If you plan on doing intense gaming, GPU computing, graphics rendering, [email protected], or crypto mining on your graphics card, you may be worried about heavy use wearing out your GPU. But will it? We’ll investigate.
Yes, but it’s complicated
Most video card lifespan information you can find online is anecdotal, though numbers can vary wildly depending on who you ask. With hundreds of different graphics card models released over the past decade, it’s difficult to summarize data on such disparate cards into easy generalizations.
So far we know the following: According to a 2020 report from a German retailer, the latest graphics cards have an overall failure rate of around 2-5% (measured in returns to retailer). And in 2021 Nvidia still provided driver updates for cards that were around 9-10 years old (like the GTX 600 series), so you can potentially expect a decade of use from a well-treated GPU card – although these could be outliers as we’ll see in the future.
Regardless of the numbers, there is hard physics at work. The materials and components used in assembling GPU cards aren’t magic: the more you use them, the faster the parts degrade and the more likely they are to fail completely. Heavy use therefore affects the service life.
Whether you see an error in your GPU card depends on a wide variety of variables, including the exact load on the GPU, the nature and degree of temperature fluctuations in the circuitry, how often the card has been powered on and off, and how clean the operating environment is .
Because a GPU card is a complex device with many parts, each can fail or degrade in different ways. We’ll go through several major parts of a GPU card and examine how they can wear down over time from heavy use.
First up: cooling fan
Of all the parts of a graphics card that are likely to fail first, we must point out the fans (or fans) which are physically moving parts. Fans keep your GPU cool by moving hot air away from the GPU chip (with a heatsink) so it can keep running.
Over time, cooling fans often become clogged with dust, reducing their ability to move air efficiently. Or the fans fail completely when an internal lubricant fails. Both scenarios increase the temperature of the GPU.
Each GPU protects itself from overheating by using thermal throttling, which slows down the GPU’s operation to lower the operating temperature. This severely limits performance. So if you have a GPU that is suddenly noisier than usual (the fan spins faster) or performs worse, give your GPU fans and heatsink a thorough cleaning with compressed air.
If a GPU fan has failed completely, you can usually replace it if you can find an equivalent fan at a computer parts supplier.
TIED TOGETHER: How to thoroughly clean your dirty desktop computer
Another suspect: Faulty thermal paste
Between each heatsink and GPU chip is a layer of thermally conductive material, e.g. B. a pad of putty or paste that helps heat transfer from the GPU chip to the heatsink.
Over time, thermal paste can crack or lose its effectiveness. When this happens, the heatsink doesn’t cool as effectively and the GPU temperature increases. As we saw in the fan section above, high GPU temperatures result in thermal throttling that slows down your GPU.
Arctic MX-4 GPU thermal compound
A highly regarded thermal compound for GPUs.
The best solution in this scenario is Replacing the thermal paste yourself. You can buy thermal grease from computer parts vendors.
Errors in other components, soldering
In addition to the GPU chip, a graphics card contains dozens of other electronic components such as capacitors, resistors, memory chips, and more. Any of these could potentially fail from heavy use or exposure to too much heat. Some fail more than others.
Capacitors are in particular error prone over time. They are sensitive to frequent temperature changes and some are defective when first produced. If you’re handy enough to fix capacitor problems, you can do that Replace possibly defective capacitors on a GPU card if you can find equivalent replacement parts.
So can the solder that connects chips and components to your GPU card’s circuit board age and break down over time B. by frequent temperature changes, rough physical handling, improper storage or running too hot. So yes, heavy GPU usage could increase the risk of solder joint failure. Repairing bad solder joints can be technically difficult, but it is not impossible.
Error in the GPU chip itself
So the question remains: can a GPU chip eventually wear out under heavy use? The answer is yes, theoretically, under extreme circumstances. But you’ll likely see the failure of another component on the graphics card well before that time.
Your graphics card’s GPU chip contains millions or billions of transistors etched into a piece of silicon. transistors age over time, which affects their performance. If enough transistors misbehave, the chip will fail.
Corresponding semiconductor technologythere is several main reasons why transistors fail over time from aging (one of them is heat), and the errors are more likely the smaller the feature size on the chip. Experts suspect that computer chips made today won’t last as long as chips made in the 1990s, but Predicting an exact lifetime is still a guess because the technology is so new.
NVIDIA is not currently publishing MTBF (Mean Time Between Failure) Estimates for their consumer graphics cards, but the company publish them for some of its industrial and business graphics accelerators. For example the data sheet for the Tesla K20X GPU Accelerator reports the MTBF for the card (at a temperature of 35°C/95°F) as 14.7 years for an “uncontrolled environment” and 23.8 years for a “controlled environment”. (Note that industrial graphics hardware is generally expected to be more robust and hold up better under heavy use than consumer graphics hardware.)
Interestingly, we can compare this theoretical figure with hard real-world data. One of the few empirical studies on GPU endurance comes from a 2020 paper titled “GPU Lifetime on Titan Supercomputer: Survival Analysis and Reliability” written by Oak Ridge National Labs. The paper reports on the reliability of the 18,688 Nvidia K20X Kepler GPU cards used in the now-retired device Supercomputer Cray XK7 Titan over a period of almost 7 years (2012-2019).
After some initial hiccups due to connectivity issues, they found relatively high reliability with the XK7’s graphics cards until 2016 (about 3-4 years later) when many started to fail. But guess what? They attributed most failures in the first batch of cards (before replacement) to a faulty resistor on the graphics card circuit board, not the GPU chip itself. Overall, the study authors found that the average MTBF of the heavily used GPU cards of the K20X is around 3 years (not 14-23 years as stated in Nvidia’s data sheet), with some of the hottest core cards failing first. They concluded: “GPU reliability depends on heat dissipation.”
So the odds are high that if you use your graphics card as intensively as one of the (then) largest supercomputers in the world, it will wear out faster and other components like fans and resistors will fail long before the GPU chip itself. Exactly how long you will receive depends on factors that we cannot predict.
Ultimately, heat is the enemy
Ultimately, according to all the sources we’ve read, the number one deciding factor in a GPU card’s lifespan is how hot it runs. The hotter the card, the faster all of its components will degrade. The hotter the card, the more it will throttle performance to prevent catastrophic failure. Good cooling extends both the life of your card and its performance.
Whether you’re mining crypto or gaming, if you’re keeping your GPU card reasonably cool with clean, working fans and effective thermal paste, you likely have a powerful card that, if you’re lucky, could hold up until then becomes obsolete and you upgrade .
If you are planning to buy a used GPU, be sure to consider its history, including how it was handled and used by its owner. More heavily used cards (which are working now) will likely work well in the short term, but are more prone to failure in the long term. We can’t give an exact number of card lifespans, but heavy use definitely wears out graphics cards faster.
TIED TOGETHER: Is it Safe to Buy Used GPUs from Cryptocurrency Miners?
This article was previously published on Source link