Using something that has moving parts will cause wear on that thing.

GPUs nearly always have fans (“active cooling”), which wear out when used. It doesn't matter if you're mining or playing games, it wears on the fans.

But the rest of the card doesn't have moving parts. However, temperatures and changes in temperature can cause wear, as well, as it causes materials to expand and contract.

While mining, the temperature generally stays the same, and isn’t high, because miners tend to down-clock their cards roughly 20–30% in order to make them more power-efficient.
For example, I have an RTX 2080 Ti that’s mining and is constantly at quite exactly 62 degrees C, +/- 2 degrees C depending on the season/ambient temps, and is set to a power target of only 67% (it draws 174 W).
This is a perfectly comfortable temperature for the components on the card.

When playing games, the load you put on the GPU varies greatly, and sometimes the card has a short spike in power draw, which can eventually wear out the VRMs (voltage regulator modules). Basically, a VRM will eventually “burn out”.
However, mining doesn’t cause such spikes, because the load is very even.

In conclusion, mining is much nicer to a GPU than gaming is, but simply using something almost always implies putting some level of wear on that thing, that’s just reality.

Luckily, fans can be replaced in most cases, even if it might either be an ugly solution or a bit more expensive than you’d like.
Keeping the fans and the heat sinks dust-free by cleaning them regularly means they don’t have to work as hard and won't wear as quickly. (True whether gaming or mining.)

Per hour no more than gaming.

Gaming is entertainment. Most gamers won’t undervolt their GPU to get the most economical performance out of it, nor will they care how hot it runs unless it hits its thermal throttle.

Mining - done properly - is a business enterprise to make money. You don’t make money by having to replace your GPUs frequently. Nor do you let the hardware, optimised for gaming, eat into your profits.

An RTX 2070 has a TDP of about 280watts and should be kept below 90*. I have two of these mining away.

• one draws 104w and has core temp 55 and memory temp 72.
• other draws 67w and has core temp 54 and memory temp 70.

These are way below the GPU’s tolerances. On the other hand if I throw Cyberpunk 2077 at them with everything turned up the 2070 is going to be pulling close to it’s maximum power and hitting higher temperatures.

That’s not to say mining doesn’t carry it’s own risks. The cooling fans are running 24/7 and the cards are pumping out heat 24/7. To be fair the fan speed and heat is almost constant (constant speed, constant temperature) so there’s no damage from fluctuations. Fans are mechanical and wear out, and thermal paste dries out over time.

But like previously stated, mining is a business and a good miner will be looking after their hardware. Fans and thermal paste are cheap and easy to replace.

It depends greatly on how it was used/abused and for how long. There are two primary concerns, overclocking and fan life. If you’re looking at a 3 year old card that was used extensively for mining, the fan or fans may need to be replaced much earlier than a card that wasn’t used for mining. Replacement fans are usually $10 to $15 dollars, but some are difficult to source. The other consideration is overclocking. If the card was substantially overclocked and over volted to get the highest possible mining hash rate, that can take a toll. But miners are at least as likely to go the other way and underclock/undervolt. Electricity costs money, and miners are acutely aware of that. There is no profit in overclocking if the increased electric bill due to overclocking costs more than the small increase in hash rate from overclocking. On the other hand, miners can significantly reduce electric consumption through underclocking and undervolting without significantly compromising hash rate.

I’ve bought 6 used gpus that I’m pretty sure were used for mining. I have had to replace fans, but I haven’t gotten a dud that didn’t work yet.

Dont use your regular GPUs for mining, theres negligible chance of this reaping you any profits

If you are serious about mining, you make make

• Check the feasibility of mining in your region, using online mining profit calculators
• Setup a safe bitcoin wallet preferably hardware wallets. You can use hardware wallets like Ledger Nano S which comes with a LED screen to check your balance and to approve transactions
• Invest in a Bitcoin ASIC, they are very powerful processors made specially for mining bitcoin. A popular bitcoin based ASIC is Antminer S9, which can provide great hash rates as compared to traditional GPUs
• Install a legit mining software
• Join a mining pool, mining on your own is not feasible since it only gives you lesser chances of getting rewards. When you join a mining pool, the combined computational power greatly increasing your chances of reaping rewards. Be clear about the policies of the mining pool that you are planning to join
• Have fun mining-

Smash the follow button for more on bitcoin

In most cases, no. Mining, to be profitable, typically requires peak efficiency and reliability. To accomplish this, the cards are usually undervolted a little bit. The cards are generally run well below the speeds and temperatures they’d see in gaming. Also, miners know that selling the cards is part of the equation for profits, so they generally take very good care of them as far as cleaning and re applying thermal paste.

