4 ms·

How did this code damage computers? Overheating? Kept fans running higher?

banachtarski 13y ago
just thermodynamics really. the cpu cycles spent mining bitcoins are in fact costing the user something.
Hilyin 13y ago
I don't think it'd damage the computers, if it did, it'd be very minor, like you said. I'd say the biggest effect would be people's electrical bills being higher.
- speeder 13y ago
  Bitcoin mining on Laptop GPU is known to destroy whole machines (usually by overheating several parts of the motherboard and ruining them... also ruining the GPU itself too)
- HarryHirsch 13y ago
  Quite a bit higher, in fact. Under full load a fanboy graphics card dissipates 100 - 150 W. A wattmeter is instructive toy.
- shawabawa3 13y ago
  I did some GPU mining for about 5 minutes once. Had to stop it immediately because the noise my GPU was making definitely wasn't good for it. The fan still rattles a bit sometimes from the 5 minutes it was being run at 100% GPUs are not really designed for 100% load for prolonged periods of time
testbro 13y ago
I believe the concern was that it ran on GPUs (mining on CPUs hasn't been cost-effective for quite some time). Since GPUs are rarely run constantly and don't see loads like mining produces during typical use, their owners might have overclocked them under those assumptions. With those assumptions violated, I wouldn't be surprised to see cards fail prematurely.
- corysama 13y ago
  Don't even need to overclock them. Given a sample size of 14000 GPUs, I would expect a couple hundred to at least bluescreen simply from being placed under a sustained serious load for the first time. Try running the FurMark stress test on any given machine. If you don't get glitches or a reboot, that GPU is above-average stability in my experience.
  - HarryHirsch 13y ago
    This is a sad comment on the state of consumer hardware. Speaking from my viewpoint as a computational chemist, a video card that fails memtestG80 is not merchantable and gets slung back at the manufacturer. After a while you have learned how fast companies deal with RMAs and you know that factory-overcooked hardware is next-to-useless.
    - ferongr 13y ago
      >This is a sad comment on the state of consumer hardware And also incorrect. A system crashing on a combined Linpack/Furmark load is defective or contains a defective component. Even extremely thermally-limited laptops do not suffer from this due to protection mechanisms.
ambiate 13y ago
While mining litecoins, a capacitor in my power supply blew up. I saw a huge jolt of blue, my monitor lost connection and finished up with some smoke.
invisible 13y ago
GPUs overheating and getting damaged from said heat. Running a GPU at 100% for days is probably a good way to burn it up.

just thermodynamics really. the cpu cycles spent mining bitcoins are in fact costing the user something.

I don't think it'd damage the computers, if it did, it'd be very minor, like you said. I'd say the biggest effect would be people's electrical bills being higher.

speeder 13y ago
Bitcoin mining on Laptop GPU is known to destroy whole machines (usually by overheating several parts of the motherboard and ruining them... also ruining the GPU itself too)
HarryHirsch 13y ago
Quite a bit higher, in fact. Under full load a fanboy graphics card dissipates 100 - 150 W. A wattmeter is instructive toy.
shawabawa3 13y ago
I did some GPU mining for about 5 minutes once. Had to stop it immediately because the noise my GPU was making definitely wasn't good for it. The fan still rattles a bit sometimes from the 5 minutes it was being run at 100% GPUs are not really designed for 100% load for prolonged periods of time

testbro 13y ago

I believe the concern was that it ran on GPUs (mining on CPUs hasn't been cost-effective for quite some time). Since GPUs are rarely run constantly and don't see loads like mining produces during typical use, their owners might have overclocked them under those assumptions. With those assumptions violated, I wouldn't be surprised to see cards fail prematurely.

corysama 13y ago
Don't even need to overclock them. Given a sample size of 14000 GPUs, I would expect a couple hundred to at least bluescreen simply from being placed under a sustained serious load for the first time. Try running the FurMark stress test on any given machine. If you don't get glitches or a reboot, that GPU is above-average stability in my experience.
- HarryHirsch 13y ago
  This is a sad comment on the state of consumer hardware. Speaking from my viewpoint as a computational chemist, a video card that fails memtestG80 is not merchantable and gets slung back at the manufacturer. After a while you have learned how fast companies deal with RMAs and you know that factory-overcooked hardware is next-to-useless.
  - ferongr 13y ago
    >This is a sad comment on the state of consumer hardware And also incorrect. A system crashing on a combined Linpack/Furmark load is defective or contains a defective component. Even extremely thermally-limited laptops do not suffer from this due to protection mechanisms.

ambiate 13y ago

While mining litecoins, a capacitor in my power supply blew up. I saw a huge jolt of blue, my monitor lost connection and finished up with some smoke.

invisible 13y ago

GPUs overheating and getting damaged from said heat. Running a GPU at 100% for days is probably a good way to burn it up.