When I use mplayer (i.e. play a movie with it) the system will sporadically freeze completely, for around 13 seconds. Everything halts as if the system is having a kernel panic or worse -- even the clock halts.
Around 13 seconds pass, everything returns to normal (the clock then jumps 13 seconds ahead, as well), and from the kernel pool log (dmesg) I observe this:

This seems to happen at sporadic time intervals, but can happen as often as every two minutes at "worst", making watching a movie clip the most annoying experience you can imagine.

There are no errors from mplayer.

What the heck is this? I have tried to update the graphics drivers to the latest version, to no avail. I also have no idea how this issue started in the first place.

I use Slackware 13.1.0 on a fast x86 (4.1 GHz Core i7, 6 GiB RAM, SSD, GTX480, etc.). There are no other known issues with the system. I don't use any sort of package managers or automatic updates. I've been trying to figure out what I did that caused this issue to start in the first place, but I have no idea.

This issue happens regardless of whether the media played is physically local on the computer, or if it is on a network.

That's your nvidia driver complaining. Usually nvidia tells users with Xid errors to make sure they are using the latest driver and have updated their computer's BIOS to the latest available.

Adam

I have the latest Nvidia drivers and motherboard BIOS installed.

I have updated either driver prior to the problem starting. If it was a driver issue I should be able to immediately observe the problem on the same day I did the update -- yet I did not.

From what I see on the net, you may be able to fix this by changing your nvidia driver version (previous versions), or kernel version (newer).

Not necessarily. Any number of factors could be playing a role, both hardware and software.

In any case, you can always ask on the linux nvnews forum. The first suggestions you'll likely hear are the ones I gave, the second suggestions are the ones from H_TeXMeX_H.

According to AaronP, one of the nvidia developers: "There are many things that can cause Xid errors, so you really do need to generate and post the bug report because they have to be dealt with on a case-by-case basis."

Check the temperature with Nvidia's X settings application.
For comparision, my GTS450 idles at 31C, and ~40-43C after a few rounds in Oil Rush (pretty GPU intesive)
My Girlfriend's GT240 idles at 32C, and 40-45C under load.
The Roommate's GT220 idles at 35C, and hits 47C with Oil Rush, 40-42C with less demanding GPU activities.
Our MythTV frontends (all Nvidia and VDPAU powered) never get above 45C, even with VDPAU hardware IVTC+Denoise.

Make sure you have a high quality power supply. The GTX480 sucks some serious wattage. You'll also want a power supply that has a high 12V rail. Some power supplies may taught 700W, have 4 12V rails, but each rail will only be 20-25 amps. This simply is not enough to power a card of that caliber. Antec (for one) makes a line of power supplies to cater to these power hungry cards. Instead of having multiple low power rails, they have a single high power rail.

Antec TruPower 650 - 2x 12V @ 22a, 2x 12V @ 25A
Antec High Current Power 620 1x 12V @ 48A

Most GTX480's I've seen require/recommend 42A on the 12V rail.

---edit---
4.1 GHz Core i7
What happens when you're not overclocking?

I do have a good PSU, there are no power related issues with the system. The overclocking is rock stable, I've had this setup for almost two years now (yes, at 4.1 GHz) with no known issues; nevertheless I actually did reduce the frequency by 1 GHz just to see if I could observe a difference.

However, the issue appears to be solved now.

I downloaded and compiled the latest kernel with the same optimized kernel configuration for my system, I have not had any issues since then.