Think of mining the same way you’d think of “highway miles” on a used car.

Several tech youtubers have run mining cards against new GPU’s and found no difference in performance. I personally run a retired mining card in my primary gaming PC.

Not like you think.

GPU’s produced for working under high load for a long time. But they are not build for working %100 load for weeks without stop.

So, heat can damage PCB. On a long run it is possible to damage GPU it self too. Connections between GPU and PCB is the weakist point here. Which this damage can be repariable.

Real problem here is electricity. Power infastructre is great on North America, Europe, Japan and some region of China. But it isn’t perfect. Time to time voltage can be rise and drop. Computers are always open are much prone to this kind a problem. And electric problems can destroy CPU, mainboard, PSU and GPU. Also users has to clean fans constantly. Those will atrach extreme amount a dust.

Dust over mainboard and GPU it self had clean up to. Because over time this dust turn in to conducter. Which is a very bad thing that destroy mainboard, GPU and CPU.

But, if you manage to keep GPU and computer cool and clean and use nice uninterruptible power supply (UPS), you can use your GPU for mining without any bad dreams.

The biggest worry would be the fans. If they are sleeve enthusiasts, they will most likely fail in a few years. They might last ten years if they are dual ball bearing.

Many of the fancier cards have 10-year caps as well, and those that don't usually advertise it on the box. If you use 10 year caps and twin ball bearing fans, and keep the cards at a reasonable temperature, they should last a long time. My 1070 cards are set to run at 52 or 53 degrees Celsius, with a 75 percent power target and 42 percent fan usage. Over the rack, I have a box fan blowing.I imagine they'll be OK for mining as long as it's profitable, and then they'll be repurposed as a gaming card until they're eventually retired because they're too old.

In reality, as a general rule of thumb, mining cards are only usable (profitable/efficient) for two generations. Gamers can get more than two generations out of a card, but most gamers won't keep it much longer than that —- at least in mainstream games. The previous generation of desktop graphics cards was the GTX 780. (980, 1080). A GTX 780 would be adequate for gaming but not for mining. I saw in a forum that a GTX 780 was barely producing enough to cover energy costs with mining in the January/December timeframe (highest mining profitability in the previous year).

It will function similarly to any other piece of computer or server gear. I know people that have a number of old power supplies or PCs, including Dell or HP OEM systems with old P4 or Dual Core Pentium processors that still work. My father uses an AMD X2 3600 motherboard that has served him well over the years. I recently upgraded him to an AMD tri-core processor. Then there are certain PCs that don't last ten years due to a hardware failure.

For further reading

My experience matches Michael Johnson’s answer below.

Mining Bitcoin and ultimately Litecoin on my videocard’s GPUs causes the system to run continuously at its rated temperature maximum for months at a time. In my experience running two cards continuously, I experienced fan failures at ~16 months, both cards failed within a few days. That’s not bad, considering it was over a year at 95% fan speed. Far longer than I expected. ~6 months later I had a dead card I was able to revive by replacing 3 bulging electrolytic capacitors. The same card failed irreparably some months after that. After that I pulled the remaining card and it’s now in my “TV” pc, and is only used for video playback.

I replaced these with a pair of R290x’s, and they mined for “many” months without incident.

If you are going to do this, schedule regular monthly maintenance to blow the dust out of the PCs. Dust is an insulator, and heat kills PCs.

If you’re mining bitcoins on a CPU you’re going to have a bad time. You want to use a GPU (or lots of GPUs, or I wonder if a TPU is the way to go? [Turns out I’m way behind - thanks Vladislav Zorov, so ignore this]).

