News Swiss Retailer Stats Reveal Which GPU Brand Has the Highest Failure Rates

[Admin]

Digitec Galaxus AG shares its statistics regarding graphics card defect rates and the time it takes to process a warranty claim.

Swiss Retailer Stats Reveal Which GPU Brand Has the Highest Failure Rates : Read more

 

[Dr3ams]

Those are useless statistics to the general consumer.

 

[pf100]

Those are useless statistics to the general consumer.

It could have an effect on a buying decision I make.

 

[ilukey77]

Ive got alot of powercolor GPU's and never once had issue with any of them if they ever did Nvidia cards i would buy theirs..

funny i would have though sapphire would have been better as they are the OG AMD AIB card provider !!

 

[deesider]

Those are useless statistics to the general consumer.

Useful to reinforce my bias against Gigabyte

 

[Dr3ams]

funny i would have though sapphire would have been better as they are the OG AMD AIB card provider !!

I own two Sapphire cards and not a single problem. Sapphire cards here in Germany are highly rated and if you look at online buyer reviews, at retailers such as Alternate and Mind Factory, you'll see around 99% positive comments.

 

[Dr3ams]

It could have an effect on a buying decision I make.

If you are buying from that retailer, then maybe their statistic would be useful. But their statistic doesn't match buyer reviews here in Germany and it probably wouldn't match reviews in the United States either. Also, some retailers don't even test the cards before they RMA them back to the manufacturer.

 

[USAFRet]

It could have an effect on a buying decision I make.

The percentage differences are so small as to be irrelevant.

 

[sstanic]

People who buy expensive Asus Strix cards are far more likely to send them back if they don't like the pink shade of their RGB compared to people who buy the cheapest Gainward or Palit cards..

 

[cyrusfox]

Matches my experience with Gigabyte, takes forever to RMA, but they did fix my card, took about 6 weeks with acceptance, shipping and fix turnaround. Not quick....

 

[bond007alex]

Wow, how convenient, the top spots are all taken by companies with little availability to the plebs or companies that sell gpu's that don't have to be installed.

I wonder why that could be? Is that because people with little tech knowledge are far more likely to destroy tech with improper handling?

Without controlling the samples this entire study isn't very useful.

 

[Amdlova]

In my life with graphics brands XFX and gigabyte have let me down... other brands last the time to the gpu be pointless.

 

[kerberos_20]

so far got zero issues with any GPU, the only two GPUs ive killed was because ive had some deformed PC case, GPU wasnt sitting properly and it moved, happened twice in row with two different GPUs, which resulted in PC not powering up when GPU inserted xD that was like 10-15yrs ago (hd 6850 and gtx 9800+)

 

[Yourhuckleberry]

I have a Gigabyte Windforce 1070 still kicking since 2016 so this is a relatively dumb chart considering they aren't taking into account the volume they have sold for each manufacturer. Cherry picking stats is well, smooth brain territory

 

[DSzymborski]

The lack of data makes it difficult to take seriously because of the spread of the probabilities. They say minimum of 300 GPUs, which is a pitifully small number in this context. At 300 GPUs, even if you knew for a fact that the underlying probability of failure was 1%, there's still about one chance in seven that the observed failure rate would be under 0.3% or above 1.7%

 

[Deleted member 2950210]

Proud owner of an Inno 3D RTX 4090 OC!

 

[bill001g]

The lack of data makes it difficult to take seriously because of the spread of the probabilities. They say minimum of 300 GPUs, which is a pitifully small number in this context. At 300 GPUs, even if you knew for a fact that the underlying probability of failure was 1%, there's still about one chance in seven that the observed failure rate would be under 0.3% or above 1.7%

But people that are bad at math just like to read the headlines and look at pretty charts. It could also be they manipulated the data and are hiding it because this is really some form of advertising under the guise of a factual report.

 

[Nicholas Steel]

I had a Gainward Geforce 6600GT back in the day, as far as I can recall it worked perfectly fine with no issues.

