• 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.

Failure rates of Hardware in 2013 - GPU's

Caporegime
Joined
24 Sep 2008
Posts
38,284
Location
Essex innit!
They also qualified that, "The list may look very NVIDIA heavy, but part of it is that we simply sell more NVIDIA cards than AMD cards, so there are more NVIDIA models that we have sold enough of to be confident in our data." In addition to this fact, even though they sell more NVIDIA cards, the NVIDIA GeForce cards on a whole were much more reliable in 2013 than AMD Radeon cards. NVIDIA GeForce cards only had an overall failure rate of 3.3% versus AMD Radeon cards which had an overall failure rate of 10%. This marked an increase in AMD's GPU failure rate, which is definitely a step back for the company. They also mentioned that Nvidia's professional cards had a failure rate of 2.05% while AMD's had an almost identical 2.17%, which isn't surprising considering that professional cards are fundamentally designed and tested to have lower failure rates.

5d52c364b6c61525d89cb8a4cb25bfe4.jpg


http://www.brightsideofnews.com/news/2013/12/27/whats-the-most-reliable-hardware-of-2013.aspx

The whole article is a good read and I first thought "Yer, AMD have more failures, as they have sold far more cards than nVidia but not the case. QC failing? I thought 3% was bad but 10% is crazy bad.
 
Amd sell more cards. End of ;)

I take it you didn't read what was posted or the article, as it clearly states they sold more nVidia cards, so where you got that from is beyond me.

we simply sell more NVIDIA cards than AMD cards, so there are more NVIDIA models that we have sold enough of to be confident in our data.
 
Failure rates are also % based so it doesn't really matter how much they sell as long as they sold enough to get a representative sample.
 
Probably due to people mining 24/7 without adequate cooling / air flow and burning out the cards.
 
Thanks for posting that Gregster, very interesting read. :)

Would be nice to check that the sample sizes are large enough for the numbers to be indicative of an overall trend (e.g. I've never heard of this company before, they might have sold 10,000 NVidia GPUs and only 10 AMD for all I know), but they'd have to be some pretty far out numbers to make the 3% vs 10% failure rate comfortable reading for me as a red team member.
 
Probably due to people mining 24/7 without adequate cooling / air flow and burning out the cards.

Not sure mining has had enough of a mainstream representation within the sample range to have much impact.

What I find quite interesting:

Puget is a very Kingston-heavy builder, only offering Kingston memory. All of the RAM that they included in their failure rates was Kingston and after looking at their numbers we totally see why. For their consumer-level desktop RAM list they offer two DIMMs, a HyperX 1600 MHz kit and a regular 1600 MHz kit even though they offer three different DIMMs, these two appear to be the most popular. Their HyperX low voltage kit had a failure rate of 0% while the non-HyperX kit had a failure rate of 0.11% which is almost as good as zero in my book.

I've started to lean towards Kingston RAM a lot lately as its been so reliable for me (touch wood), I used to use it in the past but started using OCZ stuff for awhile but back to Kingston again now, so good to see it seems to be a wider experience to.
 
Rroff said:
What I find quite interesting:
Puget is a very Kingston-heavy builder, only offering Kingston memory. All of the RAM that they included in their failure rates was Kingston and after looking at their numbers we totally see why. For their consumer-level desktop RAM list they offer two DIMMs, a HyperX 1600 MHz kit and a regular 1600 MHz kit even though they offer three different DIMMs, these two appear to be the most popular. Their HyperX low voltage kit had a failure rate of 0% while the non-HyperX kit had a failure rate of 0.11% which is almost as good as zero in my book.
Rroff said:
I've started to lean towards Kingston RAM a lot lately as its been so reliable for me (touch wood), I used to use it in the past but started using OCZ stuff for awhile but back to Kingston again now, so good to see it seems to be a wider experience to.

Yeah UK RMA base as well for Kingston RAM and SSD's if you wanted to go direct with them.

Mushkin must have a good reputation as well, two of their clients are Apple and NASA.

Intel is pretty damn good to, they will arrange courier to pick up the SSD if you need to RMA it.
 
