sometimes cause problems since the OS behavior changes in a way not anticipated by driver developers - which often means that they release driver updates to fix these problems. Jan 11 at 19:16
  • yeah, I'd suspect a system update... GL/DirectX could have updated... might have even patched a vulnerability. Sometimes another game's installer will update those things... (You should really keep everything up to date!)
    – pcalkins
    Jan 11 at 19:26

As people pointed out in the comments, corelation does not imply causation. In other words, the fact that your system was instable at a time when a new driver was available does not mean that the fact that a new driver was available is the reason that your system became unstable.

An Example

Imagine there is some function in your OS, which works in a specific way. It is not documented to work in this way, and according to the spec, one cannot rely on the function working the way it does. The nVidia driver developers disregarded this, and wrote their driver to expect this function to work the way it does - after all, that's how it worked for countless years now.

Then, Microsoft changes this function, claiming that it is not a breaking change, because they never guaranteed this behavior in the first place. The driver now expects something that is no longer true, and the system becomes instable, crashing in certain circumstances, etc.

nVidia becomes aware of the issue and releases a new version of their driver. All you observe is that your system is unstable, and that that happened around the same time that a new driver has become available. The actual reason for the instability was an OS update, or rather, the developers assuming something they should not have assumed.

The Actual Question

Do hardware manufacturers control PCs remotely? No.1

It's overwhelmingly unlikely that this is the case, as backlash against a hardware manufacturer controlling computers remotely without the user's explicit and informed consent would be immense.

1 Unless you count things like the Intel Management Engine, which are designed specifically to allow remote management.

Not the answer you're looking for? Browse other questions tagged or ask your own question.