Is heat the only thing that will damage my GPU when crypto mining? How can I optimize my setup?

[mqdfco]

I'm using my gaming pc to crypto mine, so I care quite a lot about whether or not my components will get damaged. My "GPU Load" is constantly at 100% when crypto mining. Will this damage my GPU in the long run? I think my GPU temperature is low, but could running the GPU at 100% still damage it even if the heat is low? I have not done any over/under-clocking which is what I want to look into.





These are my GPU stats after a few hours of mining. Are these good? What should I do to better optimize it for longevity, which is frankly the only thing I care about.

 

[toilet pepper]

You are correct. Heat kills electronics. Mining or any other load doesn't kill the card or shorten its lifespan. They are meant to do "stuff" and not idle.

You can also increase your fan speed for cooler temps.

Here's a list of things that can go "wrong." I'm sure there's a lot so feel free to add:

1. Fans die and take the gpu with it. Normally the card will throttle before anything bad happens.
2. High heat to low heat cycles. Might mess up the micro soldering of electronics and cause a short. (I highly doubt these happen. Most electronics does not use low heat Lead for soldering anymore.)
3. Dirty power source. Might mess up your PSU and other components.

 

[Zach_01]

I'm using my gaming pc to crypto mine, so I care quite a lot about whether or not my components will get damaged. My "GPU Load" is constantly at 100% when crypto mining. Will this damage my GPU in the long run? I think my GPU temperature is low, but could running the GPU at 100% still damage it even if the heat is low? I have not done any over/under-clocking which is what I want to look into.





These are my GPU stats after a few hours of mining. Are these good? What should I do to better optimize it for longevity, which is frankly the only thing I care about.

Your temperatures are fine. More than fine I would say.
GPU: 56C, when in gaming this could reach 70~80C
GPU Hotspot: 69C when in gaming this could reach close to 100C or even more
VRAM: 67C when this could reach 80C or more.

What's your hash rate (MH/s) with this configuration?
First of all I would lower GPU frequency to around 1200MHz and GPU voltage accordingly. 1845MHz is totally unnecessary for mining. That can save a few degrees on temp and a few watt on consumption without compromising MH/s.

And if you want the max possible mining performance you should OC VRAM to the point of start having rejected shares on the miner. Then back it down a little.
Remember that you can change frequencies when miner is running but its better to restart it after any MHz/voltage change and wait several minutes (5~10) for hashrate to stabilize. For rejected shares you have to wait a few hours to be shown if anything is wrong (i.e. excessive VRAM freq).

 

[64K]

The old-timers always say that heat and dust are the enemies of electronics. Having a quality PSU is important also.

If I were going to be running a GPU 24 hrs a day then I would blow it out for dust every 3 months. It just takes a minute.

 

[Shrek]

You are correct. Heat kills electronics.

Not arguing with this, but heat variation might also be a factor.

Reminds me of some diesel-electric trains that adopted a marine engine that had a good reputation, but they found the block was cracking early. In the marine application the engine ran at constant load while in the train application the load ramped up and down a lot and this was giving rise to thermal fatigue.

So some cards cut the fan at low loads to keep the GPU temperature up and so the variations down.

What should I do to better optimize it for longevity, which is frankly the only thing I care about.

I am surprised you are not also concerned about energy consumption as this costs money.

 

[Zach_01]

Not arguing with this, but heat variation might also be a factor.

I agree, this is a valid point. Temp variation or fluctuation if you prefer can be a negative factor to longevity.

 

[freeagent]

Its not so much heat, its the cycles. But if its running cool you should be good.

 

[ThrashZone]

Hi,
GPU's are all pretty much factory setup for quiet gaming so 60c is not unusual gpu temp.
You can always of course setup a custom fan curve I do it's pretty easy I hate hot spots.