What’s the processor (whatever kind it is) going to be doing if it’s not mining? If it’s turned off then it will last a lot longer than if its turned on. If it’s turned on and doing something else instead, then that depends on what else it is doing, how hot it is, and how hard the fan(s) is(are) running.

Also, what counts as end of lifespan? Most processors lifespan ends because they are out of date and the components around them fail. For many people a fan failure on a CPU/GPU might mean that the processor is at end of lifespan, but if you’re a bitcoin miner you probably know how to change a fan.

Those of us who have built prized gaming rigs wouldn’t dare risk them for anything, even extra income. That's why cryptomining can daunt newcomers—rumors about hardware degradation have made them fearful of killing their darlings. Lucky for them, mining doesn't degrade your GPU any more than a lifetime of l33t gaming.

Since the 2017 crypto gold rush, miners and gamers have butted heads over their predilection for the most performant GPUs. High demand led to shortages and price gouging, making an expensive hobby even more dear.

As the crypto craze cooled—and desktop miners got muscled out by huge ASICs farms—the market flooded with cheap, second-hand GPUs from remorseful investors. Despite the newfound surplus, gamers got wary that those cards might be practically worthless due to hardware degradation.

Can Mining Degrade My PC?

Cryptomining isn't the first cottage industry to turn to GPUs for processing. Their aptitude for parallelized computation makes them equally useful for more rendering particle effects in Far Cry, solving cryptographic equations, and hunting E.T. Exchanger24 - very good and safely crypto site. Go through биткоин в рубли and exchange your Crypto easily with exchanger24. It’s really a trusted site.

If you’re worried about your hardware, breathe a sigh of relief. Mining doesn’t put any more stress on your GPU than a AAA game with great graphics—it's doing the same computing tasks.

No, it decreases the lifespan of the Bitcoin miner’s chip - an integrated circuit designed and manufactured specifically for mining Bitcoin :)

But if the question is: does running a chip at peak loads for extended periods of time shorten its lifespan, compared to just idling it most of the time?

I think the answer is yes, because at the scales and currents at which integrated circuits operate the flow of electricity is like a river strong enough to knock out atoms from the conductor. When there’s more heat, this is more likely to happen (since the atoms vibrate more and are more likely to get in the way of electrons). So even if you’re within the thermal limits, higher temperatures like the ones from non-stop load are still a bad thing (even though in practice the lifetimes of CPUs are so long that you won’t really notice, even if you shorten it by a lot). Link: Electromigration

Yes it will over time. You are pushing the GPU to its maximum for as long as you are mining. This will eventually cause damage to the GPU. If you have proper ventilation then this time to failure will be longer, but it will wear out. Also as the difficulty mining Bitcoin goes up your GPU will mine fewer BTC (satoshi)

Back in 2013 I was mining BTC on my computers. But it became hard to keep up with the BTC difficulty change and was using a lot of power. I have put all my mining efforts into cloud mining.

Here is one of my mining account dashboards

I have 8TH of power and this is predicted to earn me 13.00 USD per day. Right now I am re-investing everything I earn back into mining to increase the hashpower I have. In June 2018 I will start taking out profit.

If you want to mine BTC I would recommend using Hashflarehttps://hashflare.io/r/16DBCED8

It is a slow way to get BTC but as the price of BTC goes up you earn more and you don’t have to worry about hardware software or electricity. Somebody takes care of all that for you.

Bitcoin mining typically pushes the GPU to 100% utilization. At that point, the GPU is going to be very warm after a few hours. Depending on airflow in your bitcoin mining rig and the room it’s in, the hardware could sustain permanent damage.

There are ways to manage this. Undervolting the GPU is the most common. Bitcoin mining doesn’t necessarily destroy a GPU, but it could degrade performance down the road.

Wear and damage are very different, Quora. Damage is caused by either hardware malfunction or user error, maybe a power surge. Wear is a normal process. It would be great if someone could “de merge” these questions..

Mining will not harm your GPU, at least not by itself, even if you mine all day long. However, if your GPU is in a badly ventilated case and will become too hot under load, this may decrease its lifespan (but this will happen from any other load which causes too much heat, including rendering or frequent intensive gaming, especially all day long).

