It's not often that we hear credible reports that a driver may have caused actual, physical harm to a video card, but that's precisely what an unhappy group of users has done. According to reports, the latest series of Forceware (196.75) caused certain cards to overheat or fail altogether due to a fan control problem. NVIDIA
has pulled the driver and issued the following statement:
We are aware that some customers have reported fan speed issues with the latest 196.75 WHQL drivers on NVIDIA.com. Until we can verify and root cause this issue, we recommend that customers do not download this driver. Instead, please stay with, or return to 196.21 WHQL drivers. Release 196.75 drivers have been temporarily removed from our website and we also are asking our partners and others to remove temporarily this 196.75 WHQL driver as well.
Obviously we recommend you fall back to whatever driver release you were running before, but we also wanted to see if we could demonstrate the difference between the 195.62 drivers (the set I was personally using) and the 196.75 series. To this end, we uninstalled the 195.62 drivers, ran DriverSweeper (from Safe Mode), and then installed the newer problematic driver series. We measured temperatures using Furmark 1.8 for OpenGL and World of Warcraft
for Direct3D. GPU temperatures were measured using SpeedFan.
Furmark is a GPU heating utility that both tracks and graphs temperature; our GTX 295 hit a maximum temp of 105-106C while running the utility. Here's what we saw:
As you can see, swapping from 195.62 to 196.75 made no difference to either the steepness of the temperature rise or the maximum temperature value. Our World of Warcraft
tests were similar—in both cases, the GPUs hit a maximum temperature of ~100C with a 2-4 C variation between them. The glitched Forceware's were only online for about 48 hours, but we'd still recommend double-checking if you recently updated. There's been no reports that any series besides 196.75 were affected, so anything else should be safe.