Topic: Mining Hardware Failure (Read 1258 times)
Heat and over-volting the core would be the prime longevity concerns.
I you want your hardware to last long time then don't buy sapphire. I already got one faulty graphics card even out of the box.
Through my own personal experience I would agree with this. Had to return 4 5970's in a row over a year period (3 fan problems and a DOA), and will never buy Sapphire again
I hear quite good things about Sapphire! Must have high variance with supply then :S
I've looked into liquid cooling but BLEH! It's so expensive. Deliciously awesome and sexy but I just can't justify that price tag at the moment.
I've looked into liquid cooling but BLEH! It's so expensive. Deliciously awesome and sexy but I just can't justify that price tag at the moment.
Liquid cooling is NEVER worth it unless you want silence or are going for a very high overclock. IMO, for the price of watercooling a dual gpu rig you can just get better video cards or a second rig with another gpu and that will more than make up for the mh/s increase you'll get over watercooling.
Plus.. it's one more component that can fail and if your h2o setup goes south, you bork your whole computer.
I hear quite good things about Sapphire! Must have high variance with supply then :S
I've looked into liquid cooling but BLEH! It's so expensive. Deliciously awesome and sexy but I just can't justify that price tag at the moment.
I've looked into liquid cooling but BLEH! It's so expensive. Deliciously awesome and sexy but I just can't justify that price tag at the moment.
I you want your hardware to last long time then don't buy sapphire. I already got one faulty graphics card even out of the box. People are complaining that fans in sapphire cards die after even a month. 2 months seems to be maximum expected life for sapphire cards. 
That is so odd you mentioned that! Literally a month ago I got barely used sapphire 5870 (I knew the originaly purchaser personally) and that damn thing died within 3 days of me using it for mining at stock speeds.. It took them 3 weeks to RMA it! I wish I could get the idiots to reimburse the btc I lost over that period of time.
I you want your hardware to last long time then don't buy sapphire. I already got one faulty graphics card even out of the box. People are complaining that fans in sapphire cards die after even a month. 2 months seems to be maximum expected life for sapphire cards.
the no name PSU
There in lies your problem. Spending good money on on a quality PSU keeps everything running longer in your system. The hardware today is much better than it was 10 years ago, but still. I've been running a q6600 overclocked from 2.4 to 3.4 in my main machine nearly 24/7 for the last 3 and half years. My 5770 has been running at 950mhz core for well over a month. I know this isn't the 'dawn of time."
F@H on my old gtx 260's was non stop for... must be over a year. Not sure how long really.
I guess if you are go through this much hardware. You need to buy stuff with great RMA service. XFX, Corsair, OCZ, Western Digital. Etc.
Also, you could sell the card after 6 months to some poor sap. Then buy newer faster stuff. Would keep it from aging in your hands, value wouldn't depreciate as much, etc.
If you are using name-brand parts your hardware will last a long time. Don't skimp and opt for the cheapest stuff or you will regret it. I speak from personal experience.
I just replaced my 2-year old Radeon 4870 X2 because I wanted a more powerful card. I've had that card for around two years and it got daily use. It ran great and gave me no problems.
As long as you keep your rig cool it will be fine. The problem with mining rigs and multiple graphics cards is the number of fans used to cool them all. If you are insanely serious about setting up a rig that will last a long time, look into liquid cooling. It's not for everyone though and will drive up costs.
I just replaced my 2-year old Radeon 4870 X2 because I wanted a more powerful card. I've had that card for around two years and it got daily use. It ran great and gave me no problems.
As long as you keep your rig cool it will be fine. The problem with mining rigs and multiple graphics cards is the number of fans used to cool them all. If you are insanely serious about setting up a rig that will last a long time, look into liquid cooling. It's not for everyone though and will drive up costs.
Most people think <70 is good(doesnt affect gpu life at all) and <80 is ok(gpu will live for years) for gpu. I keep mining rigs gpu's at about 75-80, my laptop gpu is normally ~90 in games so it should be fine.
Well, as I started to try out bitcoin mining, I used my old PC. It is mostly build of old spare parts.
I first used that Java applet miner. After about 1 hour there was a bad smell of burnt plastics. I just wanted to turn the PC off to check out where the smell was coming from, when it suddenly switched off. Further investigations brought me to the conclusion, that the mining just killed the cheap china 350W PSU. The CPU is an Athlon XP2800+ (yay, mining rate was below 1MHash/s with the Java miner
)
I first used that Java applet miner. After about 1 hour there was a bad smell of burnt plastics. I just wanted to turn the PC off to check out where the smell was coming from, when it suddenly switched off. Further investigations brought me to the conclusion, that the mining just killed the cheap china 350W PSU. The CPU is an Athlon XP2800+ (yay, mining rate was below 1MHash/s with the Java miner
)
Hey guys, I'm just interested in reading about the chances of our mining hardware just...'dying' from overuse.
Any horror stories about how quickly it took some of them to die? Any success stories about people who've been running their GPUs since the dawn of time?
Currently running my two GPUs at a stable 70 degrees and I'm curious as to what I might have to face in the future. On my old regular computer, which I left on almost 24/7, the no name PSU would die every quarter or so. I've had motherboards die after half a year. I've had graphics cards die in months from just minor use. And this was from what I would consider 'normal usage'!
Any horror stories about how quickly it took some of them to die? Any success stories about people who've been running their GPUs since the dawn of time?
Currently running my two GPUs at a stable 70 degrees and I'm curious as to what I might have to face in the future. On my old regular computer, which I left on almost 24/7, the no name PSU would die every quarter or so. I've had motherboards die after half a year. I've had graphics cards die in months from just minor use. And this was from what I would consider 'normal usage'!
Jump to: