NVIDIA Calls A Feature A 'Bug,' Strips Away Overclocking Option On Its Mobile Device Cards

(Mis)Uses of Technology

from the you-can-buy-it-but-you-can't-have-it dept

In theory, the marketplace for goods works like this: a purchaser hands over $$$ and in return receives a product that they own and can use as they see fit. In reality, purchasers often hand over $$$ and find that the product they purchased is still in the grips of the company that took their money but seems loathe to honor its end of the deal.

Case in point #38,909: guess what NVIDIA thinks is a “bug,” not a “feature.”

Starting with the Fermi drivers, though, a software overclock was possible in the drivers, which allowed you to adjust your laptop GPU’s clockspeeds at will. Tools like AfterBurner from Micro-Star International Comp., Ltd. and Turbomaster by ASUSTek Computer Inc. allowed users to more easily and safely tweak their GPU’s clockspeeds on select gaming laptops with cooling solutions designed to cope with the higher thermal load. Companies like the Clevo Comp., Sager, ASUS, MSI, and Dell’s Alienware regularly sold models billing overclockability as a sales feature.

What OEMs apparently didn’t expect was that NVIDIA would rob customers of that feature. But that appears to be precisely what happened.

NVIDIA pushed out new drivers last December that took away customers’ ability to overclock their cards. These were targeted at cards for mobile and hybrid devices, where the chance of overheating (and causing serious damage) was more pronounced. Those who had overclocked their cards but now were unable to do so demanded answers from the manufacturer. And wouldn’t you know it, the explanation for NVIDIA’s removal of this option cites “safety” as the primary motivator.

Unfortunately GeForce notebooks were not designed to support overclocking. Overclocking is by no means a trivial feature, and depends on thoughtful design of thermal, electrical, and other considerations. By overclocking a notebook, a user risks serious damage to the system that could result in non-functional systems, reduced notebook life, or many other effects.

There was a bug introduced into our drivers which enabled some systems to overclock. This was fixed in a recent update. Our intent was not to remove features from GeForce notebooks, but rather to safeguard systems from operating outside design limits.

“Safeguard systems from operating outside design limits” sounds an awful lot like “your purchased items are only as flexible as we allow them to be.” Sure, warranty departments handling burnt up/out devices may have been making some noise about dealing with the aftereffects of careless overclocking, but if so, they’re no less blameless than NVIDIA. Overclocking is generally one of those warranty-voiding activities, and if companies didn’t want to be replacing torched devices, they should have handled it better at their end. (And, as Daily Tech points out, they should probably stop advertising overclocking as a “feature” if it’s truly that much trouble in the warranty department.)

But NVIDIA’s action takes the purchased product out of paying customers’ hands. Most people who dabble in overclocking are technically adept and know the limits of their hardware (and the terms of their warranties). There will always be those who push too far or get in over their heads, and a few overclockers who disingenuously expect the device’s manufacturer to bail them out when things go wrong, but these customers are in the minority.

When a company takes away a feature (especially one that has been advertised by the devices’ manufacturers) and calls it a “bug,” it’s basically telling customers that they won’t ever own what they purchased. In this case, NVIDIA is hurting some of its most loyal customers — people who know their devices inside and out and will pay good money to stay ahead of the tech curve.

And NVIDIA’s being a bit disingenuous itself. It calls overclocking a “bug” when explaining why it took this feature away. But if it truly was a bug, why didn’t it issue a patch rather than eliminating the option? The obvious answer is that overclocking is no bug and NVIDIA knows it. But it has apparently chosen to placate its OEMs at the expense of some of its most reliable customers.

NVIDIA hasn’t issued any further statements on its “bug fix,” so it’s safe to assume it doesn’t really care whether it’s angered a number of its customers. Its position in the graphics accelerator market is virtually unassailable, especially in the area (mobile/hybrid) where it has just guaranteed its customers will get less product than they paid for.

Filed Under: bug, feature, overclocking, ownership, video chips
Companies: nvidia