 

[Colif]

The percentages aren't big enough to really be of concern. The number of cards sold with no problems overwhelms the ones with problems.

 

[wr3zzz]

The lack of data makes it difficult to take seriously because of the spread of the probabilities. They say minimum of 300 GPUs, which is a pitifully small number in this context. At 300 GPUs, even if you knew for a fact that the underlying probability of failure was 1%, there's still about one chance in seven that the observed failure rate would be under 0.3% or above 1.7%

Exactly. 0.3% failure rate out of 300 is 1/300. Comparing against brands shipping in the millions it is more sampling bias than anything that is statistically meaningful.

 

[PEnns]

"It's important to remember that the collected data is limited to Digitec Galaxus AG's sales. Therefore, the data does not portray the number of cases globally. For obvious reasons, Digitec Galaxus AG doesn't fully divulge the total amount of units sold. Sadly, that also means we don't have the whole panorama for the charts. For example, a lower value of failure rates doesn't necessarily mean the brand is better than the competition if it only represents a small portion of the global sales."

OK, so what's the point of this article then??

It's obviously useful only to maybe 0.01% of global GPU users!

PS: I have 2 EVGA cards with a combined life of almost 14 years. And yes, they still work flawlessly.

 

[evdjj3j]

"It's important to remember that the collected data is limited to Digitec Galaxus AG's sales. Therefore, the data does not portray the number of cases globally. For obvious reasons, Digitec Galaxus AG doesn't fully divulge the total amount of units sold. Sadly, that also means we don't have the whole panorama for the charts. For example, a lower value of failure rates doesn't necessarily mean the brand is better than the competition if it only represents a small portion of the global sales."

OK, so what's the point of this article then??

It's obviously useful only to maybe 0.01% of global GPU users!

PS: I have 2 EVGA cards with a combined life of almost 14 years. And yes, they still work flawlessly.

The point of the article is to try and get advertisements in front of your eyes.

 

[NeoMorpheus]

OK, so what's the point of this article then??

I dont know, but sounds like a subtle hit piece against the AMD brand. Something very common lately.

PS: I have 2 EVGA cards with a combined life of almost 14 years. And yes, they still work flawlessly.

My last N....dia GPU was a 970 from EVGA and like yours, still working like a champ.

Currently have a 7900 XTX Pulse by Sapphire and knock on wood, working flawlessly.

 

[eye4bear]

Useful to reinforce my bias against Gigabyte

I will second that thought about Gigabyte - got burned a few years back on a MB that never-ever worked 100%, and zero updates once they moved to a new model MB.

 

[bobalazs]

ASUS DUAL-RX6750XT-O12G Radeon RX 6750 XT 12GB GDDR6 OC PCIE Failure rate: 7,7%

SAPPHIRE 11318-01-20G Radeon RX 6750 XT 12GB GDDR6 NITRO+ Gaming OC Failure rate: 6,5%

On ipon .hu webshop site/store.

I can give links, but they're in hungarian. https://ipon.hu/shop/termek/asus-dual-rx6750xt-o12g-radeon-rx-6750-xt-12gb-gddr6-oc-pcie/1997475

https://ipon.hu/shop/termek/sapphir...50-xt-12gb-gddr6-nitro-gaming-oc-pcie/1996389

SAPPHIRE 11306-01-20G Radeon RX 6700 XT 12GB GDDR6 NITRO+ PCIE
Meghibásodási arány: 1,3%

XFX Radeon RX 6700 XT 12GB GDDR6 Speedster Qick 319 PCIE
failure rate: 1,6%

BIOSTAR VA67T6TML9 Radeon RX 6700 XT 12GB GDDR6 PCIE
failure rate: 3,5%

POWERCOLOR AXRX 6700XT 12GBD6-3DHE/OC Radeon RX 6700 XT 12GB GDDR6 Red Devil OC PCIE
failure rate 1,3%