Last edited:
Very interesting thread Greg, thanks.

I'm really surprised by the Intel CPU failure rate. I'd have thought it would be zero (like AMD). Wonder if it is to do with delidding.
 
Ive seen one thread on another forum were Intel actually replaced a 3770k that was delidded. Only cpu deaths ive seen from users on here was due to overvolting, two in the last week or so.
 
They should have been more specific about which brand and model had the highest failures as I read last year one massively skewed the result and I'm sure it was a sapphire model, but I will have to find it when im back on normal internet and not on my mobile.
 
Last edited:
They should have been more specific about which brand and model had the highest failures as I read last year one massively skewed the result and I'm sure it was a sapphire model, but I will have to find it when it back on normal internet and not on my mobile.

From memory Gigabyte had a brief run of bad 7850's.
 
AMD is a nightmare, by biggest mistake in life and I mean it. If you visit my profile, take a look on all the threats created by me, where they present stupid issues with AMD I have to fight on daily basis.

Never again.
 
From memory Gigabyte had a brief run of bad 7850's.

Actually this says return rates not failures, but still it still shows how one brand can skew the over all.

- PNY 0.94% (against 1.32%)
- MSI 1.38% (against 1.81%)
- Gainward 1.61% (against 1.27%)
- Zotac 1.70% (N / A)
- ASUS 1.81% (against 1.69%)
- Gigabyte 1.84% (against 1.54%)
- Sapphire 3.15% (against 3.51%)

PNY climbed from second to first place, while Sapphire keeps the last position. However, it should be noted that sales of PNY do not much about the high-end, more prone to failure. Proof that a good overall rate is not proof of anything, the GTX 660 card sold in volume with the greatest return is 2.86% with a PNY.

MSI also a significant rise in second position. Here are the models with a rate of return higher than 5%, they are unfortunately many:

- 12.67% Sapphire Radeon HD 7850
- 7.44% Sapphire Radeon HD 7870 OC V2
- 7.41% Sapphire Radeon HD 7870 OC V1
- 7.02% Sapphire HD 7950 With Boost (11196-16)
- 6.09% ASUS HD7750-DCSL-1GD5
- 5.82% Sapphire Radeon HD 7870 V1
- 5.65% Sapphire Radeon HD 7870 V2
- Gainward GeForce GTX 5.30% 670

In first place, we do not find a Radeon 7870 as could be expected but 7850! Sapphire model has apparently suffered the same problems as the 7870 we already widely discussed, but the OC version is saved and reached 2.39% for information. The 7870 is also in a good position, either V1 or V2, V2 corrected having been sold on the end of the period. Intruders are a fanless Radeon 7750 ASUS GTX 670 and Gainward.

If we look at the figures GPU we get:

- Radeon HD 7850: 3.74%
- Radeon HD 7870: 5.48%
- Radeon HD 7870 XT: 4.25%
- Radeon HD 7950: 5.75%
- Radeon HD 7970: 5.31%

- GeForce GTX 660: 1.01%
- GeForce GTX 660 Ti: 2.81%
- GeForce GTX 670: 2.87%
- GeForce GTX 680: 1.99%

It is easy to conclude that the cards based on AMD GPUs are less reliable than maps based on NVIDIA GPU. Without completely deny these figures, it must be stated that weighs heavily on the Sapphire 7850 and 7870, since without the Sapphire card we arrive at 1.5% and 1.64% on the GPU. In contrast to the 7950 and 7970 rates are higher if the Sapphire cards are excluded, but on sales rather scattered among various models that are below the minimum sample requested by reference.
http://translate.googleusercontent....s.html&usg=ALkJrhjasU9NgQ9JJMRv_4tbDEkNpYEJOg
 
Never heard of the company, although I do see more issues on our forums from AMD users, gpu wise at least.

Saying that most of the issues aren't actually the cards failing..

They only say that ''we" as a company sell more green than red so. I'd take those numbers with a tub of salt.

Would be great to know what sort of numbers ocuk have in regards to best selling and failure rates.
 
Back
Top Bottom