I personally mine on my new GPU (which I got in the beginning of 2020) in open PC case, so temperature is good, and it still works without any issues.

By the way, mining is relatively light compute load. Neural networks and rendering heat up GPU more - mining usually consumes 125W on my current videocard, but rendering or neural networks 175W.

I still have an old GPU - GTX 680 4GB from 2012, I used to run rendering jobs all day long on it for many years (either my own, or to help others on sheepit-renderfarm), and it still works just fine, it can pass VRAM tests or compute tests without errors, and it is still used in one of my PCs without any issues, even though it is not useful for rendering or mining anymore because it is so deprecated.

In other words, if you keep the temperature under load at reasonable level, the GPU will depreciate faster than it stops working. I recommend keeping temperature below 80 degrees.

Only very slightly, but it depends on the quality of the hardware.

Mainly, it’s the fans. They wear down as they keep running. You can never have absolutely no friction, and friction causes wear. But some graphics cards have excellent fans that are built to last a very long time.

As long as you keep the computer below something like 65 degrees C, it doesn’t damage the parts. One might even argue that constantly changing the temperature is more damaging than having it at a constant sort-of-high temperature.

When you change the temperature, the materials expand and contract, and this can cause micro-fracturing, which is what eventually kills the part, if it doesn’t out-right get burnt to death at some point.
But if you keep the temperature steady, it doesn’t expand or contract, so it’s good for the parts. That is, as long as this steady temperature isn’t too high. Below 65 degrees C should do no damage whatsoever to the actual circuitry, no matter how long you keep it there. You might be able to go even higher than that without any risk at all, as well, but I can’t really say for sure where the ceiling is.

Personally, I have been mining with my GPUs since 2016 or 2017, constantly. None of them have failed in any way, or shown any kind of damage or problems. (Two GTX 1080s, and one RTX 2080 Ti.)
I keep them at 65 degrees C.

Also, note that it only wears on the GPU. When mining with a GPU, the rest of the computer is essentially asleep. You will not hurt any other part of your computer, only potentially cooling fans for the computer case and the cooling fan on your GPU.

Clean out the dust in your computer perhaps every 3 months if the conditions are bad, and less often if the conditions are good. Don’t let dust build up, keep an eye on that every now and then.
And this is true whether you’re mining or not.

No, they don’t slow down or otherwise lose performance over time. That simply cannot happen at all.

One thing to keep in mind: it’s easy to become confused about reliability vs. performance. A GPU, just like many other pieces of computer hardware (such as hard drives, memory chips, etc.), may lose reliability after getting mishandled, splashed with water, collecting a lot of dust, etc. However, reliability is not performance. Whenever they are fully functional, they perform at the same level exactly as they did on Day 1.

Personal experience: I handled a lot of computers and buy one new computer for personal use about every 18 months. I have never witnessed a CPU or GPU losing performance over time (unless they are re-configured to do so). Also, both GPU and CPU tend to have an extraordinarily long lifespan. If they don’t have problems during their first few weeks of intense use, they will most likely last many years, possibly more than a decade or two, regardless how much time you spend running them. My 2001(or 2000?) Toshiba Satellite laptop was 100% fully functional in 2021 (about 2 decades later) when it was finally given to a government-sponsored recycling agent.

To keep this type of stationary computer hardware clean, try to get a computer case with some type dust filtration, and also also get a liquid cooling system (so you don’t need to get a lot of air-borne dust into your computer case).

My last desktop computer, custom built between 2016 and 2017, is still having a very clean interior. Its GPU, a GTX 1070 card, remains practically dust free, even after 10,000s of hours of intense use. No cleaning is ever needed. 100% functional, 100% as fast as it was on Day 1, though I don’t use it any more as it’s too old for editing and/or gaming at 4K.

Not as hard as heavy overclocked gaming. Probably not as hard as video rendering.

The GPU is able to run 24/7 without much effect on its lifespan.

A good mining rig is designed to be power efficient. The goal is to have the highest ratio of compute power to electric power used. The higher the local electric cost, the more you want to optimize power use.

Optimal means tuning power down, rather than up as you would for Max gaming performance.

When I did mining on a GTX 1080, the temperature rarely hit 60, and the fan speed wasn't maxed. Gaming pushed heat to 70, fans high to get the best speed.

As long as the computer is in a clean and cool environment, it shouldn't be too hard on its components. Mining GPUs get sold not because they are worn out, but because newer cards are more efficient. They can last at least five years, and most will still run long after they are too old for current modern games.

To expand on Vladislav’s answer:

The thermal expansion / contraction, can be mitigated, if you don’t keep switching workloads. E.g., instead of running it flat out for a few minutes, then letting it cool down for another few, and rinse repeat constantly. Instead, run it flat out for several hours before turning it off.

Not as good as simply never putting any electricity into it ever. But, ever so slightly better than keep changing its temperatures between extremes.

You may wish to adjust settings, so it never actually runs to its peak performance. In order to reduce that maximum temperature, it would reach. That’s also a way of mitigating this contraction / expansion effect causing stuff like metal fatigue.

But, this begs the question: Why did you buy a “high-end” GPU, only to nerf it, so it runs slower? Wouldn’t it make more sense to buy a cheaper one which already runs at the performance you’re going to get anyway?

That is, if you can’t (or don’t want to) install better cooling to drop the temps lower without nerfing the performance.

Other than that, which is likely the biggest cause of any hardware failure. Also look at the quality of your electricity coming into the computer. Perhaps invest in a UPS (or at least a voltage regulator / surge protector) if your electricity supply is even slightly fluctuating. If you’re in a place with terrible power, constant brownouts and blackouts, or in a zone where lightning is prevalent (in my case both, awful power company and one of the most extreme lightning zones on the planet), definitely get a UPS. It’s cheaper to replace the UPS than your entire computer every time.

Of course, this is attempting to answer the question as if by “lifespan” you mean, until it breaks.

But, really, even high-end GPUs can last years (if not decades) in this sense. You really have to do “something stupid” to make it “die” sooner.

There is another “common” way some use this “lifespan” idea: As if, by when, they would need to upgrade it.

I have a few … let’s call them “problems”, with this concept.

• What constitutes “need to upgrade”? As opposed to “want” instead? Do you “need” to have the best available? Or can you make do, as long as whatever program you run, performs acceptably?
• What use are you placing on this GPU? Games? Statistical calcs? Hashing? 3d modelling? Video editing? Something else? … Actually, scratch that, instead answer the next point:
• Do you have some foreknowledge of what software you’d be running in a year’s time? What about 5 years from now? 10? If so, do you know the requirements of that future program? If so, why ask this question?

Mining harms your GPU in the sense that one of its by-products is producing excess heat. If you run your mining setup 24/7 at a high temperature – above 80C° or 90C° - the GPU could sustain damage that will severely affect its lifespan.

However, mining is not alone in placing stress onto a GPU. Playing demanding games and processing data using a GPU can also wear out a GPU when at high temperatures. Similarly, overclocking your GPU too often – whether for better hash rates or better gaming performance – can also cause irredeemable damage to the GPU.

In fact, some cryptocurrency miners even suggest mining is less harmful to your graphics card than playing the latest AAA game is. As miners seek to reach the highest possible hash rate, the GPU may reach high temperatures but will remain at a consistent one with a consistent workload. In contrast, as playing games see the GPU fluctuate between rapid cooling and rapid heating depending on the load, this could cause more damage in the long run.

However, the major cause of damage to a graphics card – whether used for gaming or mining – is not taking proper care of the hardware components.

Dirt and such things accumulate on the heatsinks, and the fans can wear a bit over time, so the main difference is that the cooling performance drops over time. This can translate to lower speeds, but this can be fixed by fixing the cooling.

That is, even if you blow away the dust using compressed air, tiny pieces of dirt will accumulate, and it’ll probably sit too tight to be blown away and will require something like an alcohol wash to clean it off.

The VRMs (voltage regulator modules) have to take a lot of heat, which means they keep being subjected to thermal expansion, and eventually they won’t be able to take it, anymore.
It’s possible that their performance degrades, I’m not sure. However, this is not directly tied to the speed of the card, since it’ll still try to run at the same clock speed.
But if the VRMs aren’t good enough, it means they can’t provide the power the card needs, especially when the car “spikes” in power use, which can give you hardware errors, even bluescreens or system lockups.

