Associate
- Joined
- 21 Sep 2010
- Posts
- 455
Thinking from a warranty perspective, which brand out of ASUS and MSI would you favour for buying a second hand graphics card that you suspect has been mined?
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Which brand is best for a card that's been mined?
- Thread starter Dangerous_Dave
- Start date
Man of Honour
- Joined
- 12 Jul 2005
- Posts
- 21,529
- Location
- Aberlour, NE Scotland
Neither because personally I wouldn't touch a card that has been mined on. The warranty on Asus cards is not transferrable and unless MSI go by the serial number I doubt it is with them either. You are relying on the original owner to provide support when something goes wrong.
only in the sense that when you game, you do it for a couple / few hours most days (giving the card hours off between sessions) but when you mine it's 24/7 for as many months as they mined for. You could argue mining for a few months put the same 'milage' as years of regular gaming.
I believe MSI still use the manufacturing date contained in the serial number for warranty claims without a receipt.
Soldato
- Joined
- 1 Dec 2015
- Posts
- 18,512
evga US changed terms and condition for 2nd hand warranty claims because of mining - was a while back
Electronic equipment has an expected lifetime (before failure), and for something like every 10 degrees hotter it operates at, the expected lifetime is halved. Can't remember where I read that, but it was on some engineering website
Associate
- Joined
- 18 Jan 2008
- Posts
- 713
- Location
- UK
The durability of ex-mining cards is quite a tricky subject, one without may certainties. The expected lifespan of a card mostly relates to the capacitors, which will be rated for a certain number of hours at a given temperature; say, 5000 hours at 105C. Lower temperatures increase the life of the caps, higher temps reduce it significantly. So those 5000 hour caps may last 10,000 hours if run at 80C or just 1000 hours at 120C.
Temperature is the most important variable, rather than length of use. It's quite possible for a gaming card to be cooked in a badly ventilated case, and for well-cooled mining cards to have lots of life left in their caps. The best ex-mining cards come from big scale miners. If you're mining with 3 or 4 cards then any kind of jury-rigged cooling will mostly do in the short term, but 100+ cards are going to need a professional vented cooling arrangement. Plus GPUs consume less power at lower temperatures, which is an important factor spread over hundreds of cards.
There's also temperature cycles to take into account. Mining cards will have been run at one steady temperature for weeks or months at a time, whereas gaming cards are doing constant cold/hot/cold thermal cycles. That places physical stress on both the PCB and the solder balls under the GPU and memory, eventually causing cracks in the solder or traces.
tl;dr - a mining card may be worn out junk, or it may be in better condition than a used gaming card. Without knowing the exact conditions it was used in there's no way to know.
Temperature is the most important variable, rather than length of use. It's quite possible for a gaming card to be cooked in a badly ventilated case, and for well-cooled mining cards to have lots of life left in their caps. The best ex-mining cards come from big scale miners. If you're mining with 3 or 4 cards then any kind of jury-rigged cooling will mostly do in the short term, but 100+ cards are going to need a professional vented cooling arrangement. Plus GPUs consume less power at lower temperatures, which is an important factor spread over hundreds of cards.
There's also temperature cycles to take into account. Mining cards will have been run at one steady temperature for weeks or months at a time, whereas gaming cards are doing constant cold/hot/cold thermal cycles. That places physical stress on both the PCB and the solder balls under the GPU and memory, eventually causing cracks in the solder or traces.
tl;dr - a mining card may be worn out junk, or it may be in better condition than a used gaming card. Without knowing the exact conditions it was used in there's no way to know.
I wouldn't buy a second hand card if I suspected the owner was lying about it's use.
Lots of serious miners under-volt their cards.
This is actually a valid argument to be said that an undervolted mining card has seen less "miles" than a card from an enthusiast gamer (overclocked and overvolted for many hours a day).
I would have no issue buying a mining card from someone that was actually tech savy, knew what they were doing and didn't lie about it. No issue what so ever.
This is actually a valid argument to be said that an undervolted mining card has seen less "miles" than a card from an enthusiast gamer (overclocked and overvolted for many hours a day).
I would have no issue buying a mining card from someone that was actually tech savy, knew what they were doing and didn't lie about it. No issue what so ever.
You should also consider the 'cycles' the cards have. Gamers tend to load there cards up and down, hundred and hundreds of times. Miners, very few.
Also whilst mining, many cards are likely to be under less load than if they were gaming.
Also whilst mining, many cards are likely to be under less load than if they were gaming.
I wouldn't generalise on home or pro setups. Both can be good and bad.
I agree with this.
The only thing that may have more wear and tear with mining is the fan.
And even in this aspect having a constant usage with the fan in an upright positon could be better than having a fan in a standard PC case facing down.
The bearings in the fans are designed to deal with the centrifugal force created when the fan is spinning call the Radial load.
But when mounted in a case with the fans face down there is additional Axial load which the fans are not so good at deaing with (depending on design of fan).
That is where we get a lot of noise problems with squeaky fans.
So there are positive and negative points on both uses.
Even before I mined my main PC was running 24/7, thought the GPU load was idle for most of that (75% +). When I was using the the older GPUs the fans used to run constantly anyway, none of this flash zero speed stuff back then.
But every GPU I owned that was used in this way has run correctly years after I have sold them on.
The only one I know has failed was my 8800 GTX, but that was 10 years old, 7 year in my work PC. Death total, not fan realated, the 9-5, 5 days a week killed it. ;_; (Office temps probably don't help, no central heating)
When I do sell my mining cards I'd give full details including photos in rig, average temps ect, and then use them as my gaming card for a few weeks before hand.
Indeed
Do you worry about how many times you power cycle your TV? Or your kettle? Or your... anything?
Really curious as to where this who "Power cycling is most destructive!" argument comes from.
After all, I've never heard anyone say that you shouldn't turn off your TV because power cycling it is harmful
I wouldn't agree on that. Under mining cards are at 100% load.
I can only comment on my cards that sit far cooler than they do if gaming, around 10c cooler. Whilst at 100% constant, they're barely breaking a sweat in comparison.
I suppose there is "load and there is "load".
For example running AIDA64 my CPU is at 100% load but 20 degrees cooler than running Prime95, also at 100% load.
In a gaming rig the cards will be mostly at idle, briefly spun up for a couple hours gaming, then off most of the day.
No matter how cool your mining chips are running, they are both
a) hotter than idle
b) hotter than when off
The gaming rig will average much lower temperatures across a day.
Total stress = heat x time. Not just max_heat.