As the GPU itself gets older, it’ll inevitably start developing small defects, and this will cause it to start making errors, and it’ll be more sensitive to heat, so you get less “overhead” before failures happen.
This wear comes mainly from thermal expansion, so if the card is kept at a regular temperature, it’ll live longer.

Keep the card cool and it’ll likely live a long time, even if you might need to replace a fan due to mechanical wear.
While keeping the temperatures down is the most important thing, also note that constantly changing temperature is part of the problem. Changing temperatures of the material means it expands or contracts, and it will eventually cause microscopic fractures in the material.

Reasonably, GPUs aren’t supposed to last very long. If your GPU is 7 years old, you should have already upgraded it by now. I personally upgrade my GPU every 3–4 years, or so.
However, they do appear to survive quite well, even if some units fail earlier than others.

Anyone could make a case for both.

And no, Miners don’t baby their cards like they would lead you to believe. When Ethereum was minable, ETH miners would downclock and undervolt the GPU core of their card, but they would overclock the VRAM to the limit. A site called Techpowerup has a few longstanding threads that deal with some of the hapless mods miners have made to Graphics cards.

One thing you can say about mining is the cards are running consistently, so thermal cycling is not an issue. One of the reasons the old Generation 7 Gaming consoles died so often is because of Thermal Cycling on top of the Lead Free Solder being of poor quality. Expansion and contraction is the culprit here.

Short answer yes.
GPU’s used for mining run at full tilt for 24 hours a day, the bearings on fans start taking stress and eventually wear out.
If the card runs too hot then you always have the possibility of burnt components, melted wires, even some soldering can melt and cause shorts which damages the card.
However a GPU used for mining can be treated with care ad can last a long time. fans can be replaced, dust can be cleared away and the cards can be kept cool with additional cooling (or extraction)
I have some rigs running the same cards for 5years, sure they don’t produce like they used to but 5years is a long time for a gpu to be running 24/7.

People follow profits.

You can buy specialized mining equipment for cryptocurrencies like bitcoin, and for a time you’ll be able to profit over the cost of the electricity to run the machine. However it won’t be long before so many of those machines have been produced that it ceases to be profitable unless you have a very cheap source of electricity. So the game there is to hope that your machine arrives in time to still be profitable to run, and remains profitable for long enough to pay for itself. It’s risky, and once a machine ceases to be profitable it also loses its resale value.

We’ve had a bit of an arms race for a decade now where significantly more efficient equipment is developed and released, making the previous generation of equipment quickly obsolete.

Basically the problems are:

• It takes a long time for the cryptocurrency miner to arrive
• The resale value of used cryptocurrency miners is weak

With GPUs you don’t usually wait months for them to arrive. If a seller has stock then it gets to you in a matter of days. And then if the market sours to the point where mining is no longer profitable, you can expect to get about 1/3 to 1/2 of the MSRP back from selling your GPUs, making it far less risky.

However timing is still everything. A lot of people are looking at ROI as a linear function, that if it costs me $1000 for equipment that makes $10/day that I’ll be able to pay off the cost in 100 days. In reality it’s unlikely that the price of cryptocurrencies will keep up with the rate at which new GPUs or miners are produced, and it may take twice as long to get there. Or if the market crashes then it could be years.

You plug them into a normal computer in the usual way. Just as if you’re using those GPUs for graphics, video, machine learning, whatever.

If you’re asking, how to install multiple GPUs at the same time into the computer. You’re looking for PCIe splitters. Something which plugs into a PCIe x16 slot, and splits that to (say) 8 off PCIe x2 connections.

This tends to be a bad idea for other workloads, as you’re reducing the amount of throughput between the GPU and the CPU. But, since mining tends to only need small amounts of data every once in a while. It’s not going to reduce performance.

Then ensure your power supply can handle the extra load, and install the power cables to that as well.

Mount the cards wherever you like. Usually it’s better not to enclose them. Easier to clean and